Predictive Method for Spontaneous Combustion of Coal Stockpile using in Deep Neural Network

摘要

The present invention relates to a method for predicting spontaneous ignition of coal stockpiles using a deep neural network, which includes the following steps of: inputting temperature and environmental information related to spontaneous ignition of coal stockpiles into an input layer node of the deep neural network; and predicting the spontaneous ignition of coal stockpiles using a Sigmoid activation function at nodes of hidden layers 1 and 2 of the deep neural network to output a predicted result to an output layer node.