带干扰线性系统的动态补偿  

Dynamics Compensation for Linear System with Disturbance

作　　者：冯红银萍 FENG Hongyinping(School of Mathematical Sciences,Shanxi University,Taiyuan 030006,China)

机构地区：[1]山西大学数学科学学院,山西太原030006

出　　处：《山西大学学报（自然科学版）》2021年第5期851-875,共25页Journal of Shanxi University(Natural Science Edition)

基　　金：国家自然科学基金(61873153)。

摘　　要：系统观测和控制的过程本质上是信息提取和信息利用的过程。好的控制策略应建立在信息充分、合理利用的基础之上。这些信息既包含通过系统观测得到的在线信息,也包含系统的先验信息。基于上述认识,文章提出线性系统的动态补偿方法,它可以充分利用系统的动态信息。现代控制理论中的大多数控制方法都是基于模型的,控制器的设计建立在数学模型基础之上。然而在很多情况下,数学模型往往很难合理地描述真实系统。这使得基于模型的控制策略在实际应用中可能难以达到预期的效果。对系统模型鲁棒性的引入虽然可以在一定程度上缓解数学模型和实际系统之间差异造成的困难,但仍然无法从本质上解决问题。近年来,数据驱动控制的引入虽然摆脱了控制策略对模型的依赖,但是却存在系统先验信息利用不充分的问题。与这两种情况不同,线性系统动态补偿方法是一种信息驱动的控制设计策略,它从信息利用的角度设计控制器,既可以充分利用系统和干扰的先验动态信息,又摆脱了对控制模型的依赖,是一种新的控制设计范式。System observation and control are actually the process of information extraction and utilization.A good control strategy should be based on that the information can be used sufficiently and reasonably.The information consists of not only the online information that is obtained by the system observation,but also the prior information about the system and the disturbance.We propose the dynamics compensation method for the linear system with disturbance based on the aforementioned understanding,and we can use the dynamical information about the system and the disturbance sufficiently.Most of the existing controls in the modern control theory are based on the system model,and the controller design is based on the system model as well.However,the mathematical model may not describe the real world system appropriately sometimes.This fact implies that the model based controller may not reach the expected control effect.The robustness to the system model can reduce,to a certain extent,the difficulties that caused by the error between the plant and its mathematical model.However,the problem is still not solved completely.Recently,although the data driven control gets rid of the dependence of the control strategy on the model,it gives rise to the problem of the insufficient utilization of the system prior information.In contrast with the robustness to the system and the data based control,the dynamics compensation approach to the linear systems is an information driven approach.The controller is designed from the information utilization point of view.Both the system prior dynamics and the disturbance dynamics can be used sufficiently,moreover,the controller is almost model free.It is a new paradigm of controller design.

关 键 词：自抗扰控制 干扰系统 内模原理 高增益 动态信息 

分 类 号：O175.2[理学—数学]