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. PID (Proportion - Points - Differential) Controller has been more than 50 years of history as the earliest and practical controller, and is still the most widely used industrial controller. The PID controller is simple and easy to understand. Prerequisites such as system models such as system models are used, thus becoming the most widely used controller. The PID controller consists of a proportional unit (P), the integration unit (I), and the differential unit (D). Its relationship between E (T) and output U (T) is therefore its transfer function is: it is widely used due to wide, flexible, has serialized products, and only three parameters (KP, KI and KD). In many cases, all three units do not necessarily need one to two units, but the proportional control unit is essential. First, the PID application ranges. Although many industrial processes are non-linear or time-changing, the PID can be controlled by a system that can become substantially linear and dynamic characteristics without linear or time-change. Second, the PID parameters are more tuning. That is, the PID parameters KP, KI and KD can be tuned in a timely manner according to the dynamic characteristics of the process. If the dynamic characteristics of the process changes, for example, the system dynamic characteristics can be changed by the change in the load, and the PID parameters can be re-tuned. Third, the PID controller is also constantly improved in practice, and the following two improvements are improved. In the factory, you can always see that many loops are manual, because it is difficult to let the process work smoothly in the "automatic" mode. Due to these shortcomings, industrial control systems with PID are always troublesome by problems such as product quality, safety, yield and energy waste. The PID parameter self-tuning is to deal with the PID parameter to set this problem. The PID controller that is now automatically tested or tested is already a standard for commercial single loop controllers and decentralized control systems. In some cases, the PID controller designed for a particular system is well controlled, but there are still some problems need to be resolved: If the self-order should be based on the model, the re-setting online search and maintain the process model for the re-setting of the PID parameters. Difficult. When the closed loop is working, a test signal is required during the process. This method will cause disturbance, so the model-based PID parameter is not too good to tighten industrial applications. If self-tuning is based on the control law, it is often difficult to distinguish the impact of the effects caused by load interference and process dynamic characteristics. Therefore, the influence controller that is subject to interference will produce an ultra-tone, producing an unnecessary adaptive conversion. . In addition, due to the non-mature stability analysis method based on the control law system, there are many problems with the presence or absence of the parameters. Therefore, many of the PID controllers of many self-tuning parameters often work in automatic tuning mode rather than continuous self-tuning mode. Automatic settings typically automatically calculate the PID parameters according to the simple process model determined by the open-loop state. However, it is still undeniable that PID also has its inherent disadvantages: PID is not too good when controlling nonlinear, time-change, coupling, and parameters and structural uncertain processes. Most importantly, if the PID controller does not control the complex process, it is useless to adjust the parameters. Although there are these disadvantages, the PID controller is the easiest but the best controller.
