精确模型辨识的光电平台自抗扰控制器  

Active disturbance rejection controller of opto-electronic platform based on precision model identification

作　　者：殷宗迪 董浩[1] 史文杰 张锋[1] 

机构地区：[1]天津津航技术物理研究所,天津300308

出　　处：《红外与激光工程》2017年第9期308-313,共6页Infrared and Laser Engineering

基　　金：国家高分专项某项目

摘　　要：针对实际工程中的复杂光电平台对控制精度要求越来越高的需求,提出一种依据系统精确模型辨识方法的自抗扰控制器设计方法。考虑机械谐振和摩擦因素,建立光电载荷控制系统精确数学模型,并根据系统输入输出特性辨识系统的数学模型参数,在所辨识模型的基础上设计自抗扰控制器。以某型光电跟踪平台为例,设计了4阶跟踪微分器,5阶扩张状态观测器和非线性状态偏差反馈控制律组合的自抗扰控制器。在Matlab/simulink中建立系统仿真平台,对PID控制器和自抗扰控制器进行仿真对比,结果表明,采用自抗扰控制器的系统超调由1.8%减小到0.9%,系统最大跟踪误差由0.03(°)/s减小到0.013(°)/s,超调更小,响应时间更快,抗扰动能力更强。PID controller is widely used in most projects, but its weakness is more and more fatal with the development of the stabilization precision. An active disturbance rejection controller （ADRC） based on system identification was proposed to control complex opto-electronic platform which can be described by exact model. Firstly, the system model joined friction and structure resonance models were established in order to simulate full practical states in Matlab. Secondly, the rank of system was confirmed by AIC and the model of system was confirmed by linear least square method, and then the model was transformed into the first visible standard type. Finally, an active disturbance rejection controller was designed. The weakness of PID was conquered by 4th Tracking-Differentiator （TD）, 5th Extended State Observer （ESO） and Nonlinear States Error Feedback （NLSEF）. Take some complex opto-electronic platform as an example, a series of simulated contrast tests between ADRC and PID were carried out. The results showed that the overshoot was reduced from 1.8% to 0.9%,and the maximum tracking error was reduced from 0.03（°）/s to 0.013（°）/s. ADRC had a faster response with a smaller overshoot than PID. This indicates that ADRC achieve control more accurately than PID.

关 键 词：光电平台 自抗扰控制 系统辨识 非线性反馈 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]