Dynamic Prediction of Coal and Gas Outburst Based on BPNN and Case-based Reasoning

In order to realize the accurate, quick, dynamic and intelligent prediction of coal and gas outburst, a prediction method based on BP artificial neural network (BPNN) and case-based reasoning (CBR) is proposed. Aiming at the strong nonlinearity for influencing factors of coal and gas outburst, a method based on BPNN is given to determine case feature weights objectively and scientifically. The case reuse fast method based on weighted average is used to improve the prediction quickness and precision. The proposed method is validated using practical measured data. The simulation example shows that the proposed method has the higher prediction accuracy and stability, the prediction speed of the proposed method is quicker than those of the existing CBR prediction methods, and the proposed method is valid.