基于模糊神经网络的PID控制器研究

PID Controller Based on Fuzzy Neural Networks

工业控制经常会用到一些经典的PID控制,PID控制有很多优点比如应用范围广、参数较易整定、鲁棒性好等。但是随着科技的发展,工业控制对控制对象的要求越来越高,PID控制有时候很难满足控制要求。因此文章在原有的PID控制器的基础上,加入模糊控制、神经网络控制,优化其控制系统。模糊神经网络PID控制器通过实验证明控制效果有很大的提升。模糊控制、神经网络控制结合一起起到互补的效果,比单独的PID控制系统性能更好。

Industrial control often uses some classic PID control, PID control has many advantages, such as wide application range, easier parameter tuning, good robustness. However, with the development of science and technology, the requirements of industrial control on the control object is getting higher and higher, PID control is sometime difficult to meet the control requirements. On the basis of original PID controllers this article adds in fuzzy control, neural network control to optimize its control system. PID controller based on fuzzy neural networks offers greatly improved control effect proved by experiments. The combination of fuzzy control and neural network can realize complementary effect, it has better performance than that of single PID control system.