SiS Intelligent Energy Management Platt 2007


Published on

Scientists in Schools Program - Presentations from the Energy and Climate Change Symposium

Published in: Business, Technology
  • Be the first to comment

  • Be the first to like this

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

SiS Intelligent Energy Management Platt 2007

  1. 1. Intelligent Energy Management Bringing Brains to the Brawn.. (!) Energy Transformed Flagship Glenn Platt Group Leader, CSIRO Energy Technology
  2. 2. How The Market Has Changed
  3. 3. Distributed Energy • Clean, efficient, small electricity generation technology- small solar systems, gas generators, etc. • Much more intelligent control in how we use energy- demand side management & other “new” areas are part of DE • Of significant policy interest • Of significant research interest • Of significant commercial interest
  4. 4. Distributed Energy???
  5. 5. Management and Control Systems • How do we coordinate all these individuals into doing something useful? • How can we minimise greenhouse gas emissions? • How can we improve existing infrastructure? • How do we make these things cheap? • How do we deal with dynamic and changing situations? • How can we keep the customer happy? • How can we provide a “firmness” of response? • Electrical engineers, computer scientists, software engineers, mathematicians
  6. 6. Heating, Ventilation Air-Conditioning • HVAC is a very significant consumer of energy in Australia • HVAC also causes a lot of peak-load problems • How can we minimise the energy consumed? How can we smooth out the peaks? • Pre-cool the building • Coordinate with local generation • We need to learn a model
  7. 7. HVAC agent
  8. 8. The Coordination of Multiple Devices • We need to coordinate all these loads and generators • Importantly, traditional control systems are not appropriate here- scalability and depth of control • We’ve chosen to apply techniques from the discipline of multi-agent systems science to our problem. • Agents are simple, yet can achieve complex outcomes
  9. 9. The Coordination of Multiple Devices Emergent behaviour- when large numbers of programs interact in a connected environment, various phenomena occur which are not explicable in terms of the programming or behaviour of any single agent.
  10. 10. The Coordination of Multiple Devices Movie!
  11. 11. Agents in a Residential Setting
  12. 12. The Virtual Power Station • At the moment, the economics of installing your own solar system are fairly poor: • Low payback rate • Limited usefulness to the wider network • If we can aggregate lots of small solar systems in to one big coordinated response, this would have a much greater use • The “Virtual Power Station” • Can significantly improve the payback for the individual • Need to be able to provide predictions • Machine learning again!
  13. 13. Utility Simulation
  14. 14. Utility Simulation
  15. 15. Conclusions • One of the biggest opportunities we have is to simply be more intelligent in how we utilise our existing systems • Technology from artificial intelligence, electrical and software engineering, can assist the incumbent practitioners here “Its called demand management, and if your kid is looking for a job with a future, they should enrol in a course soon.” -Australian Financial Review, 23 Oct 2004 • Importantly, these are not “energy” practitioners!
  16. 16. CSIRO Energy Technology Glenn Platt Demand Side Energy Systems Phone: +61 2 49606120 Email: Web: Contact Us Phone: 1300 363 400 or +61 3 9545 2176 Email: Web: