The Neural Network PID Controller for Cement Rotary Kiln Temperature Based on FPGA

Abstract

In this paper, we discuss a method of using BP neural network to adjust PID parameters online for controlling the cement kiln temperature. First, according to the principle of the rotary kiln temperature, the BP_PID controller is designed in theory, and the neural network trainings and simulations in MATLAB are taken. Secondly, by using the top-down method in VHDL language, the modules of BP_PID controller which is the BP neural network forward transmission module, the error back propagation module, and the weigh and threshold adjustment module and so on are all implemented on FPGA chip. Eventually the simulations of program in Quartus are given in this paper. The simulation results show that this design is reasonable, the hardware implementation is correct, so it can create the conditions for a wide range of applications of hardware intelligent algorithm in the field of industrial.