Model of Cement Rotary Kiln Based on Elman Neural Network and Design of DHP Controller

Calcination process of cement clinker is a complex multi-variable large-disturbances and nonlinear system which is full of mass transfer,heat transfer,physical and chemical reactions.In order to reduce energy consumption and ensure the quality of cement clinker burning,it's necessary to explore new optimal control methods to stabilize the temperature of rotary kiln.Approximate Dynamic Programming(ADP) integrated neural networks,reinforcement learning and dynamic programming techniques,is a new algorithm for optimal control.The dual heuristic dynamic programming(DHP) is an algorithm of ADP,whose output is a partial derivative of cost function with respect to state.It has many advantages such as good dynamics,fast convergence rate,high controlled resolution and so on.Based on the detailed analysis of rotary kiln technology,the model was established by Elman neural network,and the optimization controller was designed with the dual heuristic dynamic programming.The results show that,after the fluctuations in the early control period,the temperature of cement rotary kiln tends to be stabilized and the simulation control of cement rotary kiln is realized.