A new approach to fuzzy-neural-network modeling
Pin Zhang · Journal of Xi'an University of Architecture & Technology · 2000
Based on zero order takagi sugeno model,a simple yet effective approach to fuzzy neural network (FNN) modeling for complex systems with input output data is presented.The main purpose of the present method is to solve the difficulty of structure identification in FNN.A so called linguistic model is used to find the optimal combination of system input variables.Then,in the stage of input variable selection,we neither estimate the parameters of the fuzzy model nor determine the number of the fuzzy rules.Since an FCM based self adaptive fuzzy clustering technique is employed to determine the proper structure of the FNN and set the initial weights in advance,the network can be trained rapidly.Furthermore,a modified membership function is also provided to improve the identification accuracy of the FNN.Effectiveness of the present method is verified by a simulation for a practical modeling.