基于PSOGWO的磁悬浮球系统LADRC控制方法OA北大核心CSTPCD

In order to solve the problem that maglev ball system is susceptible to external disturbance and diffi-cult to establish accurate mathematical model,a linear active disturbance rejection controller independent of precise model of the system is applied to control system.The unknown disturbances of maglev ball system are estimated and compensated by using a linear extended state observer and a linear state error feedback control law.At the same time,aiming at the problem that parameters of linear active disturbance rejection controller are difficult to be tuned to reach the optimal state,a grey wolf optimizer based on hybrid particle swarm optimization algorithm is proposed.The advantages of fast convergence of particle swarm optimization algorithm are integrated into grey wolf optimization algorithm with strong global searching ability to complete parameter optimization.The results show that the linear-active disturbance rejection strategy optimized by hybrid algorithm is superior to PID control in response speed,following and anti-interference performance.Specifically,the setting time in step re-sponse simulation is reduced to 0.033 s;the position deviation in fixed-point levitation experiment is decreased by 80％.