I 2 PhDW 2015
Organised by:
Faculty of Electrical Engineering and:
Lublin University of Technology
Institute of Electronics and
Information Technology
Electrotechnical Institute Warsaw
Under the auspices of:
Sponsored by:
The text of the various contributions in this volume were set individually by the authors or under their supervision.
No responsibility is assumed by the editors or the publisher for any injury and/or damage to person or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein.
ISBN 978-83-61956-35-8 e-ISBN 978-83-61956-36-5
Electrotechnical Institute
28 Pozaryskiego Str. 04-703 Warsaw, Poland www.iel.waw.pl
Book cover design: Maja Kocoń
©All rights reserved
Copyright by Electrotechnical Institute 2015
Program Committee:
Gratkowski Stanisław, West Pomeranian University of Technology, Poland -
Chairman
Sikora Jan, Lublin University of Technology, Electrotechnical Institute, Poland -
Co-chairman
Brykalski Andrzej, West Pomeranian University of Technology, Poland -
Co-chairman
Ziółkowski Marcin, West Pomeranian University of Technology, Poland -
Editorial board chair
Adamiak K., University of Western Ontario, Canada
Aoqui S.I., Sojo University, Japan
Bales V., Slovak University of Technology
Brauer H., Ilmenau University of Technology, Germany
Choi I.S., Hannam University, Korea
Cichocki A., Riken Brain Institute, Japan
Cieśla A., AGH University of Science and Technology, Poland
Dedkova J., Brno University of Technology, Czech Republic
Delong A., Delong instruments a.s., Czech Republic
Demenko A., Poznan University of Technology, Poland
Doležel I., Czech Technical University, Czech Republic
Domek S., West Pomeranian University of Technology, Poland
Drexler P., Brno University of Technology, Czech Republic
Fiala P., Brno University of Technology, Czech Republic
Hanzelka M., Brno University of Technology, Czech Republic
Hoła J., Wrocław University of Technology, Poland
Hotra O., Lublin University of Technology, Poland
Hotra Z., State Polytechnic University of Lvov, Ukraine
Iwaszkiewicz J., Electrotechnical Institute, Poland
Jakubiuk K., Gdansk University of Technology, Poland
Jancarik V., Slovak University of Technology, Slovak Republic
Janicek F., Slovak University of Technology
Jarzyna W., Lublin University of Technology, Poland
Jung H.K., Seoul National University, Korea
Kadlec R., Brno University of Technology, Czech Republic
Kaźmierkowski M.P., Warsaw University of Technology, Poland
Krak Y., National Technical University of Ukraine, Ukraine
Krawczyk A., Czestochowa University of Technology, Poland
Kroutilová E., Brno University of Technology, Czech Republic
Kryvonos I., Institute of Cybernetics of Ukrainian Science Academy
Lionheart B., University of Manchester, UK
Mikulka J., Brno University of Technology, Czech Republic
Nagy L., Budapest University of Technology and Economics, Hungary
3
Nilsson H.E., Mid Sweden University, Sweden
Pavo J., Budapest University of Technology and Economics, Hungary
Pozzi M., Cranfield University, United Kingdom
Ripoll J., Inst. of Electronic Structure & Laser (FORTH), Greece
Rymarczyk T., Net-Art R&D, Lublin Poland
Sankowski D., Technical University of Lodz, Poland
Schmid U., Vienna University of Technology, Austria
Skoczkowski T., Warsaw University of Technology, Poland
Slama J., Slovak Univ of Technology, Slovak Republic
Slosarčík S., Technical University of Kosice, Slovak Republic
Sokolowski J., Université de Lorraine, Campus des Aiguillettes, France
Stryczewska H.D., Lublin University of Technology, Poland
Toepfer H., Ilmenau University of Technology, Germany
Tomczuk B., Opole University of Technology, Poland
Udvary E., Budapest University of Technology and Economics, Hungary
Usak E., Slovak University of Technology
Wac-Włodarczyk A., Lublin University of Technology, Poland
Wiak S., Łódź University of Technology, Poland
Wilczyński W., Electrotechnical Institute, Poland
Wójcik W., Lublin University of Technology, Poland
Wójtowicz S., Electrotechnical Institute, Poland
Yamada S., Kanazawa University, Japan
Zakrzewski K., Electrotechnical Institute, Poland
Zymmer K., Electrotechnical Institute, Poland
Organising Committee:
Barbara Szymanik – Chairman
Maja Kocoń
Adam Łukomski
Błażej Jabłoński
Katarzyna Cichoń-Bańkowska
Aurelia Kołodziej
Paweł Frankowski
Iwona Piech
4
16:00
19:30
Thursday 14.05
19:00 Participants' registration
Banquet - Welcome Party
8:30
9:45
11:00
11:15
12:15
12:30
14:00
15:40
17:00
17:15
18:15
20:00
Friday 15.05
9:30 Breakfast
11:00 Opening and Invited Session 1
11:15 Coffee break
12:15 Oral Session 1
12:30 Coffee break
13:30 Oral Session 2
15:30 Lunch
17:00 Commercial Session
17:15 Coffee break
18:15 Poster Session 1
19:15 Poster Session 2
Banquet
8:30
10:00
11:00
11:15
12:15
12:30
14:00
15:00
Saturday 16.05
9:30 Breakfast
11:00 Oral Session 3
11:15 Coffee break
12:15 Oral Session 4
12:30 Coffee break
13:30 Invited Session 2
15:30 Lunch
Local Trip
8:30
10:00
11:00
11:15
12:15
12:30
13:30
Sunday 17.05
9:30 Breakfast
11:00 Oral Session 5
11:15 Coffee break
12:15 Oral Session 6
12:30 Coffee break
13:30 Closing Ceremony
15:30 Lunch
5
Friday 15.05 10:00 – 11:00
OZ1_1 Andrzej Demenko
Network model of 3D regions with electromagnetic field
Computational Engineering: A Challenge for
Interdisciplinary Engineering Education
OZ1_2
Hartmut Brauer and Hannes
Toepfer
Friday 15.05 11:15 – 12:15
O1_1
Arnold Kalvach and Zsolt Szabó
Isotropy analysis of metamaterials
O1_2 Maciej Burak
O1_3
Tomasz Rymarczyk,
Rymarczyk and Jan Sikora
Paweł
Inhomogeneous CTMC Model of a Telephone Call
Center with Abandonment.
Modelling and Simulation of EIT Flood Embankment
System
Friday 15.05 12:30 – 13:30
O2_1 Artur Krupa and Bartosz Sawicki Massive simulations using MapReduce model
Limiting Input Signals During Nonlinear Control of a
O2_2 Adam Lukomski
Humanoid Robot
O2_3
Hela Garbaa, Lidia Jackowska-
Strumillo, Krzysztof Grudzien and
Andrzej Romanowski
Application of Electrical Capacitance Tomography and Artificial Neural Network to rapid estimation of cylindrical shape parameters of industrial flow structure
Friday 15.05 15:40 – 17:00
CS_1 Tomasz Krupicz (COMSOL Multiphysics) Comsol Multiphysics
CS_2 Marta Hajdasz (ABB Sp. z o.o.) Welcome to ABB
CS_3 Piotr Lipnicki (ABB Sp. z o.o.) Modern Diagnostics for Industry
6
Saturday 16.05 10:00 – 11:00
O3_1
Iveta Glodová, Vladimír Gašpar and Jozef Bocko
O3_2 Pawel Niewiadomski
Analysis of frequency with elastic-plastic materials on a specific object using the finite element method in
Abaqus environment
Effect of addition of selected nanoparticles on parameters of concrete mix and hardened selfcompacting concrete
O3_3
Tomoya Abiru, Fumiaki Mistugi,
Tomoaki Ikegami, Kenji Ebihara and Shin-Ichi Aoqui
Environmental application of electrical discharge for ozone treatment of soil
Saturday 16.05 11:15 – 12:15
O4_1 Agnieszka Piekarska
O4_2
Jiří Chytil, Martin Jakube and Jiří
Sliž
O4_3
Pavel Křepelka
Application of electroporation in medicine and cosmetics
The development of universal impedance spectroscopy for use in food industry
D-shaped optical fiber and its application
Saturday 16.05 12:30 – 13:30
Application of optical wave microphone OZ2_1 Fumiaki Mitsugi
OZ2_2
Pavel Fiala, Dusan Nespor and Petr
Drexler
Numerical model of a large periodic structure
7
Sunday 17.05 10:00 – 11:00
O5_1
Gergely Mészáros and Tibor Berceli Harmonics Suppression of Microwave Oscillators
O5_2
Konstantin Weise, Matthias
Carlstedt, Marek Ziolkowski and
Hartmut Brauer
O5_3
Tamás Pető
Lorentz Force Eddy Current Testing: Theory and
Simulation
Multichannel Passive Radar Receiver Platform
Sunday 17.05 11:15 – 12:15
O6_1
O6_2
Piotr Graca and Bronisław
Active Axial Magnetic Bearing Prototype Field
Tomczuk
Vladimir Chudacik and Milan
Smetana
Calculation
Tilt-Shift Eddy Current Probe Impact on Information
Value of Response Signal
O6_3
Zoltán Szalay and Lajos Nagy
Measurement of Thin-Film Thickness by Coaxial
Resonator
8
Friday 15.05
P1_1
P1_2
17:15 – 18:15
Ayoub Saoud, Volodymyr Mosorov and Krzysztof Grudzien
Daniel Olchowy and Łukasz
Gołąbek
P1_3
Diana Olejnik and Paweł
Frankowski
P1_4
P1_5
P1_6
Eliška Vlachová Hutová, Petr
Marcoň and Karel Bartušek
Jakub Szumowski and Przemysław
Adamkiewicz
Karol Duda and Przemysław
Adamkiewicz
P1_7
Maciej Borówka
P1_8 Maciej Gniadek
P1_9 Marcin Maciejewski
P1_10 Martin Valla
P1_11
Norbert Rosman and Barbara
Szymanik
P1_12 Paweł Prokop
P1_13
Paweł Tchórzewski
P1_14 Piotr Gas
P1_15
Piotr Trubiłowicz
P1_16
Przemysław Wróblewski
P1_17 Tomas Halagan
P1_18
Tomasz Zinko, Paulina Pianko-
Oprych and Zdzisław Jaworski
Flow swirl measurement using Electrical capacitance tomography
Prototype of Measurement System in Electrical
Impedance Tomography
Wireless Control of a Measuring Robot Designed for the Tracking and Evaluation Rebars in Concrete
Structures
Effect of high voltage on the development of the plant tissue
Prototype of an electrical capacitance tomography system using a capacitance-to-digital converter
Flood embankments monitoring system model with use of electrical impedance tomography
Active EEG electrodes design and simulation – introduction
Adaptive input shaping algorithm
Feature extraction in noisy ECG signals processing.
DNA analysis using Hartree Fock methods
Objects' classification in infrared based landmine detection
Using of the Chan-vese method of image segmentation using non-linear function approximated by sections.
Adjoint equation in electrical impedance tomography
The Optimal Microwave Hyperthermia Treatment
Based on the Operating Characteristics of the Multi-
Slot Coaxial Antenna
From Images Metadata toward Ontology of Images
Optimization of MPI measurements simulation
Interdomain NFV Federation
Three-dimensional modelling of thermal stresses in a planar Solid Oxide Fuel Cell of a novel design
9
Friday 15.05 18:15 – 19:15
P2_1
P2_2
Adam Ryszard Żywica
Agnieszka Gdula
Magnetoacoustic Tomography with Magnetic
Induction for Noninvasive Imaging of Low
Conductivity Objects
The use of statistical methods for segmentation of dental x-ray images
P2_3
Anna Burek and Maciej Borówka
Algorithms of incentive in IT companies
P2_4
Jacek Grochowalski and Zbigniew
Frąckiewicz
P2_5 Jacek Kryszyn
Design Problems of High Power Static Voltage
Source for Isolated Power Grids in Shipbuilding and
Maritime Industry
FPGA-based Electric Field Solver for Electrical
Capacitance Tomography
P2_6
P2_7
Katarzyna Biernat and Waldemar
Kozłowski
Konrad Niderla and Przemysław
Adamkiewicz
Autonomous Electric Vehicle Charging Station
P2_8 Leszek Wolski
P2_9 Maciej Gniadek
P2_10
Maciej Panczyk, Beata Panczyk and
Jan Sikora
P2_11
Michal Jurcisin, Stanislav Slosarcik and Peter Balog
P2_12 Nadzeya Viktarovich
P2_13
Paweł Frankowski.
A fuzzy logic based temperature control system for the biogas digester
High frequency DC/DC converter based on SiC
Transistors for RES application.
