基于粒子群优化的低阶时滞系统辨识  

Identification of Low-Order System with Time Delay Based on Particle Swarm Optimization

作　　者：李敏花[1] 柏猛[1] 吕英俊[1] LI Minhua;BAI Meng;Lv Yingjun(Department of Electrical Engineering and Information Technology,Shandong University of Science and Technology,Jinan 250031)

机构地区：[1]山东科技大学电气信息系

出　　处：《模式识别与人工智能》2019年第6期524-530,共7页Pattern Recognition and Artificial Intelligence

基　　金：山东省自然科学基金项目(No.ZR2014FQ020)资助~~

摘　　要：为了解决低阶时滞系统阶跃响应辨识问题,提出基于粒子群优化的参数估计方法.方法主要包括参数初值计算和参数估计两部分.首先,采用积分方程方法估计时滞系统参数初值,通过设置参数初值估计误差,得到系统参数取值范围.然后,为了减小由观测噪声引起的参数估计误差,采用粒子群优化算法优化模型参数.最后,通过仿真实验分别验证文中方法在不同噪声条件下辨识低阶时滞系统的性能.实验表明,文中方法具有良好的参数估计精度和较强的抗噪能力,可有效解决噪声条件下低阶时滞系统的阶跃响应辨识问题.To solve the problem of step response identification of low-order system with time delay, a parameter estimation method based on particle swarm optimization is proposed. The method consists of the calculation of initial parameters and the parameter estimation. Firstly, an integral equation approach is utilized to estimate the initial parameters of the system with time delay. By setting an initial parameter estimation error, the parameter range of the time-delay system can be determined. Next, the particle swarm optimization algorithm is employed to reduce the influence of the measurement noise on parameter estimation. Simulation experiments are conducted to verify the performance of the proposed method in identifying the parameters of low-order system with time delay under different noisy conditions. Experimental results demonstrate that the proposed method possesses good parameter estimation precision and strong anti-noise ability and it effectively solves the step response identification problem of low-order system with time delay.

关 键 词：低阶时滞系统 参数估计 阶跃响应 积分方程 粒子群优化 

分 类 号：TP271[自动化与计算机技术—检测技术与自动化装置] TP391[自动化与计算机技术—控制科学与工程]