Mine Gas Emission Prediction based on Gaussian Process Model

For the purpose of making effective analysis and full use of gas measuring data in mines to achieve more accurate gas emission quantity prediction, studied the method for gas emission time series analysis to achieve gas emission quantity prediction based on the Gaussian process regression model. Two methods for gas emission time series analysis were proposed, the relationship between gas emission quantity and time (Q-T) model and autoregression model, considered gas emission time series as a function of time in Q-T model, while constructed the Gaussian process regression model from gas emission time series itself completely in autoregression model. The results of case study show that Gaussian process model can describe the objective laws and the developing trends of mine gas emission, its predictive results are accurate and reliable, therefore, the application of the Gaussian process regression in gas emission time series analysis is a feasible and effective method for gas emission quantity prediction, it has high practical application value.