Interaction of input shaping and LQR controller
Selected problems of wall dampness tomographic measurements
Possibilities of heat dissipation from electronic structures
Energy saving performance analysis of polygeneration of heat, cooling and power
Multi-Frequency Eddy Current Method for Phase
Based Inspection of Steel Bars in Reinforcement
Concrete Structures
P2_14
Rafal Piotuch, Ryszard Pałka and
Kamil Tarenko.
Genetic algorithm based PMSM optimization
P2_15
Rafał Borowiec and Wojciech Surtel
P2_16 Roman Hajtmanek
P2_17
Włodzimierz Kruczek
P2_18 Wojciech Matelski
Construction of a terminal device to examine the state of human activity
Robust models for uniform workload distribution problem
Practical ability to control electrical static converter valves according to patent PL168286
Low power DC/DC converter from 3 kV to 300 V: simulation analysis
10
Professor Andrzej Demenko is Head of Division of Mechatronics and
Electrical Machines at Poznan University of Technology, Poland.
Currently he is Chairman of the Committee on Electrical Engineering of the Polish Academy of Sciences (PAN), Chairman of Commission of
Electrical Sciences in Poznań Branch of Polish Academy of Science, and also Vice-Chairman of Polish Society for Theoretical and Applied
Electrical Engineering (PTETiS). Since 1970 he has been employed in research and education. He has published 5 books and over 280 conference and journal papers on electromagnetics, electrical machines, numerical methods and applied physics. His personal research is development of field method in analysis and designing of electrical machines and drives. Professor A. Demenko is active in numerous international committees, steering groups, and editorial boards. He is elected member of Board of
International Compumag Society, Chairman of Steering Committee of International Symposium on Electromagnetic Phenomena in Nonlinear Circuits, member of Advisory Board of COMPEL, and member of Steering Committees of several international conferences. Professor A. Demenko is also the Editor-in-Chief of Archives of Electrical Engineering, quarterly magazine of Polish
Academy of Science.
Fumiaki Mitsugi received the B.E. degree in department of electrical engineering and computer science, in 1997, and the M.E. and Dr.(Eng.) degrees in graduate school of engineering, from Kumamoto University,
Kumamoto city, Japan, in 1999 and 2002, respectively.
From 2002 to 2007, he was a post doctoral research fellow in electrical engineering, Oita University, Oita city, Japan. From 2007 to 2014, he was an Assistant Professor and since 2014, he has been an Associate Professor with the graduate school of science and technology, Kumamoto
University. He has contributed more than 100 papers in international journals and conference proceedings. His research interests include timeresolved observation of discharge plasmas in atmospheric pressure, pressure waves from discharges, laser-solid processes and thin film fabrication, laser ablation in supercritical fluids, optical wave microphone technique, and application of ozone for plants, soil and water.
11
Hartmut Brauer was born near Berlin, Germany, in 1953. He received the Diploma engineer and the Dr.-Ing. in 1975 and 1982, respectively, both in electrical engineering from the Technische Universität Ilmenau,
Germany. Since 1975, he has been with the Department of Advanced
Electromagnetics, Technische Universität Ilmenau, where he is currently working as Senior Lecturer and Senior Researcher.
He is author/coauthor of more than 80 papers in journals and books. His main research interests include theory and computation of electromagnetic fields, inverse field problems/optimization in electrical engineering, with applications to nondestructive evaluation, bioelectromagnetics, and educational aspects. Dr. Brauer is a founder member (1994) of the International Compumag
Society (ICS). He is member of the international steering committee of the International
Workshops on Optimization and Inverse Problems in Electromagnetism (OIPE) since 2008 and of the International Symposium of Theoretical Electrical Engineering (ISTET) since 2013.
Pavel Fiala: 1993– Deputy director at the Centre for the Modelling and
Optimisation of Fields in Electromechanical Systems, a joint project of the
Faculty of Electrical Engineering and the company ABB EJF a.s., Brno –
Doctoral degree completion (Ph.D.) at the Faculty of Electrical
Engineering and Information Science, Brno University of Technology; defence of doctoral thesis titled "Modelling of Current Transformers with
Short Circuit Tests", 2003 – Director of the Department of Theoretical and
Experimental Electrical Engineering, Faculty of Electrical Engineering and Communication, Brno University of Technology, 2005-Defence of inaugural dissertation titled "Modelling and Design of a Power Pulse Source", since 09.2014 full proffesor of The
Theoretical and Electrical Engineering.
Applied research in the high-voltage range of heavy-current electrical engineering, mainly within the field of electrical faults and discharges (1988-1990). Modelling of non-linear characteristics of materials (1992-1996); modelling, analysis and optimisation of coupled electromagnetic/thermal/mechanical problems connected with models formulated using concentric parameters; modelling of problems with turbulent flow and an electromagnetic field influenced by chemical composition of materials. Dr. Fiala has built and tested several efficient algorithms for the optimisation of non-trivial problems. Furthermore, he has practiced the modelling and measurement of non-linear material properties – hysteresis (1991-1993), and proposed or realised the design and modelling of power pulse sources PROTOTYPA a.s. and power microwave generators (PROTOTYPA a.s, Tesla Vršovice, Institute of Plasma Physics,
Academy of Sciences of the CR, Praha). Dr. Fiala has participated in educational work at the
Department of Theoretical and Experimental Electrical Engineering through teaching courses such as Theoretical Electrical Engineering, Electric Circuit Theory Electrical Measurements,
Electromagnetism
12
Friday 15.05
13
14
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Andrzej Demenko
Poznan University of Technology, Poland andrzej.demenko@put.poznan.pl
Abstract:
The paper discusses the analogies between finite element equations for electromagnetic field and equations of electric and magnetic networks (circuits). The methods using nodal and edge elements are considered. The network models of regions with magnetic field and electric field containing eddy and displacement currents are analysed. Both scalar and vector potentials are used. Two types of graphs/networks are introduced: (a) edge networks for nodal elements, and
(b) facet networks for edge elements. It is noted that edge values of vector potentials A and T are represented by loop quantities in facet networks.
Branch equations of equivalent networks are given. It is shown that the branch parameters of equivalent magnetic and electric networks are defined using the interpolation functions of edge element for edge network and using the interpolation functions of facet element for faced network.
The loop equations of presented facet magnetic and electric networks represent the equations of edge elements method for vector potentials A and T . However, the nodal equations of discussed edge magnetic and electric networks are equivalent to the equations of nodal finite element method for formulations using scalar potentials
and V .
The proposed network models for magnetic and electric field are used to form the coupled magneto-electric models. In these models the magnet and electric networks are coupled via source.
The equations of coupled via source models represent the finite element equations for formulations using magnetic and electric potentials, i.e. for A V ,
T and A T formulation. Special attention is paid to the coupled models of regions with multiply connected conductors.
Finally, the example of network model application is presented. The finite element methods of electromagnetic torque calculation are discussed. The virtual work principle is applied.
The principle is adapted to the magnetic edge and facet models. As a result the network representations of Maxwell’s stress formula are created.
15
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Hartmut Brauer, Hannes Toepfer
Ilmenau University of Technology , Germany hartmut.brauer@tu-ilmenau.de
Keywords: computational engineering, electrical engineering and information technology, interdisciplinary engineering education, curricula for Master students, PhD and joint research teams
Abstract:
Within the Research Training Group „Lorentz Force Velocimetry and Lorentz Force
Eddy Current Testing“ 12 PhD students from 6 countries (and actually 9 associated members from
5 more countries) are working at the faculties Mechanical Engineering, Electrical Engineering and
Information Technology and Computer Science and Automation of the Technische Universität
Ilmenau for their PhD. During a period of three years, they are studying together and collaborating but they have passed very different academic training courses.
In the current educational program they have to pass a course which is called
“Computational Engineering”, including aspects of computational mechanics, computational electromagnetics, computational heat transfer, computational fluid dynamics, mathematical modelling and validation, across all three faculties. Additionally, there are lectures on measurement technology, material science, numerical methods, signal processing and optimization.
Because in such interdisciplinary research teams a minimum of knowledge of all disciplines is required for successful collaboration, the establishment of the course
“Computational Engineering” is considered as a challenge, not only for the lecturers. In the presentation will be shown the challenge and the chances for renewing the current curricula, from the point of view of electrical engineering and information technology.
16
Friday 15.05
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Arnold Kalvach, Zsolt Szabó
Budapest University of Technology and Economics, kalvach@evt.bme.hu
Keywords: metamaterials, wave propagation, simulation, optics
Abstract:
Metamaterials are artificial, mostly periodic structures with subwavelength feature sizes, which are applied to control the propagation of electromagnetic waves from radio frequencies to visible light. Due to their deep subwavelength feature sizes they can be described as homogeneous materials in a given frequency range, in a similar way as natural materials are characterized with macroscopic (bulk) material parameters. However, in contrast to natural materials, metamaterials can be designed to have resonances at arbitrary frequencies and to show extraordinary properties like near-zero or negative refractive index.
Unfortunately, these properties are limited for a narrow frequency range or a narrow incident angle range. The former is due to the nature of resonances, the latter is due to the structure of metamaterials. It is difficult to find and to fabricate 3D structures that show the same macroscopic properties for any angle of incidence. Instead, 2D structures are most commonly used, where the angle of incidence is expected to be small.
In this paper we study the so called Fishnet structure and show that despite being a 2D structure, it can represent a quasi-isotropic behavior in a given frequency range. Additionally, the examined structure has a near-zero refractive index in the same frequency range which can be exploited in different applications.
To test isotropy angle-dependent effective parameters of the structure are retrieved. First,
S-parameters are obtained for a number of incident angles after applying full wave simulation on one unit cell of the structure. The unit cell is surrounded with Bloch boundary conditions, hence representing an infinite periodic structure. Then the S-parameters are compared with the Sparameters of an ideal homogeneous slab. The latter can be expressed analytically with the normal wave number within the material and the generalized wave impedance of the material, hence inverting these expressions the effective normal wave number and the effective generalized wave impedance of the metamaterial can be computed from the obtained S-parameters. Knowing the normal wave number for a number of incident angles the dispersion relation of the metamaterial can be delineated for the given frequency. If the sum of squares of the tangential and normal wave numbers are constant for a single frequency, the material is said to be isotropic on that frequency.
The examined structure is a 5-layer Fishnet structure. The layers consists of two 38um thick copper layers on 760um thick ISOLA carrier each, separated with 500um air gap. The distance of the layers is 10mm and the size of the unit cell is 20x18mm. The size of the windows in the copper layers is 11x6.5mm. The structure shows a near-zero refractive index around 10GHz, and its dispersion relation can be well approximated with that of an isotropic material for a wide incident angle range.
19
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Acknowledgement
This work has been supported by the Bolyai János Fellowship of Hungarian Academy of Sciences, the EUREKA project MetaFer and PIAC_13-1-2013-0186 of the National Development Agency
Hungary.
Maciej Burak
West Pomeranian University of Technology, Poland, maciej.burak@zut.edu.pl
Keywords: uniformization, call center, transient models, numerical methods, Markov chain
Abstract:
In this paper we consider modeling an inbound telephone call center, where the callers may balk or abandon, with an inhomogeneous continuous time Markov chain model. We further discuss the practical application of uniformization method and compare it to traditional call center modeling methods based on stationary approximations.
20
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Tomasz Rymarczyk 1 , Paweł Rymarczyk 1 , Jan Sikora 2
1 Net-art, Poland, 2 Lublin University of Technology, Poland tomasz@rymarczyk.com
Keywords: Electrical Impedance Tomography, Inverse Problem, Finite Element Method, Level
Set Method
Abstract:
This paper presents the new method examining of the flood embankment dampness by the electrical impedance tomography. Our system was prepared to determine the dampness of the test flood blank on a specially built model. Decreasing along wall height and depth humidity creates nonhomogeneous testing environment where object properties conductivity varied smoothly with one more coordinates. Surface potential measurements are performed at different angles of projection whereby the information needed to determine an approximate distribution of conductivity inside the object is obtained. The proposed algorithm is initialized by using topological sensitivity analysis. Shape derivatives and topological derivatives have been incorporated with the level set method to investigate shape optimization problems.
The coupled algorithm is a relatively new procedure to overcome this problem. The number of iterations determine the position and shape of zero level set functions. In this algorithm, there can control the process of the image reconstruction. Numerical methods of the shape and the topology optimization were based on the level set representation and the shape differentiation and were made possible topology changes during the optimization process. Level set methods have been applied very successfully in many areas of the scientific modelling. These approaches based on shape sensitivity include the boundary design of the elastic interface.
There are two features that make these methods suitable for the topology optimization.
The structure is represented by an implicit function. This function is often discretized on a regular grid that conveniently coincides with the finite element mesh. The next valid feature is the simple update of the implicit function using the Hamilton-Jacobi equation, where the velocity function is determined by the shape sensitivity of the structure.
