Newsroom

Fill in your details for the latest course guides.

  1. Please enter your first name
  2. Please enter your last name
  3. Please enter a valid email address
  4. Choose your course guide*
    Please select a course

Deakin students deliver viable, industry-valuable solutions on the grid-charging patterns of EVs.

In a research competition-first, Deakin students have delivered high-precision data-modelling that will help energy providers manage the demand of electric vehicles across Australia’s power grid. 

The Deakin EV Challenge 2022 is part of an innovative research project with industry partners United Energy (UE)and Centre for New Energy Technologies (C4NET) where multi-disciplinary student research teams developed algorithms to detect the presence of an EV at a network charging site (such as a household).

Competition provides industry solutions

Leading the research project, Deakin Business School’s (DBS) Professor Rens Scheepers says the competition – attracting a $20,000 prize pool sponsored by both UE and C4NET – provides results that will help UE understand and plan for energy demand across its grid network.

‘If a lot of EVs are charging in a particular location, it puts particular demands on the grid, so providers need to know where the uptake is and how to accurately invest in capacity, infrastructure planning and upgrades. This competition has been working with big data and providing predictive analytics solutions to help providers manage power demands, optimise the use of clean energy, and better reduce emissions,’ he explains.

Using data from de-identified sites across UE’s grid, ten Deakin teams – made up of academic mentors and up to three students from DBS and Deakin’s Faculty of Science Engineering and Built Environment – developed algorithms for UE that were evaluated for performance and ranked according to their merit criteria. 

Prof. Scheepers (who directs DBS’s Business and Technology research theme) says that in the world of machine learning, algorithms like this are developed through the provision of training data.

‘This is time-series data from (de-identified) households where known electric vehicles are being charged on the grid. The testing data was sent to the research teams – who did not know if there was an EV being charged or not – who then developed detection algorithms. Energy providers need these algorithms to be as accurate as possible – minimising false positives – about whether or not an EV is being charged at locations as that ultimately informs demand planning.’

The complete list of academic mentors and student members awarded a monetary prize for their efforts can be found at the end of this article. 

‘The EV Detection Competition introduced me to the challenges and limitations we face when dealing with real world data and served as an incredible opportunity to apply what I've learn at university to solve a real problem,’ says Yotam Barazani, a student on the winning team. ‘The challenge called upon my team's and my creativity, critical thinking and problem-solving skills.’

Associate Professor Lemai, academic mentor for the second placed team adds, ‘It was an amazing and rewarding experience. We had to explore and stretch our thinking when dealing with the unfamiliar challenging real-world problem space with many technical details from the large datasets.

'Through multiple meetings we learned domain expertise from UE and C4NET teams and received feedback on our proposed solution models. We experimented with various analytical and creative approaches and carefully evaluated the performance of our deep learners, accounting for efficiency in the training and calibration processes and we were thrilled with the outcome. My team and I feel so honoured to contribute to this real-world initiative for the greater good.'

‘While the best solution may have taken out top prize, Prof. Scheepers says it’s a competition with many winners. 

‘UE preferred a diversity of solutions and approaches and this competition – alongside the research project – is providing a range of options to explore. And while the prize money is fantastic for the students, the real benefit is gaining relevant industry work experience.’

An outstanding industry collaboration 

DBS Dean Professor Amanda Pyman says the research project and embedded competition has been an outstanding example of industry-university collaboration.

‘This is a fantastic project, at the intersection of business, technology and sustainability, which is a major focus for us at Deakin Business School. That’s why we’re excited to be invited to work with partners like United Energy and C4Net who have presented our students and researchers with an authentic data science challenge like this one. We’re incredibly proud of the work our students and their academic mentors have put into this competition and of the success of their data models. It is especially gratifying that the work of our students and academics can contribute smart solutions to deliver better outcomes for consumers, our partners and the environment.’    

But more broadly, there are important big-picture benefits as Prof Scheepers says, ‘This research provides energy providers a range of viable solutions for grid support which improves energy demand management and capacity for consumers to charge their vehicles. In turn, this benefits society as a whole’. 

Deakin’s student research talent

UE’s Head of Network Intelligence, Tobie de Villers, says the organisation has been impressed by the capabilities of Deakin’s student research teams.

‘These talented students have researched and tested potential solutions that will provide us the insight we need to accommodate more electric vehicles on our network. Knowing where electric vehicles are located and are being charged will give us the data we need to understand what investments we need to make in our network to support our customers. We value our collaborations with universities and are proud to be part of bringing together applied academic expertise with real world industry applications.’

C4NET’s CEO James Seymour says that UE has engaged some of the freshest minds in data analytics to bring new techniques and approaches to a sector-wide globally significant challenge. 

‘Supporting skills development is at the core of C4NET’s purpose, hence we're delighted to support both Deakin and UE’s innovation leadership, and in doing so contributing to the students’ development. Let’s hope for some of them it will be the first step of a long career in the energy sector. Initiatives like this create a unique opportunity to bring new thinking to the sector and highlights the benefits of enabled collaboration between universities, industry and government. For electricity providers and consumers, this means that it is real customer behaviour that informs the design of the energy grids and markets of the future.’

 

Prize Winners

 

First Prize

Student Members: 
Mr. Yotam Barazani
Mr. Benjamin Archbold
Dr. Fatemeh Ansarizadeh
Academic Mentors: 
Dr Adnan Anwar
Dr Valeh Moghaddam
Dr Sutharshan Rajasegarar 

Second Prize

Student Members: 
Mr Thuc Nguyen
Mr Bao Duong 
Academic Mentors: 
Associate Professor Lemai Nguyen
Dr Thin Nguyen

Third Prize

Students Members:
Ms. Thi Hoa (Hannah) Nguye
Mr. Islam Khalil
Mr. Nabeel Maqsood
Academic Mentor:
Dr. Prasad Sankar Bhattacharya

Joint Fourth Prize

Student Members:
Mr Yang Cao
Ms Xiaoyan Wang
Academic Mentor:
Dr Quan Vu

Joint Fourth Prize

Students Members:
Ms Lusi Xiao
Ms Brigitta Febriani
Academic Mentors:
Dr Ali Tamaddoni
Dr André Bonfrer

The Centre for New Energy Technologies (C4NET) has contributed to the funding of this project. C4NET acknowledges the major funding contribution of its Core Participants and the Victorian Department of Environment, Land, Water and Planning. The views expressed herein are those of the author and not necessarily the views of C4NET

Photography by Simon Peter Fox Photography

 

Tags

work integrated learning competitions sustainability community impact technology

Connect with us to download course information

Find out more