Acceleration of the parameters identification for dynamic models construction using parallelization

: pp. 7-12
Lodz University of Technology
Lviv Polytechnic National University
Lviv Polytechnic National University

Construction of mathematical models for nonlinear dynamical systems using optimization requires significant computation efforts to solve the optimization task. This makes it reasonable to use parallelization of calculations for optimization task solving, especially taking into account current tendency for increasing the number of CPU cores in a single chip. The effectiveness of particular parallel implementation of optimization process is the subject of investigation in this paper.

  1. L. Byczkowska-Lipinska, P. Stakhiv, Y.Kozak, Parallelization of Calculations in Construction of Mathe­matical Models using Optimization // In Proc. XIII International Conference “System Modelling and Control” SMC 2009. – Zakopane, Poland. – 2009.
  2. P. Stakhiv, B. Melnyk, Y. Kozak, Simplification of Optimization Process during Mathematical Models Creation // Przegląd Elektrotechniczny – № 12. – 2008. – P. 281-283.
  3. Y. Kozak, Modification of Rastrigin Method of Direction Cone // Elektronika i svyaz. Special edition on Problems and Physical and Biomedical Electronics. – Kyiv, Ukraine: Publishing House of Kyiv Polytechnic Institute. – 1997. – P. 424. (Ukrainian)
  4. M. Avriel, Nonlinear Programming: Analysis and Methods // New York, USA: Dover Publications. – 2003.
  5. L. Ljung, System Identification: Theory for the User. – New Jersey, USA: Prentice Hall – 1999. – 609 p.
  6. J. Nelder, R. Mead, A Simplex Method for Function Minimization // The Computer Journal. – Oxford, UK: Oxford University Press. – № 7. – 1965. – P. 308–313.