Research on application of deep learning technique with long short term memory model in the monitoring and forecasting system of coal spontaneous combustion temperature in underground mines

Abstract

Online monitoring and forecasting of coal seam temperature in underground mines with spontaneous combustion is an urgent issue that is currently receiving attention. In this paper, a method of building a model to forecast the coal spontaneously combustion temperature every hour for the next eight hours using a univariate Long Short Term Memory (LSTM) model is proposed. The parameters of the model are adjusted through tests suitable for the given problem. The monitoring system combined with this forecasting method contributes to improving production efficiency, labor safety, environmental protection and effective use of Vietnam coal resources.