HDR Scholarship - Physics Informed Machine Learning

There has been tremendous progress in modelling complex multiphysics multiscale problems using numerical methods in recent years. However, the numerical methods rely on space-time discretization of generally nonlinear partial differential equations which are often computationally expensive. On the other hand, recent developments in big data analytics, increased the interest in combining sensor data with physical models in real-time. On the other hand, the access to large amount of data from complex systems in real-time becoming more and more feasible. This has resulted in increased interest to combine real-time noisy sensor data with numerical models of underlying physical laws of the systems. Doing this while using current numerical methods, mesh generations and solutions, is challenging, computationally expensive, and not practical when real-time predictions are needed. This research will investigate using deep learning algorithms with enforced physical laws to model structural behaviour under various loading conditions. The project is done in consultation of Austal Ship building, who will providing assistance with all stages of the project.

Applications will remain open until a candidate has been appointed

This scholarship is available over 3 years.

• Stipend of $28,600 per annum tax exempt (2021 rate)
• International students only:  Tuition fees offset for the duration of 4 years. Single Overseas Student Health Cover policy for the duration of the student visa.

To be eligible you must:

• be either a domestic or international candidate currently residing in Australia. Domestic includes candidates with Australian Citizenship, Australian Permanent Residency or New Zealand Citizenship.
• meet Deakin's PhD entry requirements
• be enrolling full time and hold an honours degree (first class) or an equivalent standard master's degree with a substantial research component.

Please refer to the research degree entry pathways page for further information.