Application of LSTM approach for modelling stress–strain behaviour of soil

This paper presents a new trial to reproduce soil stress–strain behaviour by adapting a long short-term memory (LSTM) deep learning method. LSTM is an approach that employs time sequence data to predict future occurrences, and it can be used to consider the stress history of soil behaviour. The proposed LSTM method includes the following three steps: data preparation, architecture determination, and optimisation. The capacity of the adapted LSTM method is compared with that of feedforward and feedback neural networks using a new numerical benchmark dataset. The performance of the proposed LSTM method is verified through a dataset collected from laboratory tests. The results indicate that the LSTM deep-learning method outperforms the feed forward and feedback neural networks based on both accuracy and the convergence rate when reproducing the soil’s stress–strain behaviour. One new phenomenon referred to as “bias at low stress levels”, which was not noticed before, is first discovered and discussed for all neural network-based methods. •We proposed an approach to model the stress–strain behaviour using LSTM deep learning method.•The superiority of LSTM method was testified based on the designed numerical experiments.•A laboratory test on saturated sand was studied to demonstrate capability of LSTM model.•A common phenomenon called the bias at low stress levels for NN method was discussed.