Thermal Efficiency Optimization Control System Of Cement Raw Material Predecomposition Process

As the initial stage of clinker calcination,the stability of the process affects the yield and quality of cement clinker and the stability of the whole clinker calcination process.Therefore,the optimal control of thermal efficiency of cement raw meal pre-calcination process can not only improve the stability of cement clinker calcination process,but also reduce the coal consumption.However,there are many unmeasurable parameters in the raw meal pre-decomposition process,and it is difficult to realize the real-time monitoring of thermal efficiency,coupled with the characteristics of large inertia,large hysteresis and strong coupling in the raw meal pre-decomposition process,which makes it difficult to realize the long-term stable operation of the existing traditional control scheme,and the decomposer outlet temperature setting is completely dependent on the operator’s experience,which is subjective and difficult to ensure the optimal thermal efficiency of the raw meal pre-decomposition process in real time.In order to solve the above problems,this paper conducts a study on the optimal control of thermal efficiency of cement raw meal pre-decomposition process based on a large amount of production data from a cement company in Shandong.The main research contents and innovative results of this paper are as follows:(1)Aiming at the current situation of high energy consumption and low thermal efficiency of cement raw material pre-decomposition process in the new dry process cement production process,the research and application developments of numerical simulation,optimal setting of decomposer outlet temperature,modeling and control of decomposer outlet temperature in cement raw material pre-decomposition process by domestic and foreign scholars are reviewed,and the application schemes of proportional-integral-differential control,fuzzy control,neural network control,model predictive control and their combined control algorithms in improving the thermal efficiency of cement raw material pre-decomposition process are summarized,and the feasibility of each scheme is analyzed.(2)Based on the in-depth analysis of the production process of raw material pre-decomposition process and the structure mechanism of decomposer,the key factors affecting the thermal efficiency of pre-decomposition process are found by combining the energy flow theory,and the establishment of the thermal efficiency model of raw material pre-decomposition process is completed.In the thermal efficiency calculation,this paper adopts soft measurement to solve the problem that some variables are not measurable,and lays the foundation for realizing the real-time monitoring of thermal efficiency of raw meal pre-decomposition process.(3)In order to realize the accurate prediction of decomposer outlet temperature,this paper establishes the prediction model of decomposer outlet temperature by BP neural network.Regarding the selection of input variables,the variables affecting the decomposer exit temperature were initially determined based on the in-depth understanding of the raw material pre-decomposition process,combined with the operator’s experience,and verified by using gray correlation analysis to ensure the reasonableness of the variable selection.Regarding the structure selection of BP neural network,the structure of the neural network was finally determined as a double hidden layer based on the empirical formula combined with a large number of experiments,and the accuracy of the prediction model was verified by simulation.(4)In order to realize the automatic setting and stable operation of decomposition furnace outlet temperature,the two-layer structure model predictive control is proposed.In the real-time optimization layer,the automatic optimization setting of decomposition furnace outlet temperature is realized by establishing the steady-state target equation of cement decomposition furnace outlet temperature combined with particle swarm algorithm,and the set value is tracked by using model predictive control algorithm in the dynamic control layer.The experimental results show that the dynamic performance,steady-state performance and anti-disturbance performance of the two-layer structure model predictive control algorithm are better than the PID control,Fuzzy control,Fuzzy-PID control and RBF-PID control strategies which are widely used at this stage.(5)Based on the above research,the optimized control system for thermal efficiency of cement raw meal pre-decomposition process was designed and completed with the actual engineering requirements,and the application was verified and compared using field data.The results show that the thermal efficiency calculation scheme proposed in this paper can not only realize real-time online monitoring of thermal efficiency of raw meal pre-decomposition process,but also the system stability,accuracy and thermal efficiency are improved after adopting the predictive control of double-layer structure model.