Dr Adam Berry on the Virtual Power StationPosted: Monday, 12th December, 2011
Adam Berry is a research scientist at CSIRO Energy Technology and works for the Demand-Side Energy Systems group. Recently, at a public talk in Newcastle, he gave some extra insight into benefits of the Virtual Power Station. The VPS is technology that CSIRO and Lake Macquarie City Council are currently trialling, and it’s been mentioned previously on the blog here. I’ve sat in on and enjoyed the talk and, with Adam’s permission, I’ve ‘blogified’ it for you below.
◊ ◊ ◊
A few years ago I had a rush of enthusiasm and bought myself an exercise bike. For the first little while I used it a lot – but then (as I maybe should have guessed) the novelty wore off. Now I seem to spend more time cleaning cobwebs from the seat than I do sitting on it.
I mention this because there’s an analogy here for what can happen with renewable energy. We tend to get interested in the issues and motivated to help make a difference – sometimes we even install solar panels on our own rooftops – but then our engagement wanes and we forget there’s more we can do, even with the solar panels we already have.
What evidence do I have that we’re not engaged with the topic? Well, here’s anecdotal evidence from the number of Google searches carried out in NSW over the last year, comparing relative numbers of searches on ‘renewable energy’ and ‘CO2’ with another subject that’s arguably less of a critical current global issue:
And here’s some more sobering evidence, from the Garnaut Report:
As a matter of fact, Qatar, Kuwait, Bahrain, Bolivia and Brunei are the only countries in the world with higher per-capita emissions than Australia [source].
Enter the Virtual Power Station Technology Trial, a pilot project CSIRO and Lake Macquarie City Council are carrying out in the Lake Macquarie region over twenty different sites. The project aims are twofold. Firstly, we want to increase the engagement homeowners have with their rooftop solar panels, so that they get more energy – both electric and motivational – out of them as a result. And secondly, by networking the panels into a larger ‘virtual power station’, we can improve the predictability, quality and response of the power being put into the grid – in other words, changing a day’s power output from something that would typically look like this:
to something that (ideally) looks more like this:
How do we do this? The first step is to install a little black box in each volunteer’s home and connect it to their solar system. It logs the power output of the rooftop panels and beams the data to CSIRO’s central computer.
This data gets put on the internet, where homeowners can check in to keep an eye on how their system is performing, how they are contributing to the total network power output, and how their household compares to the Jones’s system up the road!
The advantage of this isn’t just that we get a smoother and more reliable power output from the aggregated system than we would have had from any individual panel. We’ve also found that it benefits the users by keeping them engaged, letting them see the value their panels add to the system (and to their income), and giving them fast notice of any faults so they can be fixed.
On a larger scale, the grid benefits too from having networked systems that can harness distributed generation units like this. Not only is there the ‘smoothing’ effect that comes from the sites being spread out, the VPS can also choose when to charge and discharge batteries for extra smoothing, or when extra power is required. And at CSIRO we’re also happy because the project gives us fantastic data streams from which we can learn a lot.
What’s next? One thing we’re working on is using our little black boxes to predict, half an hour or so in advance, how much sunshine (and hence solar panel output) we expect in a given place. When a panel is generating electricity, we know the sun’s out at that location, and if we have data from across a wide enough area we might be able to start tracking – and preparing for – cloud movements. It’s a crude system but we think that by applying forecasting techniques from computer science technology to the system we can get it to make pretty accurate short-term predictions, and our preliminary results are promising.
The upshot of the Virtual Power Station is that it can combine household solar panels into a system that’s better than the sum of its parts. It also engages the owners of the panels so they keep track of how the panels are performing, helping them get the most out of their purchase.
As for my exercise bike – I’m still waiting for a solution.