Identification of bilinear systems using differential evolution algorithm

Abstract

In this work, a novel identification method based on differential evolution algorithm has been applied to bilinear systems and its performance has been compared to that of genetic algorithm. Box-Jenkins system and different type bilinear systems have been identified using differential evolution and genetic algorithms. The simulation results have shown that bilinear systems can be successfully and efficiently identified using these algorithms.

Optimization of weighted myriad filters with differential evolution algorithm

“…Differential Evolution Algorithm (DEA) is verified to be an efficient method for solving optimization problems in literature [20][21][22][23][24][25][26][27][28][29]. Basically, DEA has six important parameters as population size (NP), generation number (GN), combination factor (CF), scaling factor (F), upper and low bounds (ULB), crossover rate (C).…”

“…Also DEA is easily implemented by computer programs. So DEA is successfully applied to many different fields and system optimization problems [22][23][24][25][26][27][28][29][30]. But DEA has several disadvantages such as unstable convergence in the last period and easy to drop into local minimum [29].…”

“…Flow chart of DEA As seenFigure 1, mutation, crossover and selection are important operators of DEA[27][28][29][30].…”

Yapay Ari Koloni̇si̇ Algori̇tmasi İle Opti̇mi̇ze Edi̇len Hammerstein Model Kullanarak Si̇stemleri̇n Ki̇mli̇klendi̇ri̇lmesi̇

“…Linear methods can be inadequate in identification of such systems and nonlinear methods are used [6][7][8][9][10][11][12][13]. In nonlinear system identification, the input-output relation of the system is provided through nonlinear mathematical assertions as differential equations, exponential and logarithmic functions [14][15][16][17][18]. Autoregressive, Autoregressive Moving Average (ARMA) models or finite impulse response (FIR) and infinite impulse response (IIR) models are used for linear system identification in literature.…”

Swarm and evolutionary computing algorithms for system identification and filter design: A comprehensive review

“…The authors have employed a frequency domain cost function and compared the results obtained with that using a simple GA. Parameter estimation of a permanent magnet synchronous machine under variable speed control has been attempted in (Liu and Zhu, 2014) using a quantum GA. Also, a few evolutionary computing based system identification methods have been presented lately for parameter estimation of bilinear (Wang and Gu, 2007;Modares et al, 2010b;Zorlu, 2011) andhysteretic systems (Kyprianou et al, 2001;Ye and Wang, 2007;Charalampakis and Koumousis, 2008;Ye and Wang, 2009;Worden 36 andManson, 2011, 2012;Worden and Barthorpe, 2012;Liu et al, 2012;Ortiz et al, 2013;Charalampakis and Dimou, 2013;Quaranta et al, 2014).…”