21
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
22
Friday 15.05
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Hela Garbaa , Lidia Jackowska-Strumiłło , Krzysztof Grudzień, Andrzej Romanowski
Lodz University of Technology, Poland lidia_js@kis.p.lodz.pl
Keywords: electrical capacitance tomography, inverse problem, artificial neural networks, flow parameters estimation
Abstract:
Electrical Capacitance Tomography (ECT) is a non-invasive and non-destructive technique, which allows to visualize a distribution of materials with different dielectric permittivity flowing inside a pipe or a vessel. ECT consists on measuring the capacitances between pairs of electrodes placed around the targeted vessel. The acquired measurements are then processed to reconstruct a tomographic image relying on an appropriate image reconstruction algorithm. The spatial distribution of the materials present inside the isolated vessel is determined.
Comparing to other techniques ECT is low cost and fast and allows to measure and monitor processes in hazardous or unreachable areas (high temperature, high pressure). Limitations of this technique lie on the poor resolution of the provided image. Researchers are still investigating to improve the performance of this technique and agree on the complexity of the task due to the difficulties with inverse problem solution. Two major characteristic obstacles of inverse problem are: its ill-posed and non-linear properties. Typical results of inverse problem solution are the reconstructed images (permittivity distribution inside the ECT sensor space). For control purposes, only a determination of important process parameters is demanded. With a parameterization the process interpretation is simplified and the inverse problem is better posed.
A new approach to solve the inverse problem of electrical capacitance tomography is presented. The proposed method is based on Artificial Neural Network (ANN) to estimate three different parameters: radius r and position coordinates ( x , y ) of a cylindrical object present inside a pipeline. This information allows estimating the distribution of material inside a pipe and determining characteristic parameters of the flow. The proposed approach has the advantages to be direct: the different parameters are estimated without imaging and rapid: reduced computational time when compared to an existing Landweber method.
ANN-based inverse model has been built on the basis of relations between the network input and output vectors. ANN of MultiLayer Perceptron (MLP) type with one hidden layer was trained applying back propagation algorithm to perform the estimation of three flow parameters
( r, x, y ) from ECT data. Tomographic data used to train and test the MLP was simulated via
ECTsim Matlab toolbox.
The obtained results highlighted the advantages of the proposed ANN-based approach.
The considered flow parameters have been estimated 26 times faster than by the use of the
25
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Landweber method. This indicates that the proposed method would be very useful for online cylindrical flow monitoring.
Artur Krupa, Bartosz Sawicki
Warsaw University of Technology, Poland
Artur.Krupa@ee.pw.edu.pl, bartosz.sawicki@ee.pw.edu.pl
Keywords: Cloud Computing, Distrubuted Systems, Stochastic optimization
Abstract
In the last few years cloud computing is growing as a dominant solution for large scale numerical problems. It is based on MapReduce programming model, which provides high scalability and flexibility, but also optimizes a costs of computing infrastructure. This paper studies feasibility of MapReduce model for scientific simulations. It will be shown that solutions of problems consisting of many independent simulations could be implemented in the cloud computing scheme.
Introduction
Distributed processing has a long history in a computer science. Since early beginning it has taken advantage of different processors available in the same time. Recently created term
”cloud computing” is not precise, but its general understanding is based on flexibility of architecture, calculations on-demand, and lower costs.
Despite objections related with privacy, cloud computing is one of the most growing elements of todays technology [2]. It provides flexibility and performance [5]. Word top companies (eg. Amazon, Microsoft, Google) are developing data-centers with hundreds of thousands of computing nodes to process their huge datasets. Academia users also investigate how new architecture could be applied into scientific problems [1].
References
[1] A. Barker, B. Varghese, J. S. Ward, and I. Sommerville, Academic Cloud Computing Research
: Five Pitfalls and Five Opportunities, in 6th USENIX Workshop on Hot Topics in Cloud
Computing, 2014.
[2] G. DAngelo and M. Marzolla, New trends in parallel and distributed simulation: From manycores to Cloud Computing, Simul. Model. Pract. Theory, pp. 126, 2014.
[3] Apache Hadoop project, http://hadoop.apache.org
[4] B. S. Kim, S. J. Lee, and T. G. Kim, MapReduce Based Experimental Frame for Parallel and
Distributed Simulation Using Hadoop Platform, in Proceedings 28th European Conference on
Modelling and Simulation, 2012.
[5] D. Kondo, B. Javadi, P. Malecot, F. Cappello, D. P. Anderson, and U. C. Berkeley, Cost-
Benefit Analysis of Cloud Computing versus Desktop Grids, 2009.
[6] G. Sakellari and G. Loukas, A survey of mathematical models, simulation approaches and testbeds used for research in cloud computing, Simul. Model. Parct. Theory, vol. 39, pp.
92103, 2013.
26
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Adam Łukomski
West Pomeranian University of Technology, Szczecin, Poland lukomski@zut.edu.pl
Keywords: nonlinear control, humanoid robot, bipedal walking
Abstract:
This paper presents a method for limiting input signals during nonlinear control of a humanoid robot. A biped robot model with arms and feet is presented together with a ground impact model. The modelling of support phases currently allows multi-point contact, including standing on toes, flat feet, sitting and arm-assisted movements. The control design uses the feedback linearisation approach, a high-gain method, and modifies it in order to limit the input signals, so that the experimental verification of the resulting motion is possible.
27
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
28
Friday 15.05
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Ayoub Saoud, Volodymyr Mosorov, Krzysztof Grudzień
Lodz University of Technology, Institute of Applied Computer Science, Poland asaoud@kis.p.lodz.pl, mosorow@kis.p.lodz.pl, kgrudzi@kis.p.lodz.pl
Keywords: Electrical Capacitance Tomography, Swirl flow, flow velocity
Abstract
This paper is an approach for analyzing flow swirl based on twin plane electrical capacitance tomography (ECT) system. An implementation to topographic data sets obtained from swirl flow simulation; based on scale invariant feature transform (SIFT) algorithm and crosscorrelation method has been developed. Velocity distribution and flow direction of the swirl flow are expected to be extracted.
Daniel Olchowy 1 , Łukasz Gołąbek 2
1 Net-Art, Poland, 2 Netrix, Poland daniel.olchowy@net-art.eu
Keywords: Electrical Impedance Tomography, Measurement Methods, Implementation of a
Measurement System Impedance Tomography
Abstract:
Electrical impedance tomography (EIT) is a technology to reconstruct the electrical conductivity distribution of an unknown object. The measurement of the electrical conductivity brings direct information about the composition of the conductive medium. The typical problem requires the identification of the unknown internal area from near-boundary measurements of the electrical potential.
The acquisition and the processing time is determined by the number of independent configurations of stimulating and recording pairs of electrodes. A prototype system has been developed, various objects were placed onto the surface of the sensor and measures of sensor performance calculated for each object. The quality of the reconstructed image depends on the quality of measurement of voltage and electric current in measurement system. The main elements of sixteen electrode data acquisition system for electrical impedance tomography system are:
A/D Converter (TI ADS1258)
31
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
General Control System (TI OMAP L-137)
Forming and Control Circurit for Power System
Analog Multiplexers Team (ADG452)
Measuring the tub with 16 measuring electrodes (chrome-nickel) and cables (50
, shielded)
Operational amplifiers (TI OPA365)
Applied analog for digital converter is ADS1258. This is 24-bit, low-noise ADC optimized for fast multi-channel, high-resolution measurement systems. The converter provides a maximum channel scan rate of 23.7kSPS, providing a complete 16-channel scan in less than 700^s.
Diana Olejnik, Paweł Frankowski
West Pomeranian University of Technology, Szczecin, Poland diana_olejnik@wp.pl, pawel.frankowski@sk.sep.szczecin.pl
Keywords: non-destructive testing, reinforced concrete structures, wireless communication, quadruped robot
Abstract:
The aim of the paper is to present new system for wireless control of a measuring robot used to evaluation of reinforced concrete structures. The communication with prototype of quadruped robot is realized in two steps. Firstly, measuring platform is moved to the selected place. In this stage robot is controlled by the user. Then independent algorithm is running and the robot searches for the nearest reinforced and tracing the selected rebar. The hardest part of this stage is a precise detection of steel bars in the structure. For this purpose a magnetometer sensor is proposed. Prepared software allows for freely switching between manual control and automatic tracing of ferromagnets.
Data transmission is working in two modes, the first one is adapted to a quick recognition of the tested area, other one to a data acquisition. In the data acquisition mode, measurements are fed directly to the user panel, what allow for fast preliminary assessment of the structure. When some irregularity are detected, more accurate methods, and sensors mounted on the robot can be used for evaluation. The prototype of the measurement platform due to the small size, will guarantee mobility in difficult to access spaces. In the paper both software and hardware details of the system are presented.
32
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Eliška Vlachová Hutová 1 , Petr Marcoň 1 , Karel Bartušek 2
1 Brno University of Technology, Czech Republic
2 Institute of Scientific Instruments of the ASCR, Czech Republic xhutov00@stud.feec.vutbr.cz
Abstract:
A current part of our thinking is that matter is comprised of particles or waves which have electromagnetic properties. Therefore it is to be expected that externally applied electrical and magnetic fields will cause perturbations in atoms or subatomic particles. Bonding, enzyme action to associate or dissociate molecules, or to alter their configuration, must all be influenced, however slightly, by electromagnetic stimuli. The question is whether electromagnetic stimuli applied externally (usually artificially) can cause changes in plant hormone location, concentration, or action, which will in turn affect growth.
In our experiment the electrical parameters that affect early somatic embryos (ESEs) were investigated. High voltage was generated by a special high voltage generator. High voltages ranging from 5 to 20 kV and frequency of 1Hz were applied longitudinal and transversal directly on the Petri dish with 2 days old ESEs of Picea abies for periods of 3 hours every day. One Petri dish was placed directly on top of the high voltage generator and on the other petri dish were fixed two copper plates for transmission of high voltage. Petri dishes were exposed to high voltage for
14 days. After this time, the influence of high voltage was evaluated.
To evaluate the experiment were used biological and chemical methods, which confirmed the changes in the growth of ESEs.
Jakub Szumowski, Przemysław Adamkiewicz
Net-art, Poland p.adamkiewicz@netrix.com.pl
Keywords: electrical capacitance tomography, capacitance-to-digital, capacitance measurement, Altium
Abstract:
Electrical capacitance tomography (ECT) is a method of imaging cross sections of vessels and pipelines containing dielectric material. Permittivity distribution is measured with a multi-electrode sensor, then converted into voltage with a CDC circuit and finally interpreted by
33
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015 software. Due to various factors, such as limited amount of data acquired or significant computing power needed for accurate image reconstruction algorithms, the results are often low quality inexact images. However it is complexity of ECT systems implementation into actual manufacturing processes and substantial size of hardware equipment that are the main reasons of their absence in industrial applications.
As an alternative to current solutions we propose the prototype of low-cost electrical capacitance tomography system, both compact in size and equally efficient as most of currently existing ones. It has been created using Altium Designer software allowing us to simulate entire building process, thus solving most of the prototyping issues in its initial phase. This paper provides description of the device as well as results of exemplary measurements. Its focused on a hardware part of the prototype as it determines functionality within the specific industrial applications. Limited hardware characteristics are widely considered to act as a barrier against further development of electrical capacitance tomography with advanced image reconstruction algorithms. In this paper we prove creating customer friendly and effective ECT system possible and, furthermore, inexpensive.
Karol Duda, Przemysław Adamkiewicz
Net-art, Poland p.adamkiewicz@netrix.com.pl
Keywords: electrical impedance tomography, embankments, EIDORS, LABVIEW, image reconstruction
Abstract:
Nowadays, it is possible to use technology, knowledge and science to protect people from destructive effects of floods. One way of protecting people and their property is using flood embankments. Furthermore, it is necessary to monitor state quality and level of water diffusion.
To monitor state of embankment Electrical Impedance Tomography may be used. It is based on measuring potential differences between electrodes placed across flood embankment. The way in which we can define state of embankment depends on the fact that every material has unique conductance. It is achieved by measuring potential differences in analized objects. It is possible to resolve ratio of materials that build embankment using EIDORS software. The more complex measurements, the better image of explored area. In this paper we present measurements of flood embankment model using EIT system. The proposed method allows to monitor condition of flood embankment.
34
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Maciej Borówka
AGH University of Science and Technology, Poland maciejborowka@gmail.com
Keywords: EEG, active electrodes, noise, CMRR, instrumentation amplifier, simulation,
SPICE, physiologic artifacts, extraphysiologic artifacts
Abstract:
This article is intended to give basic understanding of EEG active electrodes design requirements. Due to very low amplitude and frequency of EEG, noise signal is several orders of magnitude greater than desired brain’s EEG biosignal. Step by step, physic limitations, different sources of noise causing both physiologic and extraphysiologic artifacts in final signal are described and simulated using SPICE. After setting of boundary conditions, proposed solution, based on instrumentation amplifier is presented. Afterwards provide conclusion based on comparison between real model simulations and design assumptions and also recommendations for future directions of development.
