Prediction of bearing capacity of pile foundation using deep learning approaches

• Published in: Frontiers of Structural and Civil Engineering Vol. 18; no. 6; pp. 870 - 886
• Main Authors: Kumar, Manish, Kumar, Divesh Ranjan, Khatti, Jitendra, Samui, Pijush, Grover, Kamaldeep Singh
• Format: Journal Article
• Language: English
• Published: Beijing Higher Education Press 01.06.2024; Springer Nature B.V
• Subjects: Accuracy; Algorithms; Artificial neural networks; Cities; Civil Engineering; Comparative analysis; Countries; Deep foundations; Deep learning; Engineering; Information processing; Load tests; Long short-term memory; Machine learning; Neural networks; Performance evaluation; Pile bearing capacities; Pile foundations; Pile load tests; Predictions; Recurrent neural networks; Regions; Regression analysis; Research Article; Root-mean-square errors; Statistical analysis; Statistical models; Structural integrity; high-strain dynamic pile test; bearing capacity of the pile; deep learning algorithms
• ISSN: 2095-2430; 2095-2449
• DOI: 10.1007/s11709-024-1085-z

The accurate prediction of bearing capacity is crucial in ensuring the structural integrity and safety of pile foundations. This research compares the Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (BiLSTM) algorithms utilizing a data set of 257 dynamic pile load tests for the first time. Also, this research illustrates the multicollinearity effect on DNN, CNN, RNN, LSTM, and BiLSTM models’ performance and accuracy for the first time. A comprehensive comparative analysis is conducted, employing various statistical performance parameters, rank analysis, and error matrix to evaluate the performance of these models. The performance is further validated using external validation, and visual interpretation is provided using the regression error characteristics (REC) curve and Taylor diagram. Results from the comparative analysis reveal that the DNN (Coefficient of determination ( R 2 ) training (TR) = 0.97, root mean squared error ( RMSE ) TR = 0.0413; R testing (TS) 2 = 0.9, RMSE TS = 0.08) followed by BiLSTM ( R TR 2 = 0.91, RMSE TR = 0.782; R TS 2 = 0.89, RMSE TS = 0.0862) model demonstrates the highest performance accuracy. It is noted that the BiLSTM model is better than LSTM because the BiLSTM model, which increases the amount of information for the network, is a sequence processing model made up of two LSTMs, one of which takes the input in a forward manner, and the other in a backward direction. The prediction of pile-bearing capacity is strongly influenced by ram weight (having a considerable multicollinearity level), and the effect of the considerable multicollinearity level has been determined for the model based on the recurrent neural network approach. In this study, the recurrent neural network model has the least performance and accuracy in predicting the pile-bearing capacity.