RMIT/CSIRO Next-Gen PhD in Wheel-end failure detection in heavy vehicles using IOT and Machine learning

Data61/CSIRO Next Generation Artificial Intelligence Graduates program and Delta-V Experts will provide a PhD stipend scholarship for 3.5 years of $40,500 per annum. The program will also supply $5,000 per annum for training expenses and $5,000 per annum for travel expenses for the candidate. The program will further provide each candidate with a 6-month industry placement opportunity.

One (1)

The successful candidate is expected to demonstrate familiarity with machine learning methods, especially deep learning models and artificial neural networks, as well as being comfortable with above programming languages. General knowledge on mechanical vibrations, rotating equipment and dynamic systems is desirable.

Interested applicants should email Dr Hormoz Marzbani with the following information.

1. Cover letter outlining interest and alignment with the proposed research, and addressing the eligibility criteria (also see "Further Information" below).
2. CV (including the names of two referees).
3. Copy of Honours/Masters thesis and/or first-authored published peer-reviewed conference/journal paper(s).

Shortlisted applicants will then be invited to an interview. The selected applicant will be supported to develop a formal PhD application to RMIT. A scholarship application will only be successful if the prospective candidate is provided with an offer for admission from RMIT University.

Open now.

The application will close once the position is filled.

The goal of this project is to investigate, design and implement machine learning algorithms for the purpose of anomaly and fault detection in vibrating mechanical systems in real-time. This investigation includes working in vibrations lab and hands-on experience with experimental measured data, as well as implementing the AI algorithms using Python, C++, and MATLAB. The core of the project will be collecting and analysing measurement data from vibrating mechanical systems (rotating equipment mostly) and design and development of required machine learning algorithms to identify known faults by classification methods, as well as unknown faults by clustering methods. Ultimately the best suited algorithms are to be implemented in a prototype “automated” system that can notify the user of emerging failures and provide a framework for predictive maintenance of the mechanical system. The data collected throughout the project will then be used for generating data-based scheduled maintenance procedures for the operators. The successful candidate will need to demonstrate:

• Excellent written and verbal communication skills. 
• Strong computational, programming, algorithms, and data analysis skills. 
• Outstanding research skills
• Capacity to work independently and as a part of a team."

Dr Hormoz Marzbani