Maciej Gniadek
Poznan University of Technology, Poland maciej.a.gniadek@doctorate.put.poznan.pl
Keywords: input shaping, adaptive control, flexible body systems
Abstract:
Many various mechanical systems are vulnerable for oscillations. Usually this feature is an effect of flexible joints in mechanical systems. This kind of systems are represented by cranes, robotic arms, conveyers and others. Usually all of these objects can be approximated by multimass system (or in the simplest example by two-mass system). Input shaping is one of the simplest and most effective algorithms of oscillations damping. The method is based on convolution of the baseline command (reference signal) with a series of Dirac impulses. The impulses have to be applied in specified moments of time and should have specified amplitude. Tuning of this method might be problematic in physical systems. Because of this reason auto-tuning algorithms were designed. In various systems the off-line learning of input shaping algorithms is insufficient.
Because of this reason the adaptive algorithm was designed. The algorithm is based on on-line learning, identification and FFT analysis. The adaptive algorithm of this method is robust for
35
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015 object parameters varying and (using FFT tuning) for structure changing. The paper shows an algorithm of input shaping parameters adaptation. The simulations experiments were conducted on the crane model and simple transfer function as well. The research was made using
Matlab/Simulink environment.
Marcin Maciejewski
Lublin Institute of Technology, Poland m.maciejewski@pollub.pl
Keywords: ECG, signal processing, filtering, fourier, spline
Abstract:
In this paper two signal processing methods are presented. Both of these methods are used to extract features from pre-recorded ECG signal. The signal must first undergo filtering and normalisation. Afterwards, the methods are applied. As the result, various traits are calculated.
Martin Valla
Brno University of Technology, Czech Republic xvalla00@stud.feec.vutbr.cz
Abstract:
Molecular modeling is carried out either by mechanical molecular or quantum chemical methods. The quantum chemical selected a group of AB initio Hartree Fock method (HF). It is based on wave equations and their Schodinger Hartree simplify ivolving SFC (Self Consistent
Field). SCF represent three points of the charge carriers (1 and 2 interacting monitored). Negative charges, electrons. Hartree approach does not take into account the antisymmetry of the wave function. This lack corrects variance method Hartree Fock. HF method is applied to a specific sequence of the DNA molecule. It is important to observe that it has the molecular DNA level low frequencies.
36
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Norbert Rosman, Barbara Szymanik
West Pomeranian University of Technology, Szczecin, Poland szymanik@zut.edu.pl
Keywords: infrared thermography, microwave heating, landmine detection
Abstract:
Infrared thermography can be applied to buried objects’ detection, also in case of antipersonnel landmines. Research at this field indicates that shallow buried objects can be successfully located using infrared methods and image processing. In any kind of demining procedures distinction between irrelevant to this procedure objects as stones, wooden or glass fragments and lethal landmines would be crucial.
This work is aimed at practical distinction between training antipersonnel landmines and stones. Measuring system used to conduct experiments is composed of the detection section
(thermovision camera FLIR A325) and heating section based on magnetron which generates microwaves. In experiments the following parameters were diversified: the depth of burial, heating time and types of training landmines. In this paper a series of research’s outcomes are presented and discussed.
Paweł Prokop
Lublin Univeristy of Technology, Poland paw.prokop@gmail.com
Keywords: Chan-vese, segmentation, non-linear function
Abstract:
The interpretation of telemedic images is an important factor from the viewpoint of diagnosis and further treatment of the disease. This method requires a lot of commitment, concentration, focus and human effort. Necessary to reduce these factors are informatics tools.
The use of appropriate algorithms and computing power of informatics tools allows to improve work performance through faster and more accurate analysis of objects included in image.
The research material obtained from computed tomography in the vicinity of the trunk, presents a series of images in grayscale, for which an image segmentation were made . The aim
37
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015 of this study was to detect objects associated with soft tissues. For this purpose, used two approaches.
The first method was the original CT (Computed Tomography) image segmentation by the method of Chan-vese. The second approach was based on a two-stage image processing: transformation of an image using a non-linear function with approximated sections, removing objects with high brightness corresponding to the hard tissues and their contents and removing noise from the image. Image were processed and then subjected to segmentation using the method of Chan-vese.
Used in studies of non-linear scaling function approximation with the cut gates
(separating gates), caused converting each pixel value in terms of luminance image. Introduced variables were dependent on the arbitrarily assumed threshold. First of them in terms of the low value of the saddle were located between the barbs of the histogram of the image, that had an impact on changing the pixel values of bright objects, located above the accepted threshold, which were characteristic of hard tissue (bone). The second, oscillated around the first hump of histogram to remove noise. Proved to be extremely important matching the second value, for which the adopted transformation was used to sharpen image edges.
As a result of analysis of 40 images a combination of both methods, gave satisfactory results for simple medical CT images. Comparing the two approaches, a comparative analysis performed a quality indicator, which is the duration of the segmentation directly related to the number of iterations in the method of Chan-vese to detect edges of objects in the soft tissue and the number of detected independent objects. The approach of using a nonlinear function proved to be more effective than the approach without pre-processing the image and brought satisfactory results. Thus obtained objects will be analyzed further for the purpose of medical diagnosis.
Paweł Tchórzewski
Net-Art, Poland pawel.tchorzewski@netrix.com.pl
Keywords: boundary element method, adjoint equation, inverse problem
Abstract:
The electrical impedance tomography is non-destructive imaging technique, which has various applications. For example, it can be used in medical imaging. Efficient algorithms for solving forward and inverse problem have to be developed in order to use this technique for practical tasks. In our approach the algorithm of the inverse problem bases on the Boundary
Element Method (BEM), the gradient technique and the level set method. In the gradient technique so called adjoint equation has to be solved. The solution has to be obtained in each iteration step.
Numerical techniques give us opportunity to find approximate solutions of differential equations which cannot be solved by means of analytical ones. Among various numerical tools
38
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015 we concentrated our attention on BEM. BEM can be effectively employed on condition that partial differential equation can be transformed to integral form. Additionally, the Green’s function has to be calculated. In our numerical calculations for simplicity zero order approximation has been chosen. The function which describes electrical conductivity distribution on the domain possesses two different nonzero values. Finally, we have solved the boundary value problem for twodimensional adjoint equation. Therefore, the proposed numerical model has been verified.
Piotr Gas
AGH University of Science and Technology, Poland piotr.gas@agh.edu.pl
Keywords: interstitial microwave hyperthermia, multi-slot coaxial antenna, antenna impedance and reflection coefficient characteristics, S11-parameter, TM waves, Pennes equation, tissue properties, finite element method
Abstract:
Nowadays microwave thermal therapies are encountered in different branches and applications, but a common technique applicable to such problems is associated with the interstitial microwave hyperthermia. This minimally invasive method of cancer treatment is very popular due to the simple applicator construction as well as its low production costs and ease of application. Presented paper formulates the complex problem including the coupling of electromagnetic wave equation under harmonic working conditions and bioheat transfer equation taking into account the tissue cooling due to the blood circulation. Since the entire microwave antenna has cylindrical structure, the axial symmetrical model with the TM waves is used.
Moreover, the basis of the numerical implementation of such problem is the finite element method.
The objective of this paper is to find the optimal temperature and electromagnetic field distributions based on some operating characteristics of the multi-slot antenna with a 50 ohm coaxial feed.
The microwave antennas including one, two or three cylindrical air slots in their structure were revised with regard to their impedance and reflection coefficient characteristics. Evaluating the S11- parameter on the entrance of used antenna is a straightforward way to get the resonant frequencies where the best antenna impedance matching occurs. What is important, the coaxial feed is being modelled using the port boundary condition with the set value of the total microwave input power. By changing this parameter, the temperature within the human tissue can be easily controlled to not exceed the therapeutic range of 40-45°C appropriate for hyperthermia treatment.
This work shows that there are some differences between the operating frequencies of microwave antennas described in the literature and the resonant frequencies obtained from the antenna
39
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015 characteristics that may significantly affect the heating efficiency of the tissues during the microwave interstitial hyperthermia treatment.
Piotr Trubiłowicz
Net-Art, Poland piotr.trubilowicz@netrix.com.pl
Keywords: Image Metadata, Ontology, Ontology for Knowledge Sharing, Resource,
Description Framework, Web Ontology Language. Semantic Network, Knowledge Base
Abstract:
Digital Imaging and Communications in Medicine (DICOM) is the standard for the communication and management of medical imaging information and related data [DICOM specification by NEMA 2015a].
DICOM enables the integration of scanners, servers, workstations, printers, and network hardware from multiple manufacturers into a picture archiving and communication system. DICOM metadata could be treated as vocabulary.
An ontology is an explicit specification of a conceptualization [Gruber 1993]. Ontology is formalized set of concepts and relations between them. In this poster author describes issues associated with relations between specification of image metadata and ontology of knowledge.
Sets of term describing images, like DICOM or EXIF standards are vocabularies. In this work the transformation from simple list thru XML file to Resource Description Framework is presented.
There are also considerations what is needed to transform simple list of metadata into Web
Ontology Language.
40
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Przemysław Wróblewski
Warsaw University of Technology, Poland p.wroblewski@ire.pw.edu.pl
Keywords: Magnetic Particles Imaging, optimization, nanoparticles
Abstract:
Magnetic particles imaging [1] (MPI) is a new imaging molecular technique, which uses non-linear magnetization curve of superparamagnetic nanoparticles to measure their concentration in the sample. Since 2011 we are developing new magnetic particles imaging scanner in our
Institute [2]. Recently the main obstacles that we encountered was to obtain the most accurate system matrix of our scanner. Result of the measurement using our MPI scanner is the convolution of nanoparticles concentration distribution and system function of the scanner. To achieve the actual image of concentration one need to perform deconvolution of measurements with appropriate system matrix.
Unfortunately, deconvolution is highly ill posed problem. As the measurement itself contains noise, system matrix need to be flawless to even consider performing deconvolution.
First method to obtain system matrix is to calculate it using real measurements of unit sample, but as was mentioned above real measurements contain to high noise ratio for system matrix obtained that way to be useful. Therefore, I proposed another approach. By performing exact simulation of measurements, using MATLAB toolbox developed in our instituted [3], one can calculate system matrix from the results of those simulation.
However, to be able to perform such precise simulation one needs all the parameters of the measurement process, such as: diameter of the nanoparticles (D), displacement of sample in relation to center of the scanner (delt), magnetization of saturation of nanoparticles (Ms0), temperature of nanoparticles (T), strengths of gradient field (Hs) and homogenous driving field
(Hd), etc. Unfortunately, accurate values of most of those parameters are hard to obtain. Hence, main goal of this work was to perform optimization of the simulated measurements on the ground of real measurements, using parameters mentioned above as optimization parameters.
All of the optimization were conducted in MATLAB. I defined the goal function as the mean square of difference between real measurement and simulated one. Real measurements and simulations were performed for 5th harmonic of the nanoparticles signal. Initially I analyzed the parameters using elementary effects method to determine if any of them are redundant. As all of them proofed to be useful, I proceeded to actual optimization. Firstly, I performed local optimization using gauss-newton based algorithm. As starting point the based match acquired by manually choosing values of the parameters was used. Secondly, I run the global optimization to make sure the result of the local optimization wasn’t only local minimum. Genetic based algorithm and DIRECT algorithm were used. Finally as the measurements of signal from particles can be acquired for different harmonics, I conducted two criterion optimization using as the second criterion optimization for 3rd harmonic using goal obtain algorithm.
41
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Results of optimization for 5th harmonic look promising. High similarity between real and simulated measurements can be achieved. Hopefully results of those simulation can be used to calculate correct system matrix for reconstructions. Unfortunately, best optimization result for
3rd and 5th harmonics are not complementary, but result of two criterion optimization suggests some other Pareto-optimal parameters sets that can be useful for both harmonics. Further studies need to be carried out to determine this. Also more optimization using different real measurements need to be performed.
Bibliography
[1] KNOPP, Tobias; BUZUG Thorsten, M. Magnetic Particles Imaging. An Introduction to
Imaging Principles and Scanner Instrumentation. Springer, 2012
[2] SMOLIK, Waldemar; WRÓBLEWSKI, Przemysław; SZYSZKO, Jan. Development of mpi mini scanner prototype - coils design. In: Proc. 2013 IEEE International Conference on
Imaging Systems and Techniques, IEEE, 2013, p. 164-168
[3] SMOLIK, Waldemar T.; WROBLEWSKI, Przemyslaw R.; SZYSZKO, Jan. Numerical modeling of magnetic field for magnetic particle imaging. In: Imaging Systems and
Techniques (IST), 2012 IEEE International Conference on. IEEE, 2012. p. 436-441.
