摘要
研究复卷机放张力控制中经典PID控制器参数整定优化问题,PID控制器的参数决定了张力控制系统的稳定性和快速性。提出用粒子群算法优化整定PID控制器的参数,并针对粒子群算法的易陷于局部收敛的缺点,提出用遗传算法改进粒子群算法。该方法使用时间乘绝对误差准则作为目标函数,利用遗传算法的交叉、变异操作和优胜劣汰机制,保持粒子群算法的粒子的多样性,改善粒子群算法全局收敛能力。复卷机放卷过程采用闭环PID张力速度控制,做到恒张力控制,通过建立动态数学模型和利用动态转矩平衡方程,分析出卷材上的张力受到放卷轴线速度和卷径的影响。利用MATLAB软件,分别使用改进粒子群算法和普通粒子群算法仿真整个控制过程,通过对比分析控制性能指标,改进后的粒子群算法仿真结果响应速度较快、没有峰值时间和超调量、快速达到稳定输出,扰动调节能力较好。
The parameter tuning and optimization problem of the classic PID controller in the tension control of the rewinder was studied.The parameters of the PID controller determine the stability and rapidity of the tension control system.It was proposed to use particle swarm algorithm to optimize and tune the parameters of PID controller.Aiming at the shortcoming of particle swarm algorithm that was easy to fall into local convergence,genetic algorithm was proposed to improve particle swarm algorithm.The method used the time multiplied by absolute error criterion as the objective function,and used the crossover,mutation operation and survival of the fittest mechanism of genetic algorithm to maintain the diversity of particles of the particle swarm algorithm and improved the global convergence ability of the particle swarm algorithm.The unwinding process of the rewinder adopted closed-loop PID tension speed control to achieve constant tension control.By establishing a dynamic mathematical model and using dynamic torque balance equations,it was analyzed that the tension on the coil is affected by the speed of the unwinding axis and the coil diameter.Using MATLAB software,the improved particle swarm algorithm and the ordinary particle swarm algorithm were used to simulate the entire control process.By comparing and analyzing the control performance indicators,the improved particle swarm algorithm has a faster response speed,no peak time and overshoot,and reaches quickly,stable output,and better disturbance regulation ability.
作者
谭龙彪
肖金凤
Tan Longbiao;Xiao Jinfeng(Department of Electrical Engineering,University of South China,Hengyang,Hunan 421200,China)
出处
《机电工程技术》
2022年第3期182-186,共5页
Mechanical & Electrical Engineering Technology
关键词
PID控制器
粒子群算法
遗传算法
张力控制
PID controller
particle swarm algorithm
genetic algorithm
tension control
