科研之友
微信
新浪微博
Facebook
分享链接
摘要
It is well known that asymmetric cylinder fifth-order electro-hydraulic servo system is a class of typical high-order one with non-linearity, time-variation and gain asymmetry. And since dynamic performances of the servo system vary greatly with the variable loads, conventional PID is very difficult to be used to reach the upper desired performance target, but if single neuron PID controller designed in terms of BPNN Theory is adopted. the control effect is not satisfactory because the learning rate is low, the speed of convergence is slow.20-SIM simulation software is a simulation tool established on bond graph theory, and it is very suitable for simulation of non-linear systems. In this paper, based on the aforementioned characteristics, we set up a simulation model using 20-SIM simulation tools, and design a on-line self-learning reformed single neuron PID controller. Simulation results compared with the ones of conventional PID control and single neuron PID controller show that this reformed single neuron PID control strategy is better than the other two, control effect is satisfactory.