Tomas Halagan
Slovak University of Technology in Bratislava, Slovakia tomas.halagan@gmail.com
Keywords: interconnect, NFV, inter domain, external domain, high availability, federation, network function, network operation, automation server
Abstract:
Aim of this work is to create architecture to enable interconnection of several Network
Function Virtualization (NFV) domain to utilize benefits that arise from the redundancy. Users who have subscription to network service automatically counted with a high on-going operation of that service, often referred to as High Availability (HA). In current computer networks, this is existing fact, faced by Network Service Providers. Server malfunctions, on which virtualized network functions reside, caused by component failure, human error, or system malfunctions might lead to service interruption and raise the cost of running business. Service interruption is an undesirable state and it is absolutely essential to prevent this situation by appropriate measures.
Nowadays, the number of available data centers (DCs) gives us many opportunities, how to implement smooth running network operation as well as automation servers. Dedicated servers in data centers may be reserved for the purpose of providing HA virtualized network functions.
In order to come HA functionality into NFV environment, it is imperative to create a certain level of integration of several NFV domains, which may belong to different business
42
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015 entities. This integration will enable any virtual network function from within any NFV domain to be backed up, redundant or migrated into different external domain. This process of integration can be termed as a federation, thus in case of NFV domains as an Interdomain NFV Federation.
Using my proposed architecture SDN controllers can communicate in various NFV domains using the proposed control messages.
Tomasz Zinko, Paulina Pianko-Oprych, Zdzisław Jaworski
West Pomeranian University of Technology, Szczecin, Poland tomasz.zinko@zut.edu.pl
Keywords: planar Solid Oxide Fuel Cell, thermal stresses, Finite Element Method (FEM),
Computational Fluid Dynamics (CFD)
Abstract:
The modelling investigation was carried out to analyze thermal stresses in an anode supported planar Solid Oxide Fuel Cell (pSOFC) and to estimate thermal expansion of the fuel cell. The stress is due to a spatial gradient of temperature and mismatch in thermal expansion coefficients of the fuel cell components. The temperature distribution was based on previously developed thermo-electrochemical model predicting fuel cell operation. A coupled Computational
Fluid Dynamics (CFD) and Computational Structural Mechanics (CSM) with Finite Element
Method (FEM) analysis were performed using the commercial ANSYS software: Fluent with
SOFC module and Mechanical. The design of a single planar SOFC proposed by Bossel [1] consists of three membrane electrode assembly (MEA) layers (anode, electrolyte, cathode) and two cross-flow bipolar plates with 26 ribs. Bipolar plates act as flow channels and also as current collectors. The gas flows are diagonal from one corner to the diagonally opposed corner.
Moreover, the fuel and air flows are cross-wise opposed on each bipolar plate. The study allows to indicate the most vulnerable to damage area of the planar fuel cell of a proposed design at the operating conditions. It will be useful in further design modification due to an endurance optimization of the planar SOFC.
Acknowledgement
The research programme leading to these results received funding from the European Union’s
Seventh Framework Programme (FP7/2007-2013) for the Fuel Cells and Hydrogen Joint
Technology Initiative under grant agreement no [325323]. Acknowledgments are due to the partners of SAFARI project. The work was also financed from the Polish research funds awarded
43
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015 for the project No. 3043/7.PR/2014/2 of international cooperation within SAFARI in years 2014-
2016.
References
[1] U. Bossel, Rapid startup SOFC modules, Energy Procedia, 28, 2012, 48-56. DOI:
10.1016/j.egypro.2012.08.039.
44
Friday 15.05
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Adam Żywica
West Pomeranian University of Technology, Szczecin, Poland adam.zywica90@gmail.com
Keywords: magnetoacoustic tomography with magnetic induction MAT-MI, finite element of
MAT-MI, object image reconstruction
Abstract:
Many various imaging techniques are playing an increasingly significant role in biomedical research and clinical diagnosis. In the last years especially noninvasive electrical conductivity imaging has been actively investigated, because electrical properties of biological tissues are known to be sensitive to physiological and pathological conditions of living organisms.
Magnetoacoustic tomography with magnetic induction (MAT-MI) is relatively new imaging technique which combines favourable contrast of electromagnetic tomography with good spatial resolution of sonography. In such a technique a low conductivity object is placed in a static and time-varying (pulsed) magnetic field.
The time-varying magnetic field induces eddy currents in the object volume.
Accordingly, the Lorentz force (result of interaction between static magnetic field and eddy currents) is generated in the object and finally ultrasonic vibrations are emitted. Propagated ultrasonic waves can be acquired by a set of piezoelectric transducers located around the object for image reconstruction. In this paper a finite element model of MAT-MI is presented. A simple image reconstruction has been performed with the help of k-Wave toolbox in Matlab environment.
47
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Agnieszka Gdula
Net-Art, Poland agnieszka.gdula@netrix.com.pl
Keywords: k – means algorithm, k – nearest neighbors algorithm, empirical Bayes decision rule, image segmentation
Abstract:
The aim of paper is to use statistical methods for segmentation of dental x-ray images.
This type of images is different from another one because there is not sensitive to light factor but patient movement is important during making of them. This movement is main reason of artefacts.
Although they have a lot of information, only little should be chosen. It can be done by segmentation, which always relies on dividing images to disjoint areas by some criterion. There are many types of segmentation and it is difficult to say which one is the best. They usually extract different kind of information. However, it is a very important step, because it made improperly, can be the main reason for obtaining valid results at a later stage of the analysis of the characteristics. To locate conditions of soft tissues in regions of teeth of dental x-ray images, the algorithms of k – means, k – nearest neighbor and empirical Bayes decision rules are used.
Anna Burek, Maciej Borówka
AGH University of Science and Technology, Poland maciejborowka@gmail.com
Keywords: motivation, drives, benefits, needs, IT corporation, extrinsic rewards
Abstract:
It is estimated that today in Poland we need 50 000 employees more in IT sector.
According to the forecast in Europe, in 2020 we will need even one million employees in IT sector.
Therefore the IT companies trying to lure in any possible ways prospective employees. The contents of the article are structured within the framework of subject like extrinsic rewards, behavior in work organizations, personal drives, needs, interpersonal influences and the decision to work in an IT corporations. A model is suggested specifies the unique conditions under which all IT employees will become strongly motivated to perform clearly and effectively on their tasks and achieve the goals... This algorithm focuses on the interaction among two main variables.
48
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Firstly, the psychological states of IT employees that must be present for internally motivated work behavior to develop yourself, to achieve more than is expected. Secondly, we need to take care of the characteristics of jobs environments that can create those psychological states and the characteristic of individuals that influence on this, how positively a employees can react to a complex and challenging job. This algorithm is compared to other theories of job design.
Jacek Grochowalski, Zbigniew Frąckiewicz
West Pomeranian University of Technology, Poland jmgroch@gmail.com
Keywords: voltage source, inverters, power electronics, high power, shipbuilding
Abstract:
High power static voltage inverters found their application in a power conversion for isolated power grids, especially in the shipbuilding and the maritime industry. When there is a need for interconnecting ship with an external electric power source for technical (i.e. a vessel is under a maintenance is shipyard, and it cannot use its own onboard power generator) or economical (a ship is being unloaded in a harbor) reasons there are a few options. One of them is to supply electrical energy from a static power inverter, which converts energy from a public grid to the appropriate frequency and voltage needed on board.
In this paper is showed a review of problems associated with design and usage of such static inverters especially in maritime and shipbuilding industry. Most of the problems are that very often nominal power of such devices is comparable to nominal power of its load. And in case of ships, the major part of the load are cranes, pumps, fans and so on, all driven by electrical machines (most frequently induction machines). Starting current of the induction electrical machine is about 6 ÷ 7 times nominal current. And this starting current must be withstood by a static power inverter, which place great demands on its design. This contribution proposes a solution for such issues.
49
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Jacek Kryszyn
Warsaw University of Technology, Poland j.kryszyn@stud.elka.pw.edu.pl
Keywords: Electrical Capacitance Tomography, Electric Field, FPGA, Kaczmarz algorithm
Abstract:
Electrical Capacitance Tomography (ECT) is a technique used to visualize spatial electric permittivity distribution in a sensor. An image is reconstructed from capacitance measurements of electrodes placed on a tomographic sensor. ECT is a very fast imaging technique and it is suitable for dynamic process monitoring, e.g. two-phase flows monitoring. It is also a low cost technique. Low spatial resolution and low sensitivity are main drawbacks of ECT.
One of the challenges in ECT is a measurement of low capacitance values. In a typical
ECT sensor interelectrode capacitance values are in a range from tens to hundreds of femtofarads.
It is not possible to increase these values by usage of larger electrodes because this would decrease spatial resolution which is already low. Very sensitive measurement circuits which are immune to stray capacitances are required. Image reconstruction is another nontrivial problem in ECT. Due to a small number of electrodes there is much less independent measurements than pixels in a reconstructed image. In a typical sensor there are 8 to 32 electrodes which gives 28 to 120 measurements. In generally smallest reconstructed image should have 32 by 32 pixels to be treated as a tomographic image. Number of unknowns is larger by the order of magnitude than number of measurements. This means that the reconstruction of an image in ECT is an ill-posed problem with many solutions. Furthermore electric capacitance depends nonlinearly on electric permittivity. If nonlinear reconstruction is used it is necessary to calculate electric field distribution in each step to update the sensitivity matrix. This is a hard task because electric field distribution matrix is badly conditioned and can be very large, especially in case of 3D ECT. It is crucial to have a fast and reliable solver for electric potential distribution.
In this paper an FPGA-based electric field solver for ECT is proposed. A general device for 2D electric field distribution calculations has been designed and implemented in VHDL. It can be used to solve electric field distribution of any size. The solver is based on a Kaczmarz algorithm. It is an iterative algorithm for solving large linear equation systems. Tests have been made using Spartan-3 FPGA device. Results have been compared to a MATLAB calculations on a modern PC processor. It has been shown that custom solver which takes advantages of FPGA architecture can decrease simulation time over a dozen times in comparison to a PC calculations.
50
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Katarzyna Biernat, Waldemar Kozłowski
FRACO-TERM, Poland frako-info@o2.pl
Keywords: electric vehicle charging station, photovoltaic power plant, smart micro grid
Abstract:
Autonomous electric vehicle charging station generates energy from local renewable sources. Connection with the power system is not required. The station consists of components performing the functions of sources, converters and energy storage. Electrical safety devices and contactors (ESD) are located directly at the input the station from the loaded vehicle. The primary source of energy is wind power station (WPS), small photovoltaic power plant (PVP) and an additional source of electricity is small emergency power generator (EPG).
Electrical energy storage device (EESD) acts as a buffer. Autonomous charging station together with the collection of energy electric vehicles represent micro network. In the selected nodes microgrids are measured and recorded values of electrical (voltage and current), the power and the direction and value of energy flow is determined. The controller (MC) manages elements of the charging station. Infrastructure controller (CCI) can remotely manage all stations on the energetic area. Charging is done at the request of the vehicle. The station provides energy AC or
DC. It is possible the flow of energy to the vehicle and from the vehicle to the station. Similarly, the energy can be drawn from the energy storage device and put into it. The station is managed by the master controller MC. Controller takes support of devices within the station and implement strategies work station. MC controller communicates with the CCI controller vehicle charging infrastructure. Provides information on the current status of energy and adopt smart grid operating instructions.
The paper presents an analysis of the operation of the charging station as smart microgrids. Is presented a mathematical model developed fleet of vehicles operator station.
Individual profiles defined elements of microgrid: sources of energy, electrical energy storage devices and receivers. The analytical model can provide energy relations in a formalized way and evaluate certain limits. Profiles of wind and solar energy are defined by random processes. Is assumed probability density distribution type best suited to real phenomena. Limits of random distributions are tested. Similarly, the profiles are defined energy input into the vehicle battery, except that it is also a random start and end time of charging. The model allows to estimate the size of the EESD and EPG. According to the established structure of micro network created model and algorithm of simulation the energy flows in the sections 24 hour.
The aim was to create a simple tool to estimate (sizing) of controllable energy sources and storage capacity. The designer has the technical assumptions regarding future needs and possibilities of renewable energy installations. On this basis is chosen energy storage and backup source of electricity - small emergency generator. The energy stored in the tank unit (liquid or gaseous fuel) can not be restored in autonomous system. The model reproduces the actual structure
51
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015 of the charging station. Was presented algorithm and computation scheme of simulation model and simulation results that are useful in the design of autonomous charging station.
