Differential evolution and quantum-inquired differential evolution for evolving Takagi–Sugeno fuzzy models

• DOI: 10.1016/j.eswa.2010.11.107
• Publication Date: 2010-11-25T04:47:05Z
• Authors: Haijun Su; Yupu Yang
• Keywords: Differential Evolution; Benchmark (surveying)

Research highlights? DE/QDE can solve numerical and binary optimization problems. ? DE/QDE can simultaneously optimize the structure and the parameters of the model. ? A new encoding scheme is given to allow DE/QDE to be easily performed. The differential evolution (DE) is a global optimization algorithm to solve numerical optimization problems. Recently the quantum-inquired differential evolution (QDE) has been proposed for binary optimization. This paper proposes DE/QDE to learn the Takagi-Sugeno (T-S) fuzzy model. DE/QDE can simultaneously optimize the structure and the parameters of the model. Moreover a new encoding scheme is given to allow DE/QDE to be easily performed. The two benchmark problems are used to validate the performance of DE/QDE. Compared to some existing methods, DE/QDE shows the competitive performance in terms of accuracy.