面向电子压力机模糊神经网络PID控制的研究

针对电子压力机位置伺服系统的非线性和时变的不确定性,压装力、压装速度和压入深度高可控性,系统的高稳定性、适应性及较强的抗干扰能力等特点,提出将神经网络实现模糊PID自调整的控制特性应用在现存的小型电子压力机的位置伺服系统中的方法。该控制策略将模糊控制的推理能力和神经网络的学习能力进行了有效的结合,其中,PID控制器参数自调整是通过学习并记忆PID参数调整的基本规则来实现的,以满足电子压力机位置伺服系统的要求并用MATLAB软件编程进行仿真分析。仿真结果表明:相比较常规神经网络与传统PID相结合组成的控制器,模糊神经网络PID自调整控制器对于电子压力机的位置伺服系统具有更快的响应特性及更好的稳定性。

Electronic Press Oriented Research of Fuzzy Neural Network PID Control

Electronic press position servo system with the properties of nonlinear and time-varying uncertainty requires good con-trollability of press mounting force, press speed and pressed depth, good system stability, adaptability and anti-interference ability. Ac-cording to these, a method was proposed of self-tuning fuzzy proportion integration differential ( PID) control characteristics, realized by neural network, was applied to a mini electronic press position servo system. The control strategy was combined of the reasoning a-bility of fuzzy control and the learning ability of neural network effectively. Through learning and memorizing the basic rules of PID pa-rameter adjustment, the self-tuning of PID controller parameters could be realized so as to meet the requirements of the position servo system of the press, and simulation analysis was carried out by using MATLAB software programming. The simulation results show that this fuzzy neural network PID controller as compared with that composed of conventional neural network combined with traditional PID, have better stability and faster response characteristics for the position servo system of electronic press.