基于改进粒子群算法的横动伺服控制系统辨识

摘要

In view of the black box model traverse servo control system,an identification method based on improved particle swarm optimization algorithm was proposed.First,the establishment of a system of fifth order transfer function model.Secondly,based on the particle swarm optimization algorithm,the introduction of a nonlinear concave function as a strategy of inertia weight adjustment,so as to avoid the algorithm in search of falling into the local optimum in the process of optimization,the optimization of the system model.Finally,in order to validate the identification model was correct,comparing simulation and actual measurement experiment.The experimental results showed that:the algorithm model was accurate and reliable,with good control quality,for the same speed input signal,simulation and measurement of the output curve tracking was good,the error in the-0.5～0.2 rad range,the error small.%针对横动伺服控制系统的黑箱模型,提出了一种基于改进粒子群算法的辨识方法.首先,建立了系统的五阶传递函数模型,其次,在粒子群算法的基础上,引入非线性凹函数作为惯性权重的调整策略,从而避免了算法在寻优过程中陷入局部最优,实现了系统模型的优化.最后,为了验证辨识模型的正确性,进行仿真与实测对比试验.试验结果表明:该算法辨识出的模型准确度较高,具有较好的控制品质,对于同一速度输入信号,仿真与实测的输出曲线跟随性好,误差在-0.5～0.2 rad范围内,误差小.