Coal seam spontaneous combustion prediction based on General Regression Neural Network

摘要

Coal seam spontaneous combustion is one of the colliery major disasters.Accurate prediction of the tendency of spontaneous combustion in exploiting coal seam is of great importance.There exists a kind of complicated nonlinear relation between the tendency of spontaneous combustion and its influence factors,General Regression Neural Network has mighty capability of nonlinear approach and could truly show the nonlinear relation between input variables and output variables.In this paper,after roundly analyzing affecting factor of coal seam spontaneous combustion and abroad referring other prediction models,the coal seam spontaneous combustion prediction model of General Regression Neural Network is established.The prediction model and artificial neural network tool chest of MATLAB are applied to compile program and train and forecast some coal seam samples of certain coal mine.The maximum error is just 0.0764 in six prediction tests.Researches prove that this prediction model is precise enough,so it could be used to forecast coal seam spontaneous combustion.