1. PID control. PID control is one of the earliest developed and widely used control methods. It is a method based on mathematical model of object, especially suitable for deterministic control system that can establish accurate mathematical model. However, for nonlinear and time-varying uncertain systems, it is difficult to achieve the desired control effect with the conventional PID controller. Moreover, in actual production, due to the complicated parameter tuning method, conventional PID parameters tend to be poorly set and poor in performance.
2. Artificial neural network control. Artificial neural network originated in the 1940s, it reflects the basic characteristics of the human brain from some aspects, but it is not the true description of the human brain, but only its abstraction, simplification and simulation, network information processing by neurons To achieve the interaction. The key of neural network control is to select a suitable neural network model and train and learn it until it reaches the requirement, that is to find the optimal neural network structure and weight. However, neural network learning requires a certain amount of experimental samples, which must also be obtained from known and prior experiments. At the same time, the neural network's training and learning processes are sometimes complex and require thousands of runs to get the best possible structure. Sometimes obtained is a local optimal solution, rather than the global optimal solution, due to the limitations of the method, it is also difficult to achieve effective control of the object in question.
3. Fuzzy control. In the actual project, a very skilled operator, with his rich practical experience, can obtain more satisfactory control effects by judging the various phenomena in the field. Controlling complex industrial processes can also be achieved by translating empirically-based measures into appropriate control rules and developing a controller instead of these rules. Practice has proved that fuzzy controller based on fuzzy control theory (FC) can accomplish this task.
Fuzzy control is based on the fuzzy reasoning and imitation of human thinking methods, difficult to establish a mathematical model of the object to implement a control. It uses fuzzy sets in fuzzy mathematics to characterize these vague languages ​​and uses production rules, that is, "if the conditions are true, then execute" statements. The application of fuzzy control technology in China has achieved remarkable results.
4. Expert control. Expert control is an important part of intelligent control. Based on the combination of the theory and technology of expert system with the theories and methods of control theory, the expert control is to imitate the intelligence of experts in the unknown environment and realize the effective control of the system. At the core of expert control is the expert system, which has the ability to deal with various non-structural problems, especially dealing with qualitative, heuristic or uncertain knowledge and information, and achieve various control objectives through various reasoning processes.
5. Humanoid intelligent control. After 20 years of hard work, humanoid intelligent control (HSIC) has formed a basic theoretical system and a systematic design method, and has achieved success in a large number of practical applications. Its main content is to summarize the experience of human control, imitating human control ideas and behaviors, and using generative rules to describe its control and intuitive reasoning behavior. Because the basic characteristic of HSIC is to imitate the control expert's control behavior, its control algorithm is multi-modal control, and it is used interchangeably among many modal control. The algorithm can perfectly coordinate the many conflicting control quality requirements in the control system. For example, robustness and accuracy, speed and stability.