A CNN-LSTM Model for Tailings Dam Risk Prediction

Abstract

Tailings ponds are places for storing industrial waste. Once the tailings pond collapses, the villages nearby will be destroyed and the harmful chemicals will cause serious environmental pollution. There is an urgent need for a reliable forecasting model, which could investigate the tendency in saturation line and issue early warnings. In order to fill the gap, this work presents a hybrid network - Long-Short-Term Memory (LSTM) and Convolutional Neural Network (CNN), namely CNN-LSTM network for predicting the tailings pond risk. Firstly, the nonliear data processing method was composed to impute the missing value with the numerical inversion (NI) method, which combines correlation analysis, sensitivity analysis, and Random Forest (RF) algorithms. Secondly, a new forecasting model was proposed to monitor the saturation line, which is the lifeline of the tailings pond and can directly reflect the stability of the tailings pond. The CNN was used to identify and learn the spatial structures in the time series, then followed by LSTM cells for detecting the long-term dependence. Finally, different experiments were conducted to evaluate the effectiveness of the model by comparing it with other state-of-the-art algorithms. The results showed that combing CNN with LSTM layers achieves the best score in mean absolute error (MAE), root-mean-square error (RMSE) and coefficient of determination (R2).

Keywords

• Deep learning
• forecasting
• LSTM network
• real-time warning