Strip Cooling Control And Parameter Optimization Of Continuous Annealing Furnace

Serving as an vital factor of evaluating the quality and flatness of the strip, the outlet strip temperature precision of the rapid cooling section in Continuous Annealing Line(CAL) has direct affluence on strip's structure and mechanics capability. Therefore, seeking for excellent control method and optimization method of the cooling process has become an important part for the CAL. Among each section of the continuous annealing furnace, rapid cooling section dominates the result of the cooling. Besides, the techniques and control strategy of rapid cooling section are also most complicated. Thus, the paper does much research job on seeking for more accurate ways to increase the precision of the outlet temperature with a focus on control methods and process parameter's optimization.In this paper, quite a few jobs are done on the process control system of the rapid cooling section, including the mathematical model of the cooling process, the control method of the strip temperature, the preset calculation process,especially the process of presetting water rollers'positions, which is the vital parameter of the cooling process. The shortcomings of the cooling process control are analyzed, and the ways to improve the control system and the cooling process are proposed based upon artificial intelligent methods.A fuzzy-PID control strategy based upon Smith predictor is proposed in this paper to improve the problems brought forward by the delay. And a fuzzy-PID control system based upon Smith predictor is designed and realized in SIMULINK environment, the control result and the anti-disturbing ability of the system is analyzed, besides, the control results are compared between the PID-Smith method and the method designed in this paper regarding the control object parameters'variety. The simulation results show that, the control accuracy of the feedback system is effectively increased.A 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, taking sufficient technique conditions and their varieties into consideration, which has a excellent ability of processing complex non-linear issues.The model is realized in MATLAB environment, which shows a better result than the traditional method.