Konrad Niderla, Przemysław Adamkiewicz
Net-Art, Poland konrad.niderla@net-art.eu
Keywords: Biogas digester, Fuzzy logic, Matlab, Programmable Logic Controllers
Abstract:
In this work, a fuzzy logic-based model system for temperature control of the biogas digester was developed and tested. The automatic control system is based on a computer that receives all the data from the sensors through a PLC. A MATLAB program produces the fuzzy inference and it also takes the control decisions. Temperature measurements are constantly relayed to the fuzzy controller. The control system has a series of tools that could be useful to the operator such as evaluation screen and trend graphics. The operator can enter data from chemical or biological analyses with keyboard, and elaborate his own fuzzy control rules and membership functions. Moreover, it is possible to evaluate the impact of control decisions without operating on the process. Therefore, the fuzzy controller is simple, practical, and low-cost technology and allows the operator to control and modelling the system even if he is not an expert
52
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Leszek Wolski
Electrotechnical Institute Gdansk Branch, Poland leszek.wolski@iel.gda.pl
Keywords: DC/DC converter, SiC JFET transistors, RES applications
Abstract:
The paper presents a concept of the design, construction and test results of the voltage
DC/DC Buck-Boost Converter of rated power up to 1 kW with silicon carbide transistors (SiC
JFET). The inverter is designed to work with the supercapacitor energy storage bank as voltage stabilization system in multi-source renewable energy laboratory (RES). Small conduction losses and low switching energy of SiC JFET transistors make it possible to increase the operating frequency of the inverter to 500kHz. Conducted laboratory tests confirmed the high efficiency model of the inverter at 95%.
Maciej Gniadek
Poznan University of Technology, Poland maciej.a.gniadek@doctorate.put.poznan.pl
Keywords: Input shaping, two mass system, LQR control, oscillation reduction
Abstract:
Many mechanical systems are characterized by flexible joints. These kinds of systems are usually vulnerable for mechanical oscillations. This class of objects are represented by cranes, robotic arms, conveyer drives and others. In various papers these objects are approximated by multi-mass (or for simple objects by two-mass) systems. To avoid oscillations the input shaping method might be used. The input shaping method is based on convolution of the baseline command signal and series of impulses applied in specified moments of time and with specified amplitudes. Thanks to this operation the oscillations are in anaphase and eliminates each other.
The input shaping algorithm is one of the simplest and most effective algorithms of oscillations damping.
The Linear-Quadratic Regulator (LQR) is a strategy used to optimal control of dynamic system with minimal cost. LQR control may affect with oscillations of the systems output. One of the main advantages of LQR control is robustness for changes in object parameters.
53
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
The paper presents possible structures of interaction between LQR and Input Shaping.
Three basic structures of this combination are presented and tested. All simulation research is made in Matlab/Simulink environment.
Maciej Pańczyk, Beata Pańczyk, Jan Sikora
Lublin University of Technology, Poland m.panczyk@pollub.pl
Keywords: Boundary Element Method, Functionally Graded Materials, Infinite boundary elements, Impedance tomography
Abstract:
Tomographic measurements of damp wall dampness distribution belong to the group of non-destructive testing, for example, offering the ability to monitor the drying wall process. From the computational point of view, damped wall is a heterogeneous environment where humidity decreases with height and spread out in a parabolic way along wall cross-section.
Usually the problem is analyzed using a multi-layer models of finite dimensions. Two elements allows to simplify the calculation model: one is the transition from multi-layer model to a simple geometry system described only the outer dimensions of the wall by using a modified
Green's function suitable for functionally graded materials and the other is a reduction of number of elements due to the use of infinite boundary elements.
The paper presents a boundary element method related to damped wall, with implemented infinite boundary elements and functionally graded materials. Results are compared to real measurements.
54
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Michal Jurcisin, Stanislav Slosarcik, Peter Balog
Technical University of Kosice, Slovakia michal.jurcisin@tuke.sk
Keywords: heat dissipation, heterostructure, LTCC
Abstract:
Article deal with analysis of trends in the area of heat dissipation from 3D integrated structures. Analyzed methods are supported with applications (heat dissipation based on different thermal conductivity of LTCC-HTCC heterostructure, technology of cooling channels created on the interface of LTCC-HTCC heterostructure) fabricated at Department of technologies in electronics.
Nadzeya Viktarovich
Lublin University of Technology, Poland viknadya@mail.ru
Keywords: polygeneration, trigeneration, CCHP, energy efficiency, primary energy saving
Abstract:
Abstract: The paper presents the efficiency comparison of the trigeneration system with respect to the separate generation of heat, cooling and power. Trigeneration Primary Energy
Saving (TPES) indicator was adopted as an assessment criterion. The TPES indicator – is the most commonly used indicator of primary energy consumption reduction and it can be achieved by using trigeneration system instead of separate production of heat, cooling and power.
Trigeneration is considered to be high performance, if the calculated values of the relative ratio of primary energy saving (TPES), satisfy the condition: TPES ≥ 0.
Trigeneration is the simultaneous and / or consecutive production of three products of different energy grades from a single primary energy sources. In this paper CCHP system with the simultaneous production of heat, cooling and power has been examined.
The results of simulation calculations are presented, based on which the analysis of the impact of changing value of heat used for cooling as well as electrical efficiency on trigeneration system operations have been performed.
55
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Based on its calculations and data from literature review, it has been proven that the use of trigeneration system brings a number of benefits among others: primary energy saving, reduction of energy losses, increase the efficiency of the whole energy system. It has been shown that with the increase of electrical efficiency, energy-saving performance level tends to increase and it may reach more than 50%. Appropriately selected Power to Heat Ratio (PHR), depending on demand, can become a useful tool in the process of optimizing the trigeneration and achieving the best efficiency trigeneration system for chosen conditions of work.
Paweł Frankowski
West Pomeranian University of Technology, Szczecin, Poland pawel.frankowski@sk.sep.szczecin.pl
Keywords: eddy current testing, reinforced concrete structure, data mining, non destructive testing, phase
Abstract:
Reinforced concrete has become a universally dominant construction material in the past decades. However, structures of this kind relatively frequently requires inspection. Ideally, such testing should be done without damaging the concrete. The aim of this paper is to present a eddy current, multi-frequency method based on the phase. The method was designed for evaluation of rebars in reinforced concrete structures. Utilization of numerous frequency components in the excitation signal allow to fully test the structure. Every single frequency consists independent information. Therefore, impact of noise can be reduced. Moreover, tests prove that the multifrequency excitation and spectrogram system (MFES) allow to observe changes in the structure as a relation between spectrogram signal components. Therefore this method can be used both alone or supplementary to the analysis of voltage waveform shape or magnitude-based, multifrequency method. Studies show that, the phase-based method can be successfully used to evaluate a basic structure parameters like concrete cover thickness or rebar location with high accuracy and certainty. Additionally, it was observed, that differences between signals obtained from diferent transducers create opportunities for more unique identification of bars position. Paper contain also a selected results, descriptions of the used system and data mining algorithms utilized to search associations between output waveforms and structure parameters.
56
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Rafał Piotuch, Ryszard Pałka, Kamil Tarenko 1
1 West Pomeranian University of Technology, Szczecin, Poland r.piotuch@zut.edu.pl
Keywords: PMSM, optimization, pareto front, genetic algorithm, field-weakening
Abstract:
Nowadays it becomes clear that the electromagnetic devices has to fulfil multiple objectives concurrently. Thanks to the rapid advancement in the field of power electronics, digital signal processors and control algorithms PM excited and Switched Reluctance Motors are finding more and more applications and have replaced traction systems of most present hybrid electrical vehicles (HEVs) because they offer high performance over other DC and AC machines. Surface and radially-laminated Interior PM synchronous machines with conventional structure have limited or zero flux-weakening capability. Properly designed IPMSMs are capable of operating in
CPSR (Constant Power Speed Region) – such machines perform also inverse saliency – that is q axis inductance is larger than d -axis inductance. Consequently, it has an additive torque value, so called reluctance torque, that may be exploited to extend CPSR. In a multi-criteria optimization
(MCO) problem general solution is represented by a Pareto Front. Authors implemented optimization procedure in Maxwell (FE) and Matlab environments (GA). Properly designed
PMSMs are capable of operating in CPS regions with high overall parameters.
The case of study is represented by a 4-pole IPMSM with fixed stator geometry and winding parameters. Rotor is equipped with NdFeB magnets. Aim functions were maximization of mean torque value and CPSR and minimization of cogging torque maximum value. In the optimization procedure authors used heuristic technique based on struggle between individuals commonly known as Genetic Algorithm. GA mimics natural selection process were the strongest individuals survive and non-important features fades away during struggle between genes.
57
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Rafał Borowiec, Wojciech Surtel
Lublin University of Technology, Poland raff@onet.pl
Keywords: acceleration, telemedical, terminal device
Abstract:
In this article, we present the problem of building a terminal device using two sets of sensors that measure linear acceleration and angular and magnetic field as markers for assessing the activity of the person under test. In the first part, the authors focus on the description of selected elements of the device, and explain the choice of selected hardware. In the next part of the analysis system was brought closer and the results were interpreted. We analyzed the appropriate choice of placing individual sensors basing on human anatomy. We compared the recorded results of two life situations, ie rest and simulation of fainting. The study was conducted on a group of 10 people. The final section summarizes the results and presents arguments proving the possibility of further use of terminal devices, which may become part of the telemedical system in the future.
Roman Hajtmanek
University of Žilina, Slovakia roman.hajtmanek@fri.uniza.sk
Keywords: uniform scheduling, workload distribution problem, irregularity measure
Abstract:
In this paper we study problems which occur in uniform scheduling of workload distribution with uncertainty. Let us have the non-negative random matrix. Elements of that matrix can be daily loads of vehicles. We need to minimize some irregularity measure of vector of mean dispersion of row sums by permuting columns of matrix. For crisp matrices, the minimization problem for irregularity measure of row sums of the permuted matrix is NP-hard.
We study first model where entries of matrix are random with known mean and dispersion. It is shown that object function formulated as sum of dispersion of row sum about ideal row sum is independent on dispersions of entries of the random matrix. We suggest models
58
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015 inspired by a robust statistic - truncated mean, Windsor mean. The illustrative examples of models are presented.
Włodzimierz Kruczek
Warsaw Electrotechnical Institute, Poland
WKruczek@ikolej.pl
Keywords: Power electronic converters
Abstract:
Power electronic converters are integrated pulse systems. Control of such systems is realized by pulse modulation. There are four basic types of pulse modulation:
- PPM (Pulse Phase Modulation);
- PAM (Pulse Amplitude Modulation;
- PFM (Pulse Frequency Modulation);
- PWM (Pulse Width Modulation).
In each of these modulation type corresponding parameter is proportional to the analog modulation values. The first type is used for the modulation signal and is used in telecommunications. last three types of modulation are used in the power electronics. These methods are widely described in the expert literature.
This article aims to present an alternative method for generating control pulses. It is a method patented in many countries, including Poland, in 1996 under the number 168286. Author of the patent is Grzegorz Skarpetowski. Method consists of the control waveform calculation method can be described by the equation: m t
m0
c os
t
tsd t
( )
2
n
( )
n
t
Where m (t) is a modulating function, which is shape also demanded as the output signal of the inverter.
Through analysis of the spectral waveform calculated from the formula it can be proved that the resulting waveform contains the only selected bands of higher harmonics, whereas PWM and PFM methods generate broad-band control of higher harmonics. Control calculated according to the formula allows to get a job with less number of distortions of the signal. Lower content higher should also result in reduced losses in the motor winding is controlled in this way.
59
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Wojciech Matelski
Electrotechnical Institute Gdansk Branch wojciech.matelski@iel.gda.pl
Keywords: DC/DC power converters, Insulated gate bipolar transistors, Voltage control
Abstract:
Automated and fully supervised railway routes require specialized sensors and actuators situated along the tracks to control the train traffic. Most of these utilities are electrical devices that need power supply. The industrial AC power networks are very often far away from railway routes, therefore the necessary long power cable connections would be expensive. A reasonable source of power for such devices is the railway 3 kV DC traction system. But high voltage converters suitable for low power applications are not common on the market. It is difficult even to find the descriptions of these types of systems in literature. One of the main obstacles in wide application of such converters is related to the quality of the energy available in this way. It is evident that in such systems the reliability of the power supply is of great importance to security of trains. Another problem is the voltage variation in the railway grid, which is allowed by standards and can be of quite large range. This issue is emphasized by a high probability of overvoltages, requiring the high voltage power switches, which is not recommendable from the economical point of view. The better solution is to elaborate special converter topology, minimizing the level of voltage stress on power switches.
