Operation Parameters Optimization Method of Coal Flow Transportation Equipment Based on Convolutional Neural Network

Abstract

Mine coal flow transportation has some typical features of long-distance and complex environments. The transportation equipment usually adopts the mode of constant speed, which makes a large amount of energy waste. To solve these problems, the characteristics of the coal flow transportation system are analyzed. Based on a principal component analysis-convolutional neural network (PCA-CNN), the operation parameters optimization method of coal flow transportation equipment is proposed. Taking the transport time, transport cost, and equipment utilization of belt conveyors and other equipment as the optimization objectives, the multi-objective functions are established, and the operation parameters such as transport speed, transport distance, and equipment start-up time are optimized. The PCA and the CNN are respectively used to determine the weight of each objective function and iteratively train the practical production data samples under multiple constraints. The fully connected layer of CNN is constructed by the Lagrange multiplier method. The optimal production mode and operation parameters of the coal flow transportation equipment are obtained, satisfying the multi-objective functions and constraints. Finally, the practical engineering case is simulated by Plant Simulation, and the operation parameters of the coal flow transportation equipment are compared before and after optimization. The research results show that the objective function of each experiment is optimized to some degree. Furthermore, comprising other common algorithms, the advantages and effectiveness of the based-CNN operation parameters optimization method are verified. These have an important guiding significance for energy-saving and efficient coal flow transportation equipment operation.