Prediction of Explosion Power of Fuel Air Mixture in Coal Mine by Improved BP Neural Network

Abstract

In order to improve the feasibility and accuracy of predicting the explosion power of fuel air mixture, a method of BP neural network prediction is proposed by combining factor analysis method with BP neural network. Using factor analysis method, the original data of 9 fuel air mixture explosion power factors were processed by dimensionality reduction data, 2 common factors were obtained, and 2 common factors were substituted for 9 fuel air mixtures as input layer parameters of BP neural network. A prediction model of coal and gas outburst with the combination of factor analysis method and BP Neural network is established to predict the explosion power of fuel air mixture. The prediction model of the explosion power of the fuel air mixture is selected to verify the improved BP neural network prediction models, and the final results are as follows: The relative error range of four prediction samples is 0.16%-7.58%, all less than 10%. The improved BP neural network prediction method is used to solve the problem of the traditional BP neural network because of the excessive number of input layer parameters, low data processing efficiency, slow iteration rate and low precision, which provides a new research way for predicting the explosion power of fuel air mixture.