Application of Back-Propagation Neural Network in the Post-Blast Re-Entry Time Prediction

Predicting the post-blast re-entry time precisely can improve productivity and reduce accidents significantly. The empirical formulas for prediction are practical to implement, but lack accuracy. In this study, a novel method based on back-propagation neural network (BPNN) was proposed tackle drawbacks. A numerical model constructed 300 points of sample data were recorded, with consideration fresh air volume, occupational exposure limit, toxic gas volume per kg explosives roadway length. BPNN six neurons in hidden layer then developed performance discussed terms four indicators, namely, root mean square error (RMSE), coefficient determination (R2), absolute (MAE) sum squares (SSE). Furthermore, one representative formula introduced calibrated comparison. obtained results showed that had more remarkable performance, RMSE 21.45 (R2: 0.99, MAE: 10.78 SSE: 40934), compared formula, 76.89 0.90, 42.06 526147). Hence, is superior predicting time. For better application, it embedded into software.