AI-Enabled Hardware-in-the-Loop Digital Twin for Control, Hardware Validation and Fault Diagnosis of Grid-Connected Power Electronic Converters

Modern electricity networks are being transformed by renewable generation, battery energy storage, electric vehicles, microgrids and other electronically interfaced technologies. As conventional synchronous generation is displaced, grid-connected power electronic converters must provide not only efficient energy conversion, but also dynamic grid support, reliable protection behaviour, resilience during disturbances and continuous health monitoring. However, converter controllers are often developed and tested using software models that may not fully capture real hardware behaviour, communication delays, sensor noise, parameter drift, switching constraints, gate-drive limitations, protection delays, electromagnetic interference, weak-grid interactions or compatibility issues between devices from different manufacturers.
This PhD project will develop an Artificial Intelligence (AI)-enabled hardware-in-the-loop (HIL) digital twin framework for the control, monitoring and fault diagnosis of grid-connected power electronic converters. The project is not limited to software simulation. It will require strong engagement with power electronic hardware, analogue and digital circuit principles, laboratory instrumentation, controller implementation and real-time HIL testing. The candidate will work across converter modelling, embedded control, sensor and signal-conditioning interfaces, experimental measurement, fault emulation and hardware validation. The project will therefore suit a student who is comfortable moving between theory, simulation, AI algorithm development and hands-on hardware experimentation.

Aim

The project aims to improve the reliability, controllability and diagnosability of power electronic converters in renewable-rich power systems through the integration of AI-assisted control, condition monitoring, and fault diagnosis with real-time HIL validation and hardware-based experimentation. In addition, the project will provide the candidate with comprehensive training in the full converter development and validation workflow: from mathematical modelling and controller design to digital implementation, circuit-level testing, fault diagnosis and experimental verification under realistic grid conditions.

Objectives

Develop high-fidelity mathematical and real-time digital twin models of selected grid-connected converter topologies, including switching behaviour, control loops, sensing circuits, gate-drive constraints, protection logic, grid disturbances and converter/grid-side fault scenarios.
Design AI-assisted monitoring and diagnosis algorithms for early detection, classification and localisation of converter faults and abnormal operating conditions, such as open-switch faults, sensor faults, DC-link abnormalities, gate-drive issues, thermal stress, grid-voltage disturbances, weak-grid instability and control-loop degradation.
Investigate robust and adaptive converter control strategies, including model predictive control, reinforcement-learning-assisted tuning, and hybrid physics-informed/data-driven methods.
Implement converter controllers and diagnostic algorithms using embedded control hardware, such as DSPs, microcontrollers, FPGAs or equivalent real-time control platforms, with attention to analogue/digital interfacing, signal conditioning, sampling, PWM generation, protection circuits and communication delays.
Develop and validate the proposed framework through hardware experimentation and real-time HIL testing using platforms such as OPAL-RT, with benchmark case studies relevant to renewable energy integration, microgrids, battery energy storage systems, and weak-grid operation. The project will generate validated algorithms, datasets, models, experimental protocols, and hardware demonstration case studies that support future research, teaching, and industry engagement in power electronics and smart grids.

Significance

This project aligns closely with the EECMS’s sustainability agenda and priority Sustainable Development Goals (SDGs), directly contributing to the achievement of SDG 7 (Affordable and Clean Energy) and SDG 9 (Industry, Innovation and Infrastructure). It also supports the EECMS’s top priority in research capability development and the 2026–2036 Strategic Vision for the Green Electric Energy Park (GEEP), which identifies digitalisation, power electronics, renewable energy integration, and advanced testing and validation platforms as key strategic focus areas.
A significant aspect of the project is its utilisation of Curtin’s newly established real-time simulation and Hardware-in-the-Loop capability. As part of the first phase of the GEEP research infrastructure expansion, advanced HIL equipment has been acquired to support future smart grid, renewable energy, and power electronics research. This project is expected to be among the first to leverage this capability, providing an important opportunity to demonstrate the value of the new infrastructure and showcase the EECMS’s emerging research strengths in digital twins, real-time simulation, and converter validation.
The project addresses a critical engineering challenge in the clean energy transition: ensuring that converter-dominated power systems remain stable, secure, and fault-tolerant as renewable generation, energy storage systems, and distributed energy resources become increasingly prevalent. The expected outcomes include novel AI-enabled approaches for converter health monitoring, faster fault identification, safer controller testing, improved pre-commissioning confidence, and enhanced converter performance under realistic grid operating conditions.
The significance of the project further lies in its integration of AI, power electronics, digital twins, and hardware-based validation. While many AI-driven diagnostic and control approaches remain confined to simulation environments, this project aims to bridge the gap between theory and practice by embedding AI-assisted methods within real-time control platforms, analogue and digital hardware interfaces, laboratory experimentation, and HIL validation environments. The resulting methodologies, validated models, and demonstration case studies will be directly relevant to renewable energy developers, network operators, inverter manufacturers, HIL testing facilities, and researchers working on grid-forming and grid-following converter technologies.

Ideal Candidate

We are looking for a self-motivated PhD candidate with a strong foundation in electrical/electronic engineering, particularly power electronics, analogue circuits, digital electronics, and control systems. Hands-on hardware capability is essential, including experience with laboratory instruments, circuit testing, signal measurement, PCB-level debugging, embedded control, microcontrollers, DSPs, FPGAs or HIL platforms. Skills in MATLAB/Simulink, PLECS, Vivado, Verilog or machine learning are desirable. Experience with renewable energy systems, converters, gate drivers, sensors, PWM control, fault diagnosis or real-time simulation will be an advantage. Must be eligible to enrol in a PhD program at Curtin.

This project is open to International and Domestic applicants.

Internship

The project will provide opportunities for industry internships with leading organisations that apply power electronics to renewable energy generation, energy storage, and grid integration. Through these placements, the candidate will gain invaluable hands-on experience with real-world engineering challenges, industry-standard tools, and advanced converter technologies. The internship component will complement the research activities by exposing the candidate to practical implementation, testing, and operational aspects of modern power and energy systems, enhancing their professional skills and preparing them for a successful career in the renewable energy and power electronics sectors.

Scholarship

This project does not have a scholarship attached. International applicants must be able to demonstrate that they are able to self-fund their studies, including tuition fees and living expenses, for the duration of the course.

If you’re an Australian citizen, permanent resident or New Zealand citizen and you’re accepted into a Higher Degree by Research (HDR) program, you are eligible for a tuition fee exemption through the Australian Government’s Research Training Program (RTP). This exemption (known as the RTP Fee Offset) covers tuition fees for the standard duration of your higher degree by research.