Chapter

Applications of Evolutionary Computing

Volume 3907 of the series Lecture Notes in Computer Science pp 805-809

Prudent-Daring vs Tolerant Survivor Selection Schemes in Control Design of Electric Drives

  • Ferrante NeriAffiliated withDipatimento di Elettrotecnica ed Elettronica, Politecnico di Bari
  • , Giuseppe L. CascellaAffiliated withDipatimento di Elettrotecnica ed Elettronica, Politecnico di Bari
  • , Nadia SalvatoreAffiliated withDipatimento di Elettrotecnica ed Elettronica, Politecnico di Bari
  • , Anna V. KononovaAffiliated withLaboratory of Theoretical Probabilistic Methods, Yaroslavl’ State University
  • , Giuseppe AccianiAffiliated withDipatimento di Elettrotecnica ed Elettronica, Politecnico di Bari

* Final gross prices may vary according to local VAT.

Get Access

Abstract

This paper proposes and compares two approaches to defeat the noise due the measurement errors in control system design of electric drives. The former is based on a penalized fitness and two cooperative-competitive survivor selection schemes, the latter is based on a survivor selection scheme which makes use of the tolerance interval related to the noise distribution. These approaches use adaptive rules in parameter setting to execute both the explicit and the implicit averaging in order to obtain the noise defeating in the optimization process with a relatively low number of fitness evaluations. The results show that the two approaches differently bias the population diversity and that the first can outperform the second but requires a more accurate parameter setting.