Study on Adaptive Fuzzy Control System Based on Gradient Descent Learning Algorithm

摘要

For the general two-dimensional fuzzy controller, the output control value is a rigid function with respect to the error and the change in error, so it is difficult to obtain the desired effects for the plant that is uncertain and time-varying. Aimed at the problem, the paper applies gradient descent learning algorithm to correct the mean and variance of Gaussian membership functions of all fuzzy sets in the input and output universes, in this way, the fuzzy control system is adaptive. Selecting input signals as step, ramp, acceleration and sine signals, respectively, all simulation studies were carried out. The results demonstrate that the control algorithm is feasible, and its effect is better than that of the fuzzy control system without adaptability.