Combinational method for prediction of coal spontaneous combustion based on Support vector machine

Carbon monoxide, carbon dioxide, hydrocarbon organic gases and other sorts of gaseous product are released in the process of coal spontaneous combustion and all sorts of gaseous product out time and produce a different amount with the different coal temperature. Spontaneous combustion of coal can be forecasted based on corresponding relation between coal temperature and its gaseous products' concentration. Nevertheless, the corresponding relation between gaseous products and temperature is non-linear. This paper is according to the situation of Gases produced by coal spontaneous combustion. Mainly used Support vector machine (SVM) method, and comprehensively used neural network method to build model. Forecast coal combustion degree and take early action to prevent the happening of calamity.