Process control is an important branch of industrial automation. In the past decades, industrial process control has made amazing progress. Process control technology plays a very important role in improving product quality and saving energy, whether in the process of large-scale industrial production with complex structure or in the process of transforming traditional industries. In modern industrial control, process control technology is a branch with a long history. It has been in use since the 1930s. With the development of today's process control technology, the control mode has experienced two development periods from manual control to automatic control. In the automatic control period, the process control system has gone through three development stages, namely, decentralized control stage, centralized control stage and decentralized control stage. Judging from the theoretical and technical means adopted in process control, it can be roughly divided into three stages: the first stage is from the beginning to the 1970s, the second stage is from the 1970s to the early 1990s, and the third stage is from the early 1990s. Among them, the 1970s was the heyday of the development of classical control applications and the initial stage of modern control applications, and the early 1990s was the prosperous period of modern control applications and the initial stage of advanced control development. The first stage is the primary stage, including manual control. Based on the classical control theory, conventional pneumatic, hydraulic and electric instruments are used to control the temperature, flow, pressure and liquid level in the production process. Among many control systems, single-loop structure and PID strategy are the main ones, and some special control systems are established for different objects and requirements, such as proportional control of material proportioning, Smith predictor to overcome the big lag, etc. This is in line with the production level at that time. The second stage is the development stage, with modern control theory as the main basis and microcomputer and high-grade instruments as tools to control more complex industrial processes. At this stage, modeling theory, online identification and real-time control have all broken through the previous forms, and then a large number of advanced control systems and advanced control strategies have emerged, such as adaptive control to overcome uncertainties such as time-varying object characteristics and environmental interference, and predictive control to eliminate the adverse effects caused by model mismatch. The main task of this stage is to overcome interference and model changes, meet complex process requirements and improve control quality. 1975, the world's first distributed control system came out in Honeywell, USA, which turned a new page in process control. Decentralized control system is also called distributed control system, which integrates computer technology, control technology, communication technology and display technology, adopts multi-layer hierarchical structure, and completes the operation, monitoring and control of industrial process according to the principle of overall decentralized and centralized management. Due to the adoption of decentralized structure and redundant technology, the reliability of the system is extremely high. In addition, the open framework of hardware and the modular form of software make configuration and expansion very convenient. There are also many control algorithms (tens to hundreds), good man-machine interface and fault detection and reporting functions. After more than 20 years of development, it has become more and more perfect, showing outstanding style in many control systems. Therefore, it is no exaggeration to say that decentralized control system is a milestone in the development history of process control. The third stage is the advanced stage, which is coming now. At present, process control is entering an advanced stage. No matter from the history and present situation of process control, or from the necessity and possibility of process control development, process control is developing in the direction of integration and intelligence, that is, computer integrated manufacturing system (CIMS): based on intelligent control theory, computer and network are the main means to integrate the operation, planning, scheduling, management and control of enterprises, and realize the automation from raw material storage to product delivery. Its greatest feature is the imitation of human brain function, which is a return to the primary stage of manual control to some extent, but it is more a progress and leap on the basis of manual control, which is a kind of "spiral rise" from the philosophical point of view.