Multiple Input Single Output (MISO) Process Optimization Using GA Based Fuzzy Clustering

Due to the unique characteristics such as handling complex, nonlinear, and sometimes intangible dynamic systems, fuzzy systems are used in the modeling of Input-Output data of the process. In this paper clustering algorithm is implemented in the design of a Fuzzy Logic Controller (FLC) and for the determination of the optimal values of clustering parameters such as weighting exponent and the number of clusters, Genetic Algorithm (GA) is used. Steel making process, a MISO process, is chosen here as an application example and GA based Minimum Cluster Volume (MCV) algorithm is proposed which minimizes the sum of the volumes of the individual clusters based on the elimination of redundant rules in the fuzzy rule base thereby reducing the rule firing and computational time and improving optimization

[1]  Seema Chopra,et al.  Identification of rules using subtractive clustering with application to fuzzy controllers , 2004, Proceedings of 2004 International Conference on Machine Learning and Cybernetics (IEEE Cat. No.04EX826).

[2]  W. Hills,et al.  Efficient Process Optimization , 2003, Concurr. Eng. Res. Appl..

[3]  Michio Sugeno,et al.  Fuzzy identification of systems and its applications to modeling and control , 1985, IEEE Transactions on Systems, Man, and Cybernetics.

[4]  Hyungsuck Cho,et al.  Genetic algorithm-based optimization of fuzzy logic controller using characteristic parameters , 1995, Proceedings of 1995 IEEE International Conference on Evolutionary Computation.

[5]  Jongwoo Kim,et al.  Clustering algorithms based on volume criteria , 2000, IEEE Trans. Fuzzy Syst..

[6]  M. Hossein Fazel Zarandi,et al.  A systematic fuzzy system modeling for scheduling of textile manufacturing system , 2007 .

[7]  Chaned Wichasilp,et al.  Design of Fuzzy Logic Controllers by Fuzzy c-Means Clustering , 2003 .