Sw/Document type
Sida/Page
Thesis proposal: -Improved data regression methods for material characterization
Titelförslag/Thesis title
Investigate usage of Shape Language Modeling (aka Generative Modeling
Language) for material data fitting
Tidsperiod och högskolepoäng/Period of time and amount of credits
Antal studenter/Number of students
20 weeks,
1
Geografisk placering/Location
Trollhättan
Kontaktperson/Contact person
Språk/Language
1 (2)
Thomas Hansson
Startdatum/Start date
English/Swedish
Handledare/Supervisor
Thomas Hansson
Skicka ansökan till/Send application to
thomas.no.hansson@gknaerospace.com
Avdelning/Department
9633
Sista ansökningsdag/Last application date
Om oss/About us
GKN Aerospace is the aerospace operation of GKN plc, serving a global customer base and operating in North America and Europe. With sales of
£1.5 billion in 2011, the business is focused around three major product areas - aerostructures, engine products and transparencies, plus a number
of specialist products - electro-thermal ice protection, fuel and flotation systems, and bullet resistant glass. The business has significant
participation on most major civil and military programmes. GKN Aerospace is a major supplier of integrated composite structures, offers one of the
most comprehensive capabilities in high performance metallics processing and is the world leading supplier of cockpit transparencies and passenger
cabin windows.
Bakgrund till examensarbetet/Background of thesis project
GKN Aerospace has extensive mechanical testing programmes for defining material properties. Currently a curve is fitted to the raw test data using
a combination of filtering and a least squares method. In order to more intelligently fit our data to material models we wish to investigate the use of
fitting splines to our data, controlling the shape of the curve using derivatives, controlling knots, number of polynomes, smoothness, monotony etc.
The aim is to have the option of a more flexible and robust curve fitting method than currently used methods.
Uppdragsbeskrivning/Assignment description
The thesis work will focus on ,but is not limited to, these areas.


Crack propagation tests are usually run using a DC potential drop technique. The tests usually generate potential drop vs number of
cycles data points. This work is to Investigate if it is possible to improve the modeling of da/dN vs ∆K data by creating a mathematical
model for the potential drop vs cycle data for every specimen. This model will be used to create crack size vs cycle data and then finally
a da/dN vs ∆K .
Tensile tests are represented by stress vs strain datapoints for a number of test specimens. The task is to create a mathematical model
that represents the average and minimum stress-strain curves representing all the specimens tested at the same temperature.
Mål/Target

Mathematical models and evaluation techniques that improves the capability to characterize the material properties.
Kvalifikationer/Qualifications



Master student
Interested in using advanced mathematics to solve practical problems.
Interest in material science and computer programming (Matlab and possibly Python [Numpy/Scipy] are the tools suggested to be
used).
Ansök genom att/Apply by
5972 Utg 4

Förslagsvis genom att skicka in CV och personligt brev/for instance by sending CV and personal letter, etc.
5972 Utg 4