Optimization Of Parameters And Energy And Material Consumption Of Hematite Grinding Process Based On Swarm Intelligence Algorithm

The hematite grinding process is an essential part of iron smelting.Due to the complex composition of hematite,the variable ore source,and the complex dynamic characteristics of the grinding process,it is difficult to predict the grinding particle size of the key index of the process.The grinding particle size often fails to meet the technical requirements due to the reliance on expert experience in setting production parameters.The unqualified grinding particle size will cause rework and increase the energy and material consumption of the process,which will have a significant impact on the subsequent production process.Therefore,it is of great practical significance to study the comprehensive optimization methods of grinding particle size and energy consumption of the grinding process.Given the importance of grinding particle size,with the goal of maintaining stable production of the grinding particle size and reducing energy and material consumption during the grinding process,comprehensive optimization of the grinding process is studied.The main work and innovation of the paper contain the following two aspects:(1)Firstly,a prediction(optimization)model of the grinding particle size is considered.Due to the complex grinding mechanism and the excellent nonlinear mapping ability of data-driven methods such as neural networks,a backpropagation neural network model optimized by an enhanced Harris hawk algorithm is proposed for the grinding particle size,which is used to describe the relationship between the grinding particle size and various control parameters.To improve the accuracy of the model,an enhanced Harris hawk algorithm is developed to optimize the parameters of the model.In the initialization phase,the algorithm introduced the concept of “information sharing” to enhance communication and cooperation among individuals.In the exploration phase,a decay factor is introduced and some of the scaling factors are adjusted.Finally,the average optimal position is used in the exploitation phase.The experimental results show that the enhanced optimization algorithm has higher convergence accuracy and stronger global search capability than the original algorithm and other optimization algorithms,and the optimized model can accurately predict the grinding particle size.(2)On the basis of the prediction model of the grinding particle size,a comprehensive optimization method of the grinding particle size and process energy and material consumption is studied.The mean square difference between the actual production grinding particle size and the expected value,and the total water and electricity consumption of the process are used to construct the comprehensive optimization objective function.The established prediction model of the grinding particle size is used as an equation constraint.The mill water feed,the grinding concentration,the mill power and other parameters are considered as optimization variables.Since comprehensive optimization has a nonlinear prediction model as an equation constraint,an enhanced tuna swarm optimization algorithm is developed to solve the problem in order to improve the comprehensive optimization effect.The tuna swarm optimization algorithm introduces an end-elimination mechanism in the initialization phase to improve the overall hunting ability and quality of the population.In the position update phase,a correlation factor is used to make the foraging strategy more consistent with the habits of the tuna population.In the spiral foraging strategy,an average geometric position method is introduced to improve the stability of the algorithm.The effectiveness of the enhanced tuna swarm optimization algorithm is verified by experiments with benchmark test functions.Simulation experiments of the comprehensive optimization for a grinding process are conducted by data of the grinding process.The results show that the enhanced tuna swarm algorithm has more advantages than the manual setting and several other algorithms,and achieves the reduction of process energy and material consumption while ensuring the stability of the grinding particle size.In summary,the two enhanced optimization algorithms proposed in the paper both have better search accuracy and global search capability than the original algorithm,and can be effectively applied to the comprehensive optimization of the grinding process to achieve a more stable,efficient and energy-saving hematite grinding production.