| For the advantages of simple, robust and high stability, PID control algorithm are widely used in industrial process control. With the incessantly development of industrial production, some equipments which operating condition changes frequently and complicatedly and need to optimize the PID parameter to fulfill the requirement of production quality and quantity. Therefore, the paper arms at the status of industrial process control, research and exploit PID Auto-Tuning controller to resolve the tuning problem of PID parameter in complicated technics.Firstly, this paper expounds the controller blue print of design. Software design adopt the method of block design, which includes the design of initialize module,sampling module,model identify module parameter tuning module and so on. These modules can achieve the arithmetic. On the controller hardware design, ARM7 chip-LPC2132 made by PHILIPS company is used as CPU of the controller, CA320240C chip is used as LCD module controller, which accomplish analog sampling, auto-tuning PID computing, analog signal and digital signal control output, liquid crystal display, keypad settings, the BUS communications, and others functions.Through the domestic and abroad literatures, this paper analyzes and compares the actual state of PID control application and optimized technique at home and abroad, considering the embedded system technology which requires simple arithmetic and better logicality, the paper adopts a PID controller auto-tuning algorithm based on correlation coefficient identification approach. This arithmetic are composed of two parts: correlationship-coefficent identification and PID optimize tuning. The first part includes data acquisition, data processing, model identification and model verification. And PID optimize tuning is a designing process of eximious controller parameter based on quadratic objective function. This paper adopt correlationship-coefficent identification to model the plant, and then, tune the PID parameter by Hooke-Jeeves method in accordance with a certain object function. This controller can build control process models while doesn't influence product process, then auto-tune the PID parameter.Based upon the arithmetic results are simulated by Matlab, this paper debugs and validates arithmetic on the setting of liquid station for the Process Automation Lab in Dalian University of Technology, through a mass of tests and data analysis, testifies this arithmetic could make the process returns to balance fleetly when Lab module be disturbed. |