The aim of this work was to find a topology enabling the use of low cost IGBT transistors in 3 kV low power converters. In this paper a parallel connection of five identical low power “two transistor flyback converters with division of input voltage” has been proposed. Pulse transformers, incorporated into the flyback converter structure, ensure the galvanic insulation. The designed system has a modular form. The high voltage 3 kV DC is divided on an active input circuit, then further lowered by the work of the flyback converter modules. The duty cycle of power switches is automatically calculated by the control system. The operation of converter, together with the developed control method, has been tested by simulation studies using PSIM.
Results of studies showed proper functionality of the converter and generation of stable output voltage during traction voltage variations. Thanks to the input circuit and converter topology, the converter itself can be built using 1700 V transistors.
60
Saturday 16.05
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Iveta Glodová, Vladimír Gašpar, Josef Bocko
Technical University in Košice, Slovakia iveta.glodova@tuke.sk
Keywords: Abaqus, elastic-plasticity, FEA, frequency, material
Abstract:
The main task of this article is to investigate and analyze the finite element methods
(FEM) in Abaqus of elastic-plastic (hyperplastic) properties on different types of materials using frequency analysis. The selected object will be modeled in Abaqus FEA program. The choice of material is generally important in engineering analysis. Each material has a certain value of error.
Using Abaqus FEA, the elastic-plastic analysis will be done to investigate the tension compression, flexion, torsion, etc. The purpose of Abaqus FEA is to better understand the obtained frequency analysis of the elastic-plastic material behaviour on different types of materials and to understand how does the change of material conditions influence overall results.
Keywords: self-compacting concrete, modification, nanoparticles, physical characteristics, mechanical characteristics
Paweł Niewiadomski
Wrocław University of Technology, Poland niewiadomski.paul@gmail.com
Abstract:
According to literature reports, in order to improve the physical and mechanical characteristics of concrete attempts at modifying it with nanoparticles are made in many laboratories in the world. The results of this selective and highly dispersed research seem to be promising. In connection with his ongoing PhD dissertation work the author collected and critically examined the above results and then planned his own research, which he has carried out in part.
This paper will present the results of tests carried out by the author on 10 series of selfcompacting concrete modified with an addition of SiO2 and TiO2 nanoparticles in the amount of
0.5, 2.0 and 4.0% of the cement weight and with Al2O3 nanoparticles in the amount of 0.5, 1.0,
63
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
2.0 and 3.0% of the cement weight. The results relate to the rheological characteristics of the concrete mixes and the mechanical characteristics of the hardened concrete after 28 and 90 days of curing and they will be presented against those for reference concrete without any addition of nanoparticles.
Tomoya Abiru 1 , Fumiaki Mistugi 1 , Tomoaki Ikegami 1 , Kenji Ebihara 2 , Shin-Ichi Aoqui 3
1 Kumamoto University, Japan
2 Environment and Energy Laboratory, Japan
3 Sojo University, Japan mitsugi@cs.kumamoto-u.ac.jp
Keywords: Ozone, Soil treatment, Electrical discharge
Abstract:
Recently, many types of agricultural chemicals such as pesticide and nitrogen fertilizer are using in agricultural field, however, the overuse of agricultural chemicals cause chemical contamination of agricultural crop, water and soil. Therefore, alternative method is required from environmental point of view. We propose the use of ozone in agricultural field, expecting the use of ozone to replace usual agricultural chemicals as alternative method because ozone has the high effect of sterilization and ability to increase nutrient composition of soil. In this study, influence of ozone treatment on physical properties of soil such as inorganic nutrient (NO3-, NO2-, and
NH4+), pH(H2O), pH(KCl), amount of soil DNA, amount of exchangeable bases (Ca2+, K+,
Fe2+, and Al3+), fungi, and amount of nematodes were investigate.
Ozone was generated by oxygen plasma using a surface discharge and injected into soil in a quartz container which is rotated by a motor. The use of the quartz container can avoid ozone consumption into unexpected organic matters. The concentration of ozone was 78 g/m3 with the flow rate of oxygen gas of 1.5 L/min. Exhaust ozone gas was monitored with an ozone monitor.
The soil used for treatment was andosol that was taken from field for cultivation of crops. The amount of andosol used for treatment was 100 g. Treatment time was changed. We investigated influence of ozone treatment on physical properties of soil such as inorganic nutrient (NO3-, NO2-
, and NH4+), pH(H2O), pH(KCl), amount of soil DNA, amount of exchangeable bases (Ca2+,
K+, Fe2+, and Al3+), fungi, and amount of nematodes.
Physical properties of soil were influenced by ozone treatment. Inorganic nutrient in soil increased after ozone treatment, and kept at a constant value. It was revealed that ozone treatment could induce generation of nitrogen nutrient in soil. The pH(H2O) decreased after ozone treatment because organic matters in soil were decomposed by ozone, however, it recovered to initial value with time. The almost of high molecular DNA in soil were decomposed and living bacteria were also sterilized.
64
Saturday 16.05
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Agnieszka Piekarska
Czestochowa University of Technology, Poland agnieszka.piekarska88@gmail.com
Keywords: electric filed, electroporation medicine, genetic engineering, biotechnology, cosmetics
Abstract:
First attempts to use impulse electric field to increase permeability of cell membranes were conducted on animal cells during second half of 20th century. Success of the mentioned attempts quickly resulted in adaptation of electroporation in such laboratory techniques as in-vivo or in-vitro, among others, to introduce nucleic acids to cells. Properly selected parameters and efficiency enables electroporation to be, at present, used in medicine, genetic engineering, biotechnology and cosmetics. Electroporators are utilized to run electroporation. Electroporators generate impulses with use of condenser in such a manner as to receive controlled exponential or rectangular electric field run. These devices (depending on manufacturer and application) are able to generate impulses up to 3000V in high voltage circuit and up to 500 V in low voltage circuit.
The phenomenon of electroporation in medicine is applied in oncology thanks to which it is possible to optimize dozes of chemotherapeutic agents administered to the patient. It is of high significance as antineoplastic drugs are characterized by very high toxicity. Current scientific news use the term of irreversible electroporation (IRE). IRE uses degradation of cells – effect that till now was considered undesirable. Tests show that IRE completely removes sick tissue and leaves the healthy one. In addition, it does not cause any negative side effects characteristic to other similar methods that use high and low temperature to destroy neoplastic cells. This method does not require administration of drugs, thus harmful chemotherapeutic agents do not influence the patient. Electroporation is used in esthetic medicine and cosmetics to remove scars, wrinkles, acne, cases of lack of elasticity or to skin renewal.
Use of electroporation in gene therapy enables to compensate genetic defect, introduce correction of point or gene mutations, and inactivate selected genes. The advantage of this method in gene therapy is the fact that no substances that may interact with the cell in undesirable manner are introduced to the environment.
67
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Jiří Chytil 1 , Martin Jakube 2 , Jiří Sliž 1
1 Brno University of Technology, Czech Republic
2 Výzkumný Ústav Mlékárenský s.r.o., Czech Republic xchyti06@stud.feec.vutbr.cz
Keywords: Impedance spectroscopy, circuit AD5933, Pseudomonas lundensis, material of electrodes, calibration
Abstract:
The impedance spectroscopy is a very universal spectroscopic method which can be used in whole arrays of applications. Because a quick measurement and relatively cheap analysis instrumentation can portable impedance spectroscope be also used in food industry. In this paper we present the development of impedance spectroscope and basic testing of pork meat quality.
The impedance spectroscope is based on a AD5933 circuit which we have used as key element controlled by MCU. We have used the assembled spectroscope on monitoring of pork meat decay.
We have artificially infected a pork tenderloin with main meat spoiler Pseudomonas lundensis and we have measured impedance spectra during 3 days decay in a room temperature. By using statistical method which processes Real and Imaginary part of measured impedance we have successfully distinguish between spectra of fresh and 1 day old meat.
68
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Pavel Křepelka
Brno University of Technology, Czech Republic xkrepe01@stud.feec.vutbr.cz
Keywords: optical fiber, evanescent wave, coupler, tapper
Abstract:
A technique for gentle removal of optical fiber’s cladding is presented. Fiber core denudation is important for fabrication of Bragg gratings (FBG), evanescent-field-based sensors and fiber couples. Especially in sensing applications, the tapering amplify evanescent field in fibers, which makes the strong interactions with surrounding analyte possible. The multi-D-shaped optical fiber can be also used as a high sensitive refractive-index sensor to detect changes in the surrounding refractive index. Light generated on fiber surface must be coupled back. It requires a special care during fabrication process. It is theoretically proved that the higher modes make it easy to excite the evanescent field by the side-polishing technique.
Typical application for this type of fiber is Fiber-loop ring-down spectroscopy. Light pulse is coupled into an optical fiber. The pulse is introduced too fiber loop by evanescent coupler.
The light pulse is attenuated in evanescent wave-based sensor, decoupled by the coupler and detected. Once the light is coupled into the fiber loop, light pulse makes a large number of round trips before their intensity is below the detection threshold. The D-shaped optical fiber seems to be ideal tool for Fiber-loop ring-down spectroscopy due to its relatively simplicity in terms of evanescent wave creation.
69
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
70
Saturday 16.05
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Fumiaki Mitsugi
Kumamoto University, Japan fumiaki_mitsugi@hotmail.com
Keywords: Optical wave microphone, Fraunhofer diffraction, pressure wave
Abstract:
Calculating an equation on Fraunhofer diffraction of laser that passes in refractive index changes, the solution consists of time-independent and time-dependent terms. An optical wave microphone uses the time-dependent component, detecting very small changes in refractive index or pressure of atmosphere that is so called acoustic waves or pressure waves. The most fundamental setup of the optical wave microphone is composed from a laser, a lens, and a photo detector. When the laser beam crosses perpendicularly to where refractive index changes at the beam waist of the laser, slightly Doppler-shifted diffracted waves of the laser are generated and propagate with and in the transparent beam through the Fourier optical lens, reaching the detector set up in the Rayleigh zone or at the back focal plane of the Fourier lens. Although the optical wave microphone covers frequency range of acoustic waves from audible to ultrasonic, it has an advantage in the detection of ultrasonic region (>100 kHz), where conventional microphones lack the sensitivity, because it does not use any mechanical diaphragm.
There are several applications that we have utilized the optical wave microphone in various fields. The most simple one is that in the form of normal microphone for the detection of audible or ultrasonic sound, however, it is more suitable to set up the optical wave microphone in strong electric or magnetic field, or tiny space where conventional microphones can not approach.
I will introduce the application of the optical wave microphone in plasma fields and material-laser processing. We succeeded to detect tiny shock waves as well as ultrasonic waves emitted from each micro-plasma for atmospheric pressure discharges. For material-laser processing, photoacoustic waves caused by gas piston were detected after a pulsed laser irradiation of solid surface.
The development of a multi-channel fibered optical wave microphone which improves upon the normal optical wave microphone with regard to SN ratio and a visualization system which uses computer tomography is also introduced.
73
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Pavel Fiala, Dusan Nespor, Petr Drexler
Brno University of Technology, Czech Republic fialap@feec.vutbr.cz
Keywords: Nanomaterials, graphene, signal sensing, signal transmission, signal/noise, useful signal, large model, periodic structure
Abstract:
The contribution discusses the design and analysis of a numerical model graphene coaxial line suitable for sub-micron sensors of magnetic fields. In relation to the presented concept, the target areas and disciplines include biology, medicine, prosthetics, and microscopic solutions for modern actuators or SMART elements. The proposed numerical model is based on an analysis of a periodic structure with high repeatability, and it is built upon a graphene polymer having a basic dimension in nanometers. The model simulates the actual random motion in the structure as the source of spurious signals and considers the pulse propagation along the structure; furthermore, the model also examines whether and how the pulse will be distorted at the beginning of the line, given the various ending versions. The results of the analysis are necessary for further use of the designed sensing devices based on graphene structures.
74
Sunday 17.05
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Gergely Mészáros, Tibor Berceli
Budapest University of Technology and Economics, Hungary meszaros@hvt.bme.hu
Keywords: negative resistance oscillator, microwave oscillator, harmonic supression, phase noice reduction
Abstract:
Microwave oscillators generate not only the fundamental frequency but the upper harmonics as well. The upper harmonics are undesirable in most applications because of its negative effects. In many cases the power of the upper harmonics are decreased by another circuit like filters or series/parallel resonant circuits. The commonly used bandpass filter with lumped elements cannot be used at microwave frequencies because of the parasitic of each element. The lumped elements of the filter can be synthesized using open- and short-circuited transmission line stubs. The effect of the parasitic is neglected by this method but the passband of this filter is repeated periodically due to the nature of Richards’ transformation. The effect of the periodic passband can be eliminated by a well-chosen center frequency of the fundamental passband. These filters reflect back the undesirable frequency components so the power of these components will be reduced at the output of the filter. Another way to reduce the power of the harmonics is the use of a filter which absorbs the power instead of reflecting it back. In this article we show two methods for suppressing the harmonics of a microwave oscillator. In the first method a lumped element absorption network will be used at the output of the oscillator while in the second method a special filter will be applied. The AWR Microwave Office has been used to simulate these harmonic suppression methods.
