Estimation of Loading Capacity of Piles Using Machine Learning

The current proposed project seeks to enhance the accurate estimation of the loading capacity of piles through advanced machine learning (ML) methods. Accurate prediction of the loading capacity of piles under axial, lateral, and torsional loading is critical for ensuring structural integrity and safety.
Recent advancements in artificial intelligence (AI), particularly in ML and deep learning (DL), have significantly improved predictive modeling in geotechnical engineering. This research will investigate various machine learning techniques, including Deep Neural Networks (DNN), Convolutional Neural Networks (CNN), Recurrent Neural Networks (RNN), Long Short-Term Memory (LSTM), and Bidirectional LSTM (BiLSTM) algorithms. The study will specifically focus on applying these approaches to predict the loading capacity of piles in clay, sand, and rock environments.

This project aims to investigate the loading capacity of piles using machine learning techniques. By leveraging advanced machine learning approaches, the project seeks to reduce uncertainty in estimating the loading capacity of pile foundations, addressing the various geotechnical challenges inherent in their design and construction.

Expected outcomes of the project include:
• Conducting a rigorous evaluation of machine learning (ML) and deep learning (DL) techniques, providing a comprehensive review of their application to this geotechnical challenge.
• Strengthening the collaboration between Flinders University and Geotechnical Engineering Technology Consulting Pty Ltd.
• Offering valuable training opportunities for PhD students, equipping them with expertise in applying ML and DL to geotechnical engineering problems.