DETERMINISTIC METHOD FOR OPTIMIZING VSWR
Jozef BUCHWALD, **Juraj PÚČIK
*
*Department
of Computers and Informatics, Faculty of Electrical Engineering and Informatics,
Technical University of Košice, Letná 9, 042 00 Košice, tel. 095/602 3175, E-mail: buchwald@tuke.sk
**Department of Theoretical Electrotechnics and Electrical Measurement, Faculty of Electrical Engineering and Informatics,
Technical University of Košice, Letná 9, 042 00 Košice, tel. 095/602 2707, E-mail: pucik@tuke.sk
ABSTRACT
In the article presented we deal with behaviour system of an observed rotor that is a model of the finite element method
at a specific time. Time behaviours of faulty and failure-free controlling units (shaft), respectively, with different fault places
represent the dynamic behaviour. In such cases effects of the own weight and unbalance transpire. Thereafter, the induced
system is considered, with the nominal signal depending on the amplitude of harmonic or segmenting aberration. Model of
the finite element method stands for the "experimental" investigation method of the dynamic system behaviour if there are
neither the decomposition of systemic arrangement nor the accurate systemic reactions present.
It is impossible to anticipate and lay out all of the possible cases of faults. There is the measurement being the result of
last experiences attained on one side and the deterministically and stochastically unexpected factors on the other hand.
Besides, one´s experiences from the past are not proven and cannot serve as a prediction for the future. For example,
material detrition is slow and hardly to be calculated in the stage of projecting and planning. Countermeasures are needed
not only when projecting and planning but also during the operation. At the beginning they are obviously difficult to be
perceived by senses. Faults, as a part of this category, lead to shaft breaking and cause aggregate damages. This especially
concerns cases when broken parts are coupled with other part of the shaft. This is the problem in continuous operation.
Therefore, this is a way to optimally recognise fault symptoms and to localise them so that during operation avoided could be
danger to both operators and the environment and consequential costs could be saved. It is not easy to solve the problems
mentioned for there is still lack of universal methods of solving the fault diagnosis. Effective methods are still to be
elaborated [1].
Keywords: diagnosis, observer, fault, rotor, intensity, simulation.
fra  fzik  KzUi k (t )  DvUi k (t ), ik  7
1. INDRODUCTION
Kinetic behaviour of turborotor is followed up
at a rotor containing two hydrodynamically
adjustable bearings laid at the end of shaft. The rotor
consists of seven independent controlling units with
equal
average,
length
and
homogeneous
characteristics of the material. Systemic data for
geometry and material of an individual rotor unit are
described in the following parts of this paper.
2. BEARING
The bearing chosen can be characterised as
dynamic segmented bearings connected massively
with journals of a shaft at its both ends. The total
bearing weight is made up of journals of the shaft,
saucepans, segments and the coach-box bearing.
Shaft load, as well as an alternatively effective load
upon system is carried to coach-box bearing.
The average of the journals is identical with the shaft
average. Based on the theory of a short bearing [1],
we choose the close bearing, neglecting both
the fundamental oscillation and the oscillation
of bearing bearer. Equation (1) and (2) express that
the assumed dynamic intensity is equal to that
of journals (regardless bearings shifting) [2]
fla  fzik  KzUi k (t )  DvUi k (t ), ik  1
(1)
(2)
where U is axis of shifting, D is the matrix
of damping, K is matrix of the bearing resistant to t
fault and F is power of unbalance.
2.1. Systemic parameters
Systemic parameters comprise the geometry
of data and material and the operating parameters
of a resonant (oscillating system) system. Shaft
systemic parameters are as specified below:
1. Average: d e  0,25m
2. Distance of an individual controlling unit:
l e  2m
3.
4.
Number of individual controlling units: 7
Material of controlling centre: Sl 54
5.
E - modules: E  2,1  10 N / m
11
2
3. TIME BEHAVIOUR OF THE NO-FAULT
ROTOR
As it is evident from Fig. 1, the system
approaches in its centre the harmonic motional
reactions. Maximum coordinate divergence U 2 of
about 3.8 mm, being a result of inert system since
there is not only the effecting unbalanced intensity
but also the weight intensity in vertical direction.
Microwave and Wireless Technology 2004, September 13-14, Košice, Slovakia
8
The maximum amplitude U1  3.5 mm is less when
compared with U 2 . Missing weight intensity
represents physical reasons. In Fig. 1 there are also
the U 3 and U 4 – being the reactions of moments
at [Nm]. Maximum values are 0.5 Nm.
Figure 2 presents behaviours of the left
and the right bearing. Reactions of both
of the bearings resemble the reactions at axis
of U1 and U 2 , respectively. The same anomalies
may be considered the dynamic dislocation
of a bearing. Maximum values constitute half
of values of the centre (2 mm). The changing
coordinates moments U 3 and U 4 have opposite
marks but identical values. The reason can be found
in the symmetric arrangement. The motion bearing
reactions differ from the motion reactions
at the middle.
and quick abolishment an absolute must. In complex
technical facilities special attention should be paid to
a systemic fault due to possible massive damages.
Thanks to the computer capacity and thereby
increased exploitation of computer monitoring it is
possible to diagnose the fault simultaneously and
visually already at present.
REFERENCES
[1] Mrkvička, J : Diagnosis methods of mechanical
machineries using the parallel systems, Acta
Mechanica Slovaca, Volume 4, No. 3/2000,
Vienala, pp. 133-140, ISSN 1335-2393
(in Slovak)
[2] Petržlen, M : Modellbildung eines Komplizierten Systems mit Riß. Fascicola Matematica
– Informatica, Buletinul Stiintific al Universitatii din Baia Mare, vol. XVI, Nr. 11-12, 2000,
pp. 64-70.
BIOGRAPHY
Fig. 1 Time behaviour in the middle area of fault.
Fig. 2 Time reaction on the left and right bearing.
4. CONCLUSION
The fault in a complex system is considered
systemic fault. This fault is a risk factor because of
people security. Not only for safety and economics,
but also from ecological reasons is the diagnosing
Jozef Buchwald was born on 17.11.1966. In 1991 he
graduated (MSc.) with distinction at the department
of Computers and Informatics of the Faculty of
Electrical Engineering and Informatics at Technical
University in Košice. He defended his PhD. in the
field of programming device and systems in 2000;
his thesis title was "Diagnosis of compound systems
using the Data Flow applications". Since 1995 he is
working as a tutor with the Department of
Computers and Informatics. His scientific research is
focusing on parallel computers of the Data Flow
type. In addition, he also investigates questions
related with the diagnostics of complex systems.