Research On Intelligent Coal Blending Model Of Fengfeng Group

With the promotion of China's"green energy"and the modernization development of industries such as coking,steel,electric power industries and etc.,the requirement for coal qualities and the optimization and utilization of coal resources must be taken seriously.The research is against the shortage of coal blending technology research,this paper studies the coal blending technology applicable to multi-coal types in multi coal preparation plants.Then,establish coal blending optimization model for industry indicators and G of coal via improvement particle swarm optimization algorithm,which can optimize product structure and avoid waste of resources.It has important economic value for coal enterprises.In this paper,a series of optimized coal blending experiments were designed and carried out according to the actual situation.As to the results of coal blending experiments,the relationship between coal qualities and coal blending characteristics is analyzed.It is shown that the A₍(6?9?,V₍?9?61??and S₍,?9? of the mixing coal basically meet the linear requirement while the M₍(6?9? is not satisfied due to the environmental impact.Then,based on linear regression,prediction models were established combining with the theory of coal blending.For the nonlinear characteristics of G,Gaussian transformation of the V₁)and Vis carried out to establish a nonlinear prediction model.Because V₁)and Vhave a great influence on the difference between the measurement G and prediction G.The model parameters are solved by sequential quadratic programming.It is shown that the model can improve the prediction effect to a greater extent.The aim is to minimize the price of mixing coal,minimize the use of high-quality coal and maximize the use of low-quality coal.The mathematical model constraints are determined according to the requirement of mixing coal and high-quality coal stocks.So that the coal quality characteristics are predicted by linear and nonlinear models.Then,based on genetic algorithm,particle swarm optimization algorithm and adaptive particle swarm optimization algorithm,an optimization model of coal blending structure is established.Comparing the prediction results,it is shown that the PSO algorithm is easier to obtain lower cost price than the GA;the adaptive PSO algorithm has more optimization curves than the standard one,which is more stable to solve the coal blending issue.