Application of intelligent control based on process identification in beryllium-bronze heat treatment

Beryllium-bronze is one kind of reed material. In order to achieve a good mechanical performance, the process control of its heat treatment is very strict. With the traditional PID control strategy it is difficult to meet the practical control demands. Firstly, based on the relationship between a linear system’s output and its input, the Least Squares method is applied to work with the Correlation Theory to identify such parameters of the process model as constant delay etc. Using this method the constant delay obtained is more precise than by means of the delay response length of step input while it holds a strong capability of resisting interference. Secondly, an optimized tuning of PID parameters is carried out using optimized method of many variables with no restriction based on the target of integral of time multiplied absolute error (ITAE). Thirdly, in the light of the analysis of the variable trend of routine PID control parameters in different control stages and thanks to the expertise in PID control field, the algorithm in the controller is designed to correct the tuned PID control parameters in order to improve PID effectiveness. The simulation result shows that this method can be achieved easily and the control effect is better than by the traditional PID control method. By using this method in temperature control system of beryllium-bronze heat treatment, the problems of manual tuning which is difficult and time consuming are solved, as a result the unification of celerity and precision of control comes true.