A novel approach to forecast dust concentration in open pit mines by integrating meteorological parameters and production intensity.

Abstract

Mine dust pollution poses a hindrance to achieving green and climate-smart mining. This paper uses weather forecast data and mine production intensity data as model inputs to develop a novel model for forecasting daily dust concentration values in open pit mines by employing and integrating multiple machine learning techniques. The results show that the forecast model exhibits high accuracy, with a Pearson correlation coefficient exceeding 0.87. The PM2.5 forecast model performs best, followed by the total suspended particle and PM10 models. The inclusion of production intensity significantly enhances model performance. Total column water vapor exerts the most significant impact on the model's predictive performance, while the impacts of rock production and coal production are also notable. The proposed daily forecast model leverages production intensity data to predict future dust concentrations accurately. This tool offers valuable insights for optimizing mine design parameters, enabling informed decisions based on real-time forecasts. It effectively prevents severe pollution in the mining area while maximizing the use of natural meteorological conditions for effective dust removal and diffusion.