Today's automatic control technology is based on feedback concept. Features of feedback theory include three parts: measurement, comparison, and execution. Measuring the variables of interest, compared to the expected value, correct the response of the control system with this error. The key to automatic control of this theory and application is how to make the system after making correct measurement and comparison. Robust Controls have begun in the 1950s. In the past 20 years, robust control has always been a research hotspot in international self-control. The so-called "robustness" means that the control system maintains certain properties of the characteristics of certain properties under the parameters of (structures, sizes). According to different definitions of performance, it can be divided into stable robustness and performance robustness. The fixed controller obtained by a closed loop system is called a robust controller. Due to the change of working conditions, external interference and modeling error, the accurate model of the actual industrial process is difficult, and the various faults of the system will also lead to uncertainty of the model, so the uncertainty of the model can be said to the control system. Wide existence in. How to design a fixed controller, enabling the object with uncertain to meet the quality of control, which is robust control, and become a research topic of domestic and foreign researchers. Early study of robust control, mainly for the uncertainty of single variable system (SISO) under small intramography, is representative of Zames proposed differential sensitivity analysis. However, the fault in the actual industrial process causes changes in the parameters in the system, and this change is a boundary instead of infinity. Therefore, there is a modern robust control of the system performance holding and controlling the content under the discussion parameters. Modern robust control is a controller design method that focuses on reliability research of algorithm. Its design goal is to find the required requirements for ensuring the minimum requirements for security requirements control system in the actual environment. Once this controller is designed, its parameters cannot be changed and control performance can be guaranteed. Robust control method is to assume information about process dynamic characteristics and its variation range for time domain or frequency domain. Some algorithms do not require a precise process model, but some offline identification is needed. The design of the general robust control system is based on some worst cases, so the general system does not work in the optimal state. Common design methods include: ina method, while calming, integrity controller design, robust control, robust PID control, robust pole configuration, robust observer, etc. Robust control methods apply to stability and reliability as a primary target, while the dynamic characteristics of the process are known and uncertainties can be estimated. The control of the aircraft and space aircraft is an example of such systems. In process control applications, certain control systems can also be designed with robust control methods, especially those more critical and (1) uncertainty factors; (2) Objects with small margin. However, the design of the robust control system should be done by senior experts. Once the design is successful, there is no need for too many artificial interventions. On the other hand, if you want to upgrade or make major adjustments, the system will be redesigned.
