- Home
- About veski
- veski board
- veski innovation fellows
- Timothy Scott
- Benjamin Marsland
- Pierluigi Mancarella
- Vihandha Wickramasinghe
- Jon Shah
- Roger Pocock
- Richard Sandberg
- Colby Zaph
- Kenneth Crozier
- Ethan Goddard-Borger
- Colette McKay
- Luke Connal
- Mark Dawson
- Cameron Simmons
- Tiffany Walsh
- Seth Masters
- Christopher McNeill
- Matthew Call
- Edwin van Leeuwen
- Mark Shackleton
- Ross Dickins
- Ygal Haupt
- Sarah Hosking
- Michael Cowley
- Alyssa Barry
- Gareth Forde
- Marcus Pandy
- Andrew Holmes
- veski fellows
- organisational structure
- veski annual review
- veski impacts
- veski standard
- veski pin
- Contact us
- veski foundation
- Fellowships
- Programs
- News & Events
- News
- Events
- Galleries
- Newsletters
- in conversation
- veski twitter
- veski family in the media
- veski's portraits of innovation
- A banquet of problems to be solved
- A novel approach
- A very special challenge
- At the crossroad of sport and science
- Engineering a better quality of life
- Everything at her fingertips
- Forward propulsion
- Going to the ends of the earth to cure melanoma
- His link to the past and bridge to the future
- Hitting the right note
- Holding up his side of the bargain
- Lighting the way to better child cancer outcomes
- Links and reconnections
- Mining his talent to make a difference
- Putting Melbourne's science on the global stage
- Ready, set, go: the future of locomotion
- Setting his own path
- Springboarding into a slam-dunk for science
- The lens of experience
- Where dreams are made
- veski videos
- People
- veski board
- veski innovation fellows
- Timothy Scott
- Benjamin Marsland
- Pierluigi Mancarella
- Vihandha Wickramasinghe
- Jon Shah
- Roger Pocock
- Richard Sandberg
- Colby Zaph
- Kenneth Crozier
- Ethan Goddard-Borger
- Colette McKay
- Luke Connal
- Mark Dawson
- Cameron Simmons
- Tiffany Walsh
- Seth Masters
- Christopher McNeill
- Matthew Call
- Edwin van Leeuwen
- Mark Shackleton
- Ross Dickins
- Ygal Haupt
- Sarah Hosking
- Michael Cowley
- Alyssa Barry
- Gareth Forde
- Marcus Pandy
- Andrew Holmes
- Victoria Prize recipients
- Victoria Fellows
- veski sustainable agriculture fellows
- veski inspiring women fellows
- veski connection
- PAHMR recipients
Supercomputers to propel Richard Sandberg’s jet engine research
Melbourne University researcher Richard Sandberg is hoping the three millenniums worth of research he plans to do in a single year will help develop cleaner and more environmentally friendly jet engines.
Professor Sandberg will use 100 million core computer hours on some of the world’s fastest computers to conduct research into the flow of gases in aircraft engine components such as low-pressure and high-pressure turbines.
He believes the use of supercomputers made possible by the US Department Of Energy’s INCITE program and Swiss supercomputing centre CSCS will allow him to make advances in turbine technology that is difficult to obtain with existing tools.
The machines are up to 50 times faster than those currently available in Australia. The researcher has estimated he would need 25 million hours, or 2845 years, to do the same work on a desktop or laptop computer.
The chair of computational mechanics at the Melbourne School of Engineering is working with US engine manufacturer General Electric on two projects.
“We’re looking at the turbine and trying to understand really the very detailed physics of the flow going through the turbine,’’ he said. “By learning what exactly happens, we can help the designers tweak these kinds of components to be more efficient.
“But the other aspect of the research is really also to develop better models from these data sets that we generate.’’
Professor Sandberg said it was impossible to spend 50 million hours of computing time on each iteration of the design process and the industry needed lower-order models that could do design work much faster. These tools would help produce the next generation of more efficient engines with fewer emissions.
“The current generation of the models that they use are good and calibrated for particular situations but as soon as you go off design they don’t really trust them any more and you can’t rely on them.
“What we want to do is really use those data bases that we’re generating with the high-fidelity approaches to improve the models that industry can use.’’
Work has already begun on the Swiss supercomputer and will start on the US system, the world’s second-biggest in terms of operations per second from January 1.
Some of the simulations will take weeks, or even months, despite being run on two of the world’s biggest computers.
Professor Sandberg said the two computers were similar in that they used graphics processing unit (GPU) accelerators in conjunction with central processing units (CPU) to give them their processing speed. The development of GPUs was driven by the computer games industry, but they are now being used to benefit other areas such as science.
The power of computers is continuing to increase and Professor Sandberg said today’s supercomputers could be readily available to engine designers in 10-15 years.
Professor Sandberg’s project will account for almost half Australia’s 2016 National Computational Merit Allocation Scheme allowance.
Source: The Australian
veski connection members in the news
Apr 2020 | Royal Society
Prof Jane Visavader, 2018 Victoria Prize for Science & Innovation recipient, elected to the Royal Societyin 2020
“The real benefit of increasing fabrication rates is the transition from prototyping, making one offs, to actually going into production.”
Assoc Prof Timothy Scott
Nov 2019 | Bionics Institute
Dr Thushara Perera, 2016 Victoria Fellow, received the prestigious AMP Foundation’s Tomorrow Fund
Tweets from @veskiorg
Tweets by @veskiorg