A Gas Emission Prediction Model Based on IQPSO Algorithm Optimized KELM

To improve the prediction accuracy and efficiency of gas emission from the mine retrieval workings, a kernel-based extreme learning machine (KELM) gas emission prediction model based on the improved quantum particle swarm optimization (IQPSO) algorithm is established. Redesigning the particle evolution equation in the QPSO algorithm, and the regularization coefficient and kernel parameters of KELM are optimized using the IQPSO algorithm to enhance the adaptive nature of the prediction model. The factors influencing the amount of gas emission are screened for principal component analysis processing and used as model inputs for gas gushing prediction. The proposed model is compared with QPSO-KELM and PSO-KELM models, and the results show that the error of the proposed model is controlled within 0.15 m
³
/ min, with an average relative error of 0.81%, and the prediction accuracy is 9 and 16 times better than other models. It proves that the IQPSO-KELM model has higher prediction accuracy and efficiency for the prediction of gas emission.