A comparative study of Space Search Evolutionary Algorithm (SSEA) and Differential Evolution (DE) in the design of IG-based fuzzy models
Wei Huang, 오성권 · 대한전기학회 학술대회 논문집 · 2010
In this study, we propose a space search evolutionary algorithm (SSEA). In comparison with “conventional”” optimization algorithms (such as the well-known differential evolution), SSEA leads to better search capabilities in case of search spaces of high dimensionality. Next, we use SSEA as the optimization vehicle for the design of fuzzy inference systems. A hybrid identification method of fuzzy inference systems is proposed. The overall hybrid identification comes in the form of two optimization mechanisms: structure identification and parameter identification. The structure identification is supported by the SSEA and C-Means while the parameter estimation is realized via SSEA and a standard least square error method. The fuzzy models are designed based on information granulation (IG) which is realized with the aid of the C-Means. Information granulation helps determine the initial values of the apex parameters of the membership functions of the fuzzy models. The evaluation of the performance of the proposed model is carried out by using a series of examples such as NOx emission process and Mackey-Glass time series. A comparative study of SSEA and DE demonstrates that the SSEA leads to improved performance. The proposed model is also contrasted with the quality of some “conventional” fuzzy models already encountered in the literature.