Abstract: As influencing factors of gas extraction rate at coal face are complicated, nonlinear relation which are dynamic and vague exist between each influencing factors, prediction model is hard to establish on traditional prediction methods. Neural network has the characteristics of self-organization, self-adaptive, parallel processing and strong nonlinear approximation ability. The prediction model was established based on functional relationship between gas extraction rate at coal face and its influencing factors. The result indicates that the precision of the prediction model could meet the request, and the model is reasonable and feasible.

Key words: Gas extraction rate, Neural network, Prediction