Research on Rapid Prediction of Disaster Factors in Mine Tunnel Fires Based on WOA-BP

To achieve rapid prediction of fire disaster factors such as temperature, visibility, and CO concentration in mine roadways during a fire, a rapid prediction model for mine roadway fire disaster factors based on the Whale Optimization Algorithm combined with Backpropagation Neural Network (WOA-BP) is proposed using numerical simulation with Fire Dynamics Simulator (FDS). A single rectangular roadway was selected as the research object, and a dataset was established through FDS numerical simulation for training and testing the BP neural network. The WOA was then integrated to optimize the prediction of fire disaster factors in mine roadways. The prediction performance and error characteristics of the WOA-BP model and the BP neural network model were compared and analyzed. The results indicate that the WOA-BP model can instantaneously simulate mine roadway fires under various conditions and rapidly and reliably predict the temperature, visibility, and CO concentration throughout the entire space and time domain of the roadway during a fire event. Compared with the BP neural network model, the WOA-BP model achieved an average reduction of 9.4, 2.1, and 1.12 in Symmetric Mean Absolute Percentage Error (SMAPE), Root Mean Square Error (RMSE), and Mean Absolute Error (MAE), respectively, and an average increase of 13.49% in the coefficient of determination (<tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{R}^{\mathbf{2}}$</tex>).