A fuzzy neural network system modeling method based on data-driven
Keyong Shao, Xin Fan, Shengmei Han, Shaofeng Li · 2010
The algorithm utilized only input-output data from the system to determine the proper control model, and not require a mathematical or identified description of the system dynamics. A fusion algorithm that based on subtraction clustering and fuzzy C-means algorithm(FCM) was proposed to identify the former network, automatically obtained precise cluster number and membership parameters, used the steepest descent method to train the weights of the after network, thereby set up a T-S fuzzy neural networks system model, a nonlinear system was used to illustrate this method. Simulation results demonstrate the effectiveness of the proposed identification methods.