Modeling And Optimization Of Energy Consumption System In Cement Production Based On Data-Driven

With the acceleration of my country’s urbanization process and the continuous expansion of the scale of infrastructure construction,the production and consumption of cement are increasing day by day.Cement industry is my country’s main high power consumption industry,and energy saving and emission reduction have gradually become the focus of this industry.Therefore,cement production enterprises clarify the energy consumption generation links of cement production,filter energy consumption dependent variables;adopt reasonable modeling methods to establish accurate mathematical expressions between dependent variables and electrical energy consumption,and realize effective prediction of electrical energy consumption.The trend of changes in power consumption;and targeted optimization of the production system is of great significance to effectively reducing production energy consumption and reducing production costs.This paper uses a data-driven modeling and optimization method for cement production energy consumption system to establish a firing system model.The main contents are as follows:(1)Cement production process analysis and energy consumption parameter selectionStarting from the production process of the new dry-process cement production,the process of raw meal preparation,clinker firing,cement grinding and bagging in the production line is introduced in detail,combined with the analysis of the source and composition of power consumption,19 cements are identified The influencing factors of power consumption in production are the dependent variables of comprehensive power consumption,which provide input variables for establishing the comprehensive power consumption model of cement production system,that is,data preparation is made for modeling.(2)Establish energy consumption model based on combined neural networkDue to the extremely complex process of the cement industry,the cement kiln calcination system involves many links and equipment.It is a typical complex non-linear system.The parameters involved have the characteristics of many variables,strong correlation,large lag,and large inertia.It is completely clear that the process occurs.The mechanism is not easy,so it is difficult to establish a model based on mechanism analysis.Based on data modeling,this paper combines the advantages of two types of neural networks,feedforward neural networks and feedback neural networks,and establishes a model by combining a combined neural network of BP network and Elman network.The simulation experiment verifies that the method can retain BP neural network.The good static approximation effect of the network also retains the advantages of the good dynamic approximation effect of the Elman neural network,and it has higher prediction accuracy than a single neural network model.(3)Neural network model based on principal component analysis and Markov process optimization combinationIn view of the complex structure of the combined neural network model and the low accuracy of power consumption prediction,this paper adopts principal component analysis(PCA)algorithm to simplify the structure of the combined neural network,and replaces 19 of the original energy consumption model with 9 principal components.Input variables to obtain an energy consumption model with a simpler structure and shorter training time;then a Markov correction process is used to further improve the prediction accuracy of the model.The optimization of the combined neural network model is realized,and the improved combined neural network model is obtained.(4)Optimize energy consumption dependent parameters based on genetic algorithmThe change of each energy-dependent parameter in the production process will affect the electric energy consumption of the production process.How to adjust the energy consumption dependent parameters to the optimal value in order to reduce the electric energy consumption of the cement production system has become the research focus.Since the energy consumption model of the actual cement production system is difficult to meet the condition that the objective function is continuously differentiable or even high-order differentiable,traditional optimization methods usually cannot achieve the optimization of this type of model.Aiming at this problem,this paper uses intelligent optimization algorithms Establish a genetic algorithm optimization model on the basis of the trained improved combined neural network model,avoiding the process of derivation and differentiation of the objective function,and optimizing the energy consumption dependent parameter recommended values,which can make cement production enterprises better Targeted optimization of the production system,effectively reducing the power consumption in the production process.