MSC-1DCNN-based homogeneous slope stability state prediction method integrated with empirical data

The mechanism of slope stability prediction is formulated based on its material, geometrical and environmental situation, and slope stability prediction has been accepted as a tool for analyzing and predicting future structure stability based on geotechnical properties and failure mechanisms. However, the study of slope instability is complex and usually difficult to explain by mathematical methods. The number of slope cases limits the accuracy of slope stability prediction, and the variability of soil or rock parameters of slopes poses new challenges for prediction using conventional algorithms. To improve the accuracy of slope stability state prediction, this paper proposes an efficient slope stability state prediction method with a highly robust convolutional neural network named the multiscale, multichannel, one-dimensional convolutional neural network (MSC-1DCNN) and substantial empirical data collected worldwide. The collected dataset is amplified. Additionally, the probability of failure is calculated considering the variability of soil or rock parameters. Compared with some state-of-the-art prediction methods, the MSC-1DCNN presents high prediction accuracy. The proposed method is applied to a slope, and the results indicate that this paper provides a reliable slope stability state prediction method for homogeneous slopes worldwide.