An Plurality Adaptive Evolutionary Programming and Extracting Fuzzy Rule Bases Automatically
Debao Chen · Dianzi xuebao · 2007
A new adaptive evolutionary programming,in which population is encoded by binary and complex-valued,is presented.Two layers chromosomes are arranged to express rule bases.In the first layer,binary gene is used to express the structure of fuzzy rule bases.In the second layer,complex-valued gene is used to determine the parameters of fuzzy rule bases.Adaptive evolutionary programming is used to optimize the structure and parameters simultaneously.The simulation experiments to tap fuzzy rule bases of nonlinear system and fuzzy PID controller automatically are done when prior experience knowledge is unknown.The results demonstrate that the given method is efficient.