Abstract
A gas outburst prediction model based on the improved beluga whale optimization algorithm (IBWO) optimized temporal convolutional network (TCN) is proposed for the problem of low accuracy of various gas outburst prediction models in underground mines. Firstly, the gas outburst data are preprocessed by using the kernel principal component analysis (KPCA) method. The beluga whale optimization algorithm (BWO) is improved by introducing Logistic chaos mapping, dynamic adaptive factor and Cauchy Gauss mutation. After that, the relevant hyperparameters of the temporal convolutional network (TCN) are optimized by using the improved beluga whale optimization algorithm and the IBWO-TCN gas outburst prediction model is established. Finally, real data from a coal mine in Shanxi were selected for experimental comparison and analysis, and the results showed that the prediction accuracy of each gas outburst hazard of the IBWO-TCN model was 100%, 95%, 94.12%, 90%, and 93.34%, respectively, and the comprehensive gas outburst prediction accuracy was 95%, with better comprehensive performance than other models, which verified that the method of this paper can effectively improve the accuracy of gas outburst prediction in underground mines.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call