EAL, Institut für elektrische Antriebe und Leistungselektronik

Institut für elektrische Antriebe und Leistungselektronik

JKU, Johannes Kepler Universität Linz

Sprache: DE

Publikations Einzelansicht

Hybridization of Multi-Objective Evolutionary Algorithms and Artificial Neural Networksf or Optimizing the Performance of Electrical Drives


Autor(en): Alexandru-Ciprian Zăvoianu, Gerd Bramerdorfer, Erwin Lughofer, Siegfried Silber, Wolfgang Amrhein und Erich Peter Klement
Journal: Engineering Applications of Artificial Intelligence 26 ( 2013) 1781–1794
Editor: www.elsevier.com/locate/engappai Elsevier
Jahr: 2013
Band: 26
Seite(n): 1781–1794
Zusammenfassung: Performance optimization of electrical drives implies a lot of degrees of freedom in the variation of design parameters,which in turn makes the process overly complex and sometimes impossible to handle for classical analytical optimization approaches. This,and the fact that multiplenon-independent design parameter have to be optimized synchronously, makes a soft computing approach based on multi-objective evolutionary algorithms(MOEAs)afeasible alternative.In this paper,we describe the application of the wellknown Non-dominated Sorting Genetic Algorithm II(NSGA-II)in order to obtain high-quality Pareto-optimal solutions for three optimization scenarios. The nature of these scenarios requires the usage of fitness evaluation functions that rely on very time-intensive finiteelement(FE)simulations.The key and novel aspect of our optimization procedure is the on-the-fly automated creation of highly accurate and stable surrogate fitness functions based on artificial neural networks(ANNs).We employ these surrogate fitness functions in the middle and end parts of the NSGA-II run (hybridization) in order to significantly reduce the very high computational effort required by the optimization process.The results show that by using this hybrido ptimization procedure,the computation time of a single optimization run can be reduced by 46–72%while achieving Pareto-optimal solution sets with similar,or even slightly better,quality as those obtained when conducting NSGA-II runs thatuse FE simulations over the whole run-time of the optimization process.

« zurück