Research And Application Of Multi-Objective Optimization Of Process Parameters For Vertical Mill

Vertical mill is the most energy-consuming grinding equipment in the production process of mineral powder.The setting of process parameters not only affects its own energy consumption,but also has a significant impact on the output of mineral powder.In the traditional process of ore powder,the process parameters of vertical mill are usually set according to the processing manual or processing experience.Unreasonable technological parameters leads to the problem of high energy consumption and low ore powder output.How to optimize the process parameters of vertical mill to reduce the energy consumption and increase the output of mineral powder is an urgent problem to be solved at present.This thesis takes vertical mill as the research object.In order to reduce the energy consumption of vertical mill and increase the output of mineral powder,combining machine learning technology with intelligent optimization algorithm,this thesis studies the multi-objective optimization problem of process parameters.Firstly,the process parameters of vertical mill related to energy consumption and mineral powder output of vertical mill are screened.Secondly,the prediction model between the process parameters of vertical mill and the energy consumption of vertical mill and the output of mineral powder based on machine learning is constructed.Then,the problems existing in the intelligent optimization algorithm are improved,and the improved optimization algorithm is used to optimize the process parameters.Optimization results are compared and analyzed.Finally,a multi-objective optimization system of vertical mill process parameters is established to guide the production of mineral powder.The main contents are as follows.(1)In response to the problem that it is difficult to select the process variables related to energy consumption and mineral powder output of vertical mill because of the many variables inside the vertical mill equipment and the characteristics of non-linearity and strong coupling,the method of screening process variables by combining process flow with grey entropy correlation is proposed.According to the process flow of grinding mineral powder by vertical mill equipment,the influence law of each process parameter on vertical mill energy consumption and ore powder output is qualitatively analyzed.In order to further screen the process variables,the relationship between vertical mill process parameters and vertical mill energy consumption and mineral powder yield is quantitatively described by grey entropy correlation analysis.The experimental results show that this method can effectively determine the process variables related to the energy consumption of vertical mill and the output of mineral powder.(2)Aiming at the complex nonlinear function relationship between process parameters of vertical mill,energy consumption of vertical mill and mineral powder yield,improved genetic algorithm optimization back propagation model is proposed to solve the problem of difficult modeling and low prediction accuracy in the traditional mechanism modeling method.The genetic algorithm is improved by introducing a variable iteration strategy based on sine-cosine algorithm and adaptive crossover-mutation probability.The improved genetic algorithm is used to optimize the initial weight threshold of back propagation neural network to improve the prediction accuracy of back propagation neural network model.Experimental results show that compared with the mainstream prediction model,the proposed algorithm has higher prediction accuracy and smaller root mean square error and other indicators.(3)In order to solve the problem that SPEA2 is prone to fall into local optimum and unequal distribution of Pareto front in the optimization process,strength Pareto variable adaptive chaotic differential evolution algorithm is proposed.It introduces variable chaotic mapping strategy and adaptive crossover and mutation operators to SPEA2.Experimental results show that compared with the SPEA2 algorithm,the proposed algorithm has better convergence and distribution when solving ZDT test function,and can find better solutions when solving the optimization problem of vertical mill process parameters,and the Pareto front distribution is more uniform.Compared with the original process parameter design,the energy consumption of vertical mill is reduced by 11.47%,and the ore powder production is increased by 18.36%.(4)In order to apply multi-objective optimization of vertical mill process parameters to industrial field,a multi-objective optimization system for vertical mill process parameters is developed based on the Matlab App Designer platform.The system adopts modular design,which can realize the functions of database connection,data reading,data preprocessing,prediction model training and process parameter optimization,etc.Finally,the optimization system is deployed in the supervision system of a mineral powder line in Tianjin to provide guidance for the mineral powder production of enterprises.