Abstract: In order to solve the multifactor mine gas emission prediction problem, a wavelet neural network prediction model is proposed based on quantum particle swarm optimization (QPSO) algorithm. The proposed model utilizes the feature extraction capability of wavelet neural network (WNN), and applies the QPSO algorithm to determining the optimal initial parameters of the WNN. Simulation results show that the optimized WNN prediction model has the advantages of fast convergence, good fitting ability, high prediction accuracy and the ability to give the only prediction result. Additionally, by comparing the simulation data in the experiments, the reason of instability during the network prediction, the contradiction between the network training and prediction accuracy are analyzed. Then the key decision factors and evaluation methods are given for the prediction ability of the proposed model.