基于偏置继电反馈的闭环辨识方法

多变量过程对象通常具有耦合性等特点,而在对其解耦过程中一般存在一些回路含有不可操作的变量,这就给对象的辨识带来了难度。文中提出了一种基于偏置继电反馈的闭环辨识方法,通过对回路进行偏置继电反馈实验实现对象辨识。文中方法将实验获取对象的输入输出信号进行分析和研究,获取对象在重要频率范围内的频率特性,从而辨识出对象模型。仿真实验表明,文中方法适用于广泛的过程对象,且具有较好的准确性和鲁棒性。

Closed-loop Identification Algorithm Based on Biased Relay Feedback

Multivariable process object has a coupling characteristic frequently.There usually are some loops with inoperable variable,when decoupling those processes.This will make the identification of objects more difficult.A close-loop identification algorithm based on biased relay feedback is proposed in this paper.The identification of objects is realized by biased relay feedback.This method can find a main frequency band of the objects after analyzing and discussing the input and output signals.Thereby it will get an accurate model.Simulation examples show accuracy and applicability of the identification method for a wide range of processes.