关键词: 粒子群优化(PSO), 前向神经网络, 液压系统, PID

Abstract:

In view of the high energy consumption and low response of the traditional hydraulic system, a permanent magnet synchronous motor driven constant pump hydraulic system was designed instead of common motor, which was easy to realize closed-loop feedback control. Because of the serious random interference, multi-variable, nonlinear, strong coupling, it was difficult to establish precise mathematical model of hydraulic system. General PID was difficult to meet high precision control requirements, in this respect a forward neural network PID controller based on PSO-BP hybrid optimization algorithms for adaptive control of hydraulic system was proposed to realize precise control of flow under complex conditions. The control parameters of PID controller were adjusted adaptively by neural network, the weights of neural network were optimized by the mixed learning methods. An optimized non-linear combined control rule of P, I, and D was built by the self-learning ability of neural network and to strengthen the ability of changes in working condition. Simulation and experimental results show that: the controller has fast tracking ability, small overshoot and strong robustness. A new method is provided for the high-precision flow control of hydraulic system.

Key words: particle swarm optimization(PSO), feed-forward neural network(FNN), hydraulic system, PID