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 forecast model, which could investigate the variation trend of stability coefficient of tailing dam and issue early warnings. In order to fill the gap, this work presents an hybrid network -Wavelet-based Long-Short-Term Memory (LSTM) and Convolutional Neural Network (CNN), namely Wavelet-CNN-LSTM netwrok for predicting the tailings pond risk. Firstly, we construct the especial nonlinear data processing method 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. After using the discrete wavelet transform (DWT) to decompose the original saturation line data into 4-layer wavelets and de-noise the data, the CNN was used to identify and learn the spatial structures in the time series, followed by LSTM cells for detecting the long-short-term dependence. Finally, different experiments were conducted to evaluate the effectiveness of our model by comparing it with other state-of-the-art algorithms. The results show that Wavelet-CNN-LSTM achieves the best score both in mean absolute percentage error (MAPE), root-mean-square error (RMSE) and R 2 .

A CNN-LSTM network with multi-level feature extraction-based approach for automated detection of coronavirus from CT scan and X-ray images

“…However, in the proposed CNN network, the sequence of deep features passes to the LSTM layer rather than directly passing them through the fully connected layer for classification. The CNN network efficiently extracts and recognizes the image’s local and global structures in the pixel series, while the LSTM network detects long-short-term dependencies [57] . In order to benefit from the characteristics of the two models, the proposed approach introduces a combined CNN-LSTM network for auto-identification and classification of Covid-19.…”

A Neural Network Ensemble With Feature Engineering for Improved Credit Card Fraud Detection

“…Meanwhile, * can represent f , i, or o to denote the specific gate or c for the memory cell. Therefore, V * and W * are the weight matrices, h * represent the hidden state, b * is the bias, h t is the output vector at time instant t. Furthermore, σ and tanh are the sigmoid and tanh activation functions [15]. The operator ⊗ represents the Hadamard or element-wise product.…”

“…Despite the numerous studies proposed to handle imbalanced data, this problem remains a challenge, especially in credit card fraud detection [6]. Since the advent of deep learning, recurrent neural networks (RNN), such as long short-term memory (LSTM) and gated recurrent units (GRU), have shown enormous potential in modelling sequential data [15]- [17]. Conventional machine learning algorithms have not been successful in credit card fraud detection because they do not adapt to the dynamic shopping trends of credit card clients, which results in misclassifications when used for fraud detection systems [18].…”

New Robust Regression Method for Outliers and Heavy Sparse Noise Detection via Affine Transformation for Head Pose Estimation and Image Reconstruction in Highly Complex and Correlated Data: Applications in Signal Processing

“…Zhang et al [33,34] proposed a low-rank-sparse subspace representation for robust regression (LRS-RR) method to find the clean lowrank part by low-rank subspace recovery along with regression to deal with errors or outliers lying in the corrupted disjoint subspaces. To resolve this, Zeng et al [35] addressed labeled-robust regression, but its performance is not yet promising to denoise the high dimensional images, particularly in signal processing; to tackle this, Wu et al [36,37] addressed the sparse prior information.…”