In the first method a series and a parallel LC tank is used to suppress the power of the upper harmonics. This network absorbs the power of the 2nd and the 3rd harmonics instead of reflecting them back. The input reflection coefficient of the absorption network is -9 -16 dB at the mentioned harmonics respectively while the attenuation is about 11 dB at these frequencies.
Unfortunately this network has about 1-2 dB insertion loss at the fundamental harmonic but this is negligible compared to the upper harmonic attenuation.
In the second method a special filter is used to suppress a specified harmonic component. This filter reflects a specified harmonic in opposite phase back to the oscillator. In this paper we show two cases of this method. In the first case the 2nd harmonic and in the other case the 3rd harmonic is reflected in opposite phase. These filters are synthesized using open-circuited transmission line stub. These filters have a high reflection in a wide frequency range (-3 dB or higher in 2 or 3 GHz bandwidth) but only the specified frequency is reflected in opposite phase. Using this filter the phase noise of the fundamental harmonic has been improved and its power has not been changed unlike in the case of a common filter or absorption network.
The effect of the absorption network and the special filters are simulated and measured in the case of a negative resistance microwave oscillator. BFP420 bipolar transistor is used as an active element of the oscillator because this device is potentially unstable up to 4 GHz. The
77
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015 oscillator was designed to 2 GHz startup frequency while the steady-state oscillation frequency will be smaller. The oscillators were manufactured on FR4 substrate because this substrate is widely used in circuit design. However, this substrate is not optimal for microwave frequencies above 2 or 3 GHz. The steady-state frequency of the manufactured oscillator is about 1.78 GHz but the effect of the absorption network and the filters are similar to the simulation results.
Konstantin Weise 1 , Matthias Carlstedt 1 , Marek Ziółkowski 1,2 , Hartmut Brauer 1
1 Technische Universität Ilmenau, Germany
2 West Pomeranian University of Technology, Szczecin, Poland
konstantin.weise@tu-ilmenau.de
Abstract:
The present contribution addresses the underlying theory of the non destructive testing technique Lorentz force eddy current testing (LET) [1-2]. The technique is applied on moving nonmagnetic conductors to detect buried defects and inclusions by means of the Lorentz force acting on a magnet system which is in relative motion with respect to the object under test. The fundamentals of motion induced eddy currents are described by evaluating analytical expressions of the Lorentz force and eddy current density of a cylindrical conductor moving through a ring magnet system. The reaction of the magnetic field generated by the induced eddy currents is taken into account in the present analysis in order to provide accurate predictions even for high velocities. The presented results are supported by finite-element simulations.
In the presence of a defect, the induced eddy currents and hence, the Lorentz force acting on the magnet is perturbed. This response depends on the geometry and magnetization direction of the applied magnet system. In an optimization framework, the magnet system is optimized in order to maximize the absolute defect response signal. The optimal design depends on the defect parameters such as its size and location and hence on the predefined detection goal. Distinct cases of subsurface and deep lying defects are studied and optimal designs are presented.
Further studies are related to investigations on the role of uncertainties in the input parameters on the resulting Lorentz force. In this sense, the remanence of the magnet, the lift-off distance as well as the conductivity and velocity of the specimen were modelled as uniform distributed random variables. The uncertainty intervals are estimated by our laboratory setup. A non-intrusive generalized polynomial chaos expansion [3] is performed by means of numerical simulations in order to quantify the impact of each of the input uncertainties. Most important parameters are identified in order to improve the accuracy of our laboratory setup.
References
1. Brauer H., Ziolkowski M., Eddy current testing of metallic sheets with defects using force measurements , Serbian Journal of Electrical Engineering, vol. 5, no. 1, pp. 11-20, 2008.
78
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
2. Brauer H., Porzig K., Mengelkamp J., Carlstedt M., Ziolkowski M., Toepfer H., Lorentz force eddy current testing: a novel NDE-technique, COMPEL, vol. 33, no. 6, pp. 1965-1977, 2014.
3. Xiu D., Fast Numerical Methods for Stochastic Computations: A Review , Communications in
Computational Physics, vol. 5, no. 2-4, pp. 242–272, 2009.
Tamas Peto
Budapest University of Technology and Economics, Hungary peto@evt.bme.hu
Keywords: passive radar, beamforming, SDR, DVB-T, DPIS
Abstract:
Passive radars exploit the already existing signals in the environment to detect targets.
The available sources of illumination has a very wide range both in frequency of operation and bandwidth. The capability of fast and effective reconfiguration has especially great importance on this field. Software defined radio systems can be flexible enough to receive any of the illuminator signal with the desired bandwidth. In the same time multichannel operation is also a key component of passive radars as they must able to apply digital beamforming techniques to achieve high dynamic range. Combining these technologies the presently available passive radars surveillance capabilities can greatly increase. This paper addresses the development of a multichannel software defined passive radar receiver. The effective operation of the receiver platform is presented through the design, implementation and measurement results of a DVB-T
(Digital Video Broadcast - Terrestrial) based passive system.
79
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
80
Sunday 17.05
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Piotr Graca, Bronisław Tomczuk
Opole University of Technology, Poland p.graca@doktorant.po.edu.pl
Keywords: axial active magnetic bearing, finite element method, electromagnet, magnetic flux density, magnetic force, current stiffness, displacement stiffness
Abstract:
In recent years, Active Magnetic Bearings (AMBs) have become a very good alternative for mechanical (rolling-element) bearings, oil bearings or gas bearings, normally used in highspeed rotary machines. Thanks to such advantages as lack of lubrication, lack of friction, vacuum operation, wide temperature ranges and easy integration with electric drives, the AMBs are nowadays applied in many different industry branches.
One of the important application of AMBs is an electro-spindle used in high-speed milling machines. A classical construction of the electro-spindle consist of two radial and one axial
AMBs. The radial bearings are responsible for suspending the rotor along machine axis and to control a constant air gap between rotor and stator. The axial bearing holds the axial position of the rotor. Such a construction is currently investigated at Department of Industrial Electrical
Engineering at Opole University of Technology. Most of the research focuses on magnetic field calculations and control system development.
In this paper, a prototype of Axial Active Magnetic Bearing (AAMB) is presented. It has been designed as a pair of two identical round shaped electromagnets, used to suspend a rotor in magnetic field. Based on the analysis of the magnetic field and calculation of its integral parameters, applicable dimensions of the particular parts have been proposed. Magnetic flux density calculation has been estimated using Finite Element Method (FEM) where nonlinear characteristic of ferromagnetic core was also taken into account. Magnetic force was calculated using the field distribution. The calculation results were confirmed by experimental tests on a physical model and allowed to estimate the coefficients like current and displacement stiffness.
Finally, the obtained coefficients have been used to develop a calculation model of the object, including PID controller which stabilises the rotor position. The simulation and test results are presented in the paper as well.
83
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Vladimir Chudacik, Milan Smetana
Department of Electromagnetic and Biomedical Engineering,
Faculty of Electrical Engineering, University of Zilina, Slovakia vladimir.chudacik@fel.uniza.sk
Keywords: eddy current testing, electromagnetic non-destructive method, response signals
Abstract:
Eddy current method (ECT) belongs to the electromagnetic non-destructive testing
(NDT) and evaluation (NDE) methods. The main purpose of NDT is detection of various material inhomogeneities or defects, without mechanical damage of a tested object. Non-destructive evaluation extends that purpose, because the maintenance in industry requires not only detection but also characterization of a detected defect. In NDE it is ambition to classify and quantify defects that can be detected in various structures. Many industrial applications require complex inspection where the main aim is to prevent a failure of the components, devices and equipments. Non-
Destructive methods offer this possibility of periodic and thorough inspection of industry products, which ensures their safe, effective and long-term operation. ECT method can be applied to conductive materials. The advantages include high sensitivity, rapid scanning, contactless inspection and versatility. Development of the computer technology caused the problems that had been difficult to solve practically can now be solved by numerical methods. And it is possible to very precisely and rapidly perform forward numerical simulations concerning different applications of the ECT.
The full paper will present the numerical simulation and experimental results performed on the conductive plate specimen, using ECT probe. The main aim of the work is to show the effect of material defect geometry variations on the ECT response signals. The eddy currents are excited by the absolute circular coil type. The probe is positioned normally to the plate surface with lift-off of l=1 mm and at angle of φ= 45° at different positions, respectively. It is expected that more complex information about the defects will be obtained.The numerical simulations will be performed using the OPERA 3D software. A conductive plate specimen will have the electromagnetic parameters of the stainless steel AISI 316L, with conductivity of σ = 1.35 MS/m and relative permeability of μr = 1. Several defects will be tested and compared with standard position. To verify the simulation results, experimental measurements will be taken on the same material. The experimental measurements will be compared each other and also with simulated data to the overall assessment of the impact and utility of this approach.
84
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
Zoltán Szalay, Lajos Nagy
Budapest University of Technology and Economics, Hungary szalay@mht.bme.hu
Keywords: dielectric thin-film, non-destructive testing, RF measurement, coaxial resonator
Abstract:
Recently, the measurement of layer-thickness becomes more and more important in industrial and civilian applications. There are numerous different methods (i.e. resonant, capacitive, inductive, time domain reflectometer, etc.) which are suitable to determine the thickness of a dielectric thin-film covered by metal at one side in a nondestructive way. However, the various phenomena provide distinct accuracy and measurement range. The presented quarterwavelength, open-ended coaxial based method belongs to the class of the resonant techniques. In previous articles, it has been showed that the resonator probe is able to measure the distance between the open-end and a dielectric or metal load with high resolution. Measuring the transmission of the resonator in a frequency range, a resonance curve can be created. The resonance frequency and the quality factor of the cavity change when the probe is loaded. The variations of these parameters are analytically expressible by the complex permittivity and thickness of the sample material. The developed analytical model is based on transmission-line theory and makes it possible to evaluate the results efficiently.
A 3D FEM (Finite Element Method) model of the cavity was also created to compare the results of the analytical model with numerical calculations. Based on the simulation results the highest resolution is approximately 1 micron while the active range of this method is 0.5 cm. By increasing the distance between the probe and target, the accuracy becomes lower. The measurement requires smooth-surfaced sample, which is wider than the diameter of the probe.
As opposed to VNA (Vector Network Analyzer) based sensors, our system measures the scalar scattering parameter utilizing a coaxial resonator, as well as two commercial sub-gigahertz radio modules which are developed for general communication. The excitation and probing of the resonator are realized by two magnetic loops, which are formed from the inner conductor of the coaxial line. The induction of the resonator is performed by a transmitter radio module, while the magnitude of the transmission parameter is measured by a receiver module. To create the resonance curve, a frequency sweep is performed with sinusoidally varying field between 480-
900 MHz.
The perturbation based operating principle of the sensor has been tested using distinct loads.
The experimental and the theoretical results show prominent matching, therefore we propose to measure the layer-thickness of metal covered dielectric material by the coaxial resonator based method. In addition to high accuracy and range, the simple and cost-efficient evaluation of the thickness makes the presented sensor system particularly interesting for industrial and common applications.
85
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
86
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
87
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
88
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
89
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
90
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
EMC test stand Tests of discharge lamp
Ulbricht sphere. Measurement of light flux Electro-tools noise test
Short-circuit generator at power of 2500
MVA
Short-circuit transformers
Internal arc fault in the switchgear.
View of the arc
View of the „very hot copper rain” from busbars
91
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
The aim of the International Interdisciplinary PhD Workshop is to gather PhD students and encourage them to discuss on problems related to their scientific work. Scope of topics is very wide and includes all research areas realised at the Faculties of: Electric, Electronic, Automatic, Mechatronics,
Bioengineering and Computer Sciences. Taking part in the Workshop can be valuable experience for
PhD students in their career as a researcher. It also gives an opportunity of integration with scientific society and helps to develop informal contacts. Session chairs are the most famous experts in the fields the Workshop is devoted for. Being at the Workshop is the only way to meet them and ask some questions.
The organizers are: Electrotechnical Institute Warsaw, Institute of Electronics and Information
Technology of Lublin University of Technology, AGH University of Science and Technology, Brno
University of Technology the Faculty of Electrical Engineering and Communication and Slovak
University of Technology in Bratislava Faculty of Electrical Engineering and Information Technology.
92
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
93
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
94
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
95
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
96
Interdisciplinary International PhD Workshop 2015 — IIPhDW 2015, Międzyzdroje 14 — 17 May 2015
97