Compound fuzzy neural network controller with application to the vibration control of the mold of continuous casting machinery

摘要

The mechanism and the characteristics of the three kinds of way (symbol and logic, connection, and evolution) to realize intelligent control are surveyed, a through research on the fuzzy neural networks (FNN) controller with application to electrohydraulic servo systems is carried out. Based on analyzing the characteristics of the current two kinds of FNN and their problems, a new type of FNN ocntroller is presented, and the synthesis design method of the control law and fast speed learning algorithm on line of the parameters of networks are put forward. The output of this kind of controller is composed of two parts, part one is derived on basis of the principle of slidingcontrol, it only requires the lower order model and the estimated parameters of the plant, part two is derived on basis FNN, it is used to compensate the uncertainties of the systems. The control law and the formula of its adjusting parameters are presented by means of a new type of FNN performance index function and teh synthesis method of control law. The rightness of the conclusions is verified by the experiment of an electrohydraulic position servo system of the mold of the continuous casting machinery.