|
Continuous annealing furnace is an important equipment of metals heat treatment, and its main function is to heat the strip to the setting annealing temperature, insulate and cool the strip to the setting temperature at an appropriate cooling speed to achieve the requirement of strip heat treatment process. In this whole process of annealing, the strip temperature is decreased rapidly in the rapid cooling section; the outlet strip temperature precision of the rapid cooling section has a great influence for the strip performance. So, it has very important practical significance to control the outlet strip temperature in rapid cooling section of continuous annealing furnace.In view of practical use, this paper is taking the rapid cooling section in continuous annealing production line of No.2 cold rolling plan of Benxi Iron & Steel Company as controlled plant; the thesis studies the process of presetting water rollers’positions, which are the vital parameter of the cooling process and the control method of the strip temperature. The main research work and contribution of this dissertation are as follows:(1) According to the analysis and research of the technology and process control of rapid cooling section in the continuous annealing production line of No.2 cold rolling plan of Benxi Iron & Steel Company, we can see that on the one hand, the process of presetting water rollers’positions for the main cooling equipment uses table-investigating method, but this method exists the problem of classification of thickness; moreover, because of other technological conditions of instability, the result of actual control has a significant deviations; On the other hand, the strip temperature control system is a big time-delay and time-varying plant, it often can not obtain a good effect by the conventional PID controller.(2) For the shortcoming of original presetting water rollers’positions, an improved BP neural network model is established to predict the presetting position of the water cooling rollers as a substitute for the traditional table-investigating method. And combined with mathematical models, the result will be used for the calculation of water roller. Through the simulation analysis, the strip temperature control accuracy has a greatly improved.(3) Design for the strip temperature controller. At first, the method of Smith-Predictor is used to for the problem of time-delay, but Smith-Predictor is sensitive for the mathematical model error. If the plant model has a big change, it often has a bad control quality, even leads to unstable. Fuzzy logic has feature of expressing knowledge easily, while artificial immune has good self-learning adaptive ability. In order to enhance the system’s adaptive ability and noise immunity, a kind of new controller-fuzzy immune PID controller has been designed by combing the feature of these two techniques.Compared through simulation, it shows that fuzzy immune PID-Smith controller has good disturbance-rejection, strong robustness and good dynamic and static performance. In addition, it has few fuzzy rules, simple reasoning and other good features; all of this laid the groundwork for the practical application of this arithmetic in industrial control.
|