Prediction of peak particle vibration velocity based on intelligent optimization algorithm combined with XGBoost

Meishan Iron Mine in Nanjing is one of the mines closest to high-rise buildings in China. However, the ground vibration caused by the mine blasting operation will seriously impact the stability of surrounding buildings and the daily life of nearby residents. Therefore, accurate prediction of peak particle velocity (PPV) caused by blasting is a key issue in mining engineering. In this study, an improved intelligent optimization algorithm is proposed to predict the ground vibration caused by the mine blasting. To construct the prediction model, this study collected 358 data points and selected 6 key parameters as input variables, and PPV as output variables. At the same time, to improve the prediction accuracy of the model, this study uses Tent chaotic mapping and Weibull random disturbance strategy to optimize the Osprey optimization algorithm and compares the performance of various optimization algorithms in adjusting model parameters. The determination coefficient, root mean square error, mean absolute error, and variance were used to evaluate the performance of the model. The results show that the proposed optimization model is significantly better than the traditional machine learning model in all indicators, showing higher accuracy and reliability. Furthermore, based on Shapley Additive Explanations, this study quantitatively assesses the contribution and importance of each input parameter to the PPV prediction, revealing the intrinsic relationships between the input variables and model performance. This research not only provides an efficient tool for PPV prediction but also offers insights and references for ground vibration studies in other mining blasting operations.