关键词: 煤矿瓦斯涌出量, 最小二乘支持向量机, 仿真实验, 预测精度

Abstract: In order to improve the prediction accuracy of coal gas emission, a novel prediction model of coal gas emission is proposed based on selective forgotten training samples to solve problem of training sample selection. Firstly, the forgetting factor is introduced to weaken the old training samples and highlight the role of new data simultaneously, and training samples are updated reasonably, and then the least squares support vector machine is used to establish the prediction model of gas emission, and finally, the simulation analysis is carried out to test the performance of model. The results show that the proposed model improves the modeling efficiency of coal gas emission and can obtain good coal gas emission prediction results.

Key words: coal gas emission, least squares support vector machine, simulation experiment, prediction precision