基于阶跃响应和遗传算法优化高阶加时滞模型的辨识方法

Identification Method Based on Step Response and Genetic Algorithm to Optimize Higher-Order Plus Time-Delay Model

针对工业生产中含有时滞环节的高阶过程对象,由于控制器结构复杂,采用直接设计控制器和传统模型降阶的方法实现起来比较困难,加上噪声对控制器产生的干扰,导致这些方法往往得不到满意的结果。文中采用基于阶跃响应的辨识方法,通过分析对象阶跃响应的输入和输出数据,建立二阶加纯滞后模型,利用遗传算法自适应全局搜索能力的优点来优化模型的静态增益、时间常数和纯滞后系数,从而对高阶加时滞对象进行精确的模型辨识。MATLAB仿真结果表明,该方法具有精度高、鲁棒性强和适用性广等优点。使用该方法辨识文中的模型在单位阶跃输入信号下的ITAE指标分别为18.1385、6.2715和167.8892。

For high-order process objects with time-lag in industrial production,due to the complex structure of the controller,it is difficult to implement the method of directly designing the controller and reducing the order of the traditional model.In addition,noise will interfere with the controller.The above factors lead to unsatisfactory results using these methods.To solve these problems,the identification method based on step response is adopted in this study.By analyzing the input and output data of the step response of the object,a second-order plant with dead time model is established.The advantage of genetic algorithm to adapt the global search ability are used to optimize the static gain,time constant and pure hysteresis coefficient,so as to perform accurate model identification of high-order time-delay objects.MATLAB simulation results show that the proposed method has the advantages of high accuracy,strong robustness and wide applicability.Using this method to identify the model in the text under the unit step input signal,the ITAE indicators are 18.1385,6.2715 and 167.8892,respectively.