52 Prediction of Coal Mine Gas Concentration Based on Constructive Neural Network

This paper analyzes gas concentration data from different angles (levels) through a variety of size for the characteristic of gas concentration time series data set based on quotient space theory. Meanwhile, we deal with the information by quotient space and neural network model, and propose a constructive neural network model, which combines the covering algorithm and BP neural network model. Experimental results show that the model is better than the traditional method by the coal gas data obtained from some coal gas.