基于Levant微分器的磁浮球控制算法研究
Maglev Ball Control Algorithm Based on Levant Differentiator
张振利 1汪永壮 1秦耀 1杨杰1
作者信息
- 1. 江西理工大学电气工程与自动化学院,江西赣州 341000;江西理工大学永磁磁浮与轨道交通研究院,江西赣州 341000
- 折叠
摘要
针对永磁电磁混合悬浮系统控制中因信号突变和噪声干扰引起的控制效果不理想等问题,提出一种改进型Levant微分器和PID相结合的控制方法(improve Levant-PID,ILevant-PID).融合PID控制适应性强和Levant微分器对输入噪声的鲁棒特性优势,解决系统输出抖振问题,利用模拟退火-粒子群优化解决ILevant-PID控制器参数多、关联性强等制约问题.仿真实验表明:与传统PID控制方法相比,ILevant-PID控制方法在阶跃输入下,启动更为平缓,调节时间减小了41.19％,超调量减小了 40.36％.ILevant-PID控制器在阶跃输入时有无噪声条件下的稳态误差分别为±0.37 mm及±0.23 mm,相对于PID降低了 87％以上.在跟踪方波输入时,ILevant-PID可实现PID无法对8 mm给定信号无超调跟踪,较好地改善PEMS(permanent magnet electro magnetic hybrid suspension)系统控制性能.
Abstract
To solve the problem of unsatisfactory control effect of permanent magnet electromagnetic hybrid suspension system caused by signal mutation and noise interference,the control method ILevant-PID,the combination of an improved Levant differentiator and PID,is proposed.The proposed method combines the strong adaptability of PID control and the robust characteristic of Levant differentiator on input noise to solve the chattering problem of the system output.The simulated anneal-particle swarm optimization is utilized to solve the constraints of the ILevant-PID controller,such as multiple parameters and strong correlation.The simulation results show that compared with the traditional PID control method,the ILevant-PID control method starts more gently under the step input,the adjustment time is reduced by 41.19％,and the overshoot is reduced by 40.36％.Experimental verification shows that under the condition of noiseless step input,the steady-state errors of the ILevant-PID controller are±0.37 mm and±0.23 mm respectively,which are more than 87％lower than PID.When tracking square wave input,ILevant-PID can realize non-overshoot tracking of 8 mm given signal that cannot be achieved by PID,which can improve the control performance of the PEMS system.
关键词
混合悬浮系统/ILevant-PID控制器/模拟退火算法/粒子群优化算法/控制性能Key words
PEMS(permanent magnet electro magnetic hybrid suspension)system/ILevant-PID(improve Levant-PID)controller/simulated annealing algorithm/PSO algorithm/control performance引用本文复制引用
出版年
2024
NETL
NSTL
万方数据