Predictive Control Research On Temperature Of Cement Calciner Based On Temporal DBN-ARX And Parameter Optimization

The temperature of the decomposition furnace is an important variable affecting the quality of cement.Proper and stable of the decomposition furnace' temperature is one of the keys to the operation of cement pre-decomposition system.In this paper,a generalized predictive control method based on time series DBN-ARX model and a parameter optimization algorithm are proposed.The specific research contents are as follows:Firstly,the specific process of the new dry process cement pre-decomposition process is studied,and the difficulties and main characteristics in the temperature control process of the decomposition furnace are analyzed and summarized.The main factors affecting the decomposition rate and the stable operation of the decomposition furnace are discussed,and then several main characteristic variables affecting the temperature of the decomposition furnace are obtained.Combined with the time-delay characteristics of the system,the timelag relationship between multiple characteristic variables and the temperature of the decomposition furnace is analyzed,which lays a theoretical foundation for the proposed temperature control model of the cement decomposition furnace.Secondly,according to the system characteristics of the temperature control process of the decomposition furnace,the cross-correlation function analysis theory is used to determine the maximum correlation time period information of the variable,and the input belief layer structure of the deep belief network is reconstructed,and the time series DBN with the timing characteristics of multiple related variables is established as input.Decomposition furnace steady state model.Through the dynamic gain,the time series DBN steady state model is combined with the ARX dynamic model to construct the time series DBN-ARX combination model of the cement decomposition furnace temperature control system,which lays a model foundation for the proposed temperature control method of the cement decomposition furnace.Then,based on the time series DBN-ARX combination model,the generalized predictive control model output formula of the decomposition furnace temperature control system is derived,and then the furnace temperature prediction control algorithm based on the time series DBN-ARX combination model is designed by constructing error feedback correction and online rolling optimization.At the same time,aiming at the problem that the parameters in the decomposition temperature control process are difficult to be set,a control parameter optimization method based on GGA is proposed,and a temperature prediction control system for the decomposition furnace with external parameter optimization structure is constructed.Finally,based on the decomposition temperature control algorithm proposed in this paper,the field data is used for simulation experiments.The different models,optimization methods and control methods are compared,and the simulation and stability of the system are carried out to verify the effectiveness and feasibility of the proposed method.The experimental results show that the temperature control algorithm of the cement decomposition furnace proposed in this paper is more robust and more stable,and can achieve stable control of the temperature of the decomposition furnace.