Abstract
This paper focuses on the effective analysis of the mine gas emission monitoring data, so as to realize the accurate and reliable mine gas emission prediction. Firstly, a weighted multiple computing models based on parametric t–norm is constructed. And a new mine gas emission combination forecasting method is proposed. The BP neural network model and the support vector machine were used as the single prediction models. Finally, genetic algorithm and least square method were used to determine the parameters of t-norm in the combination forecasting model, and realized the optimal combination of single models. The experimental analysis shows that the new model has less error than BP neural network model and support vector machine model in the evaluation indexes. It can be concluded that the new combined forecasting model is more suitable for the coal mine gas emission forecast.