Research Of Neural Network PID Control Based On Grey Prediction In The Control Of Tunnel Kiln

High-temperature tunnel kiln is an important part of the production of shaped refractory products. It contains refractory drying, firing, cooling and other processes, and influences the performance, the production of the refractory material, and the economic efficiency of the corporation greatly. And its control level affects the quality of the material directly. In the recent years, our refractory industry develops rapidly, and the structure design of the furnace, the combustion systems, etc, got a greate improvement. However, the control level of the tunnel kiln has developed slowly, most tunnel kilns has still adopted instruments control method, and the degree of automation is very low. Therefore, how to improve the automatization level of tunnel kiln is really a problem to resolve.The background of this paper is a110meters high-temperature tunnel kiln. Firstly, the thesis analyzes the structure, the working principle and the control level of high-temperature tunnel kiln. After that, we designed a computer control system of high-temperature tunnel kiln, and control the high-temperature tunnel kiln by the control arithmetic of neural network PID control based on grey prediction. The control system adopts the structure of3-level control levels, and realized automatic and manual control of the temperature, the pressure and the flow of the tunnel kiln and the function of monitoring of flow chart, real-time andhistorical trend and real-time alarm, and modification of control parameters, etc.In this paper, we combined the BP neural network and PID control, and the arithmetic can overcome some shortcomings of tradition PID control. Grey prediction uses the forecasting value from GM(1,N) based on a few samples to control the system behavior. Due to the forecasting values provide us some important information about future condition of system, so it is possible to control the system before. In addition, the ahead control of the grey prediction can resolve the large lag of the tunnel kiln well.Finally, we made simulation, and obtained effect charts of traditional PID control, neural network PID control, and neural network PID control based on grey prediction separately. Then, we made a comparison and analysis to verify the control scheme feasibility.