Hybrid method for prediction of coal and gas outburst based on data fusion and soft sensor

Based on introduction of the background and the limitations of present methods for coal and gas outburst, a hybrid method for prediction of coal and gas outburst based on soft sensor and data fusion combining many associated dynamic and static influence factors is proposed. In the method, the data fusion method based on arithmetic mean and batch estimation is used to process the dynamic influence factors data obtained by multiple sensors, and the soft sensor model based on fuzzy BP ANN predicts the dangerous status of coal and gas outburst according to the static factors data and the processed dynamic factors data. The application results show that the proposed method has high accuracy, and it is a practical method to dynamically and accurately predict coal and gas outburst.