Episode 107      23 min 07 sec
Getting it on the grid: Integrating renewable energy into our power supplies

Energy and climate change analyst Dr Roger Dargaville weaves together diverse technical, economic and environmental factors to produce a model for better, smarter use of our energy supplies. With science host Dr Shane Huntington.

"The goal is to look at how to design a system of renewable energy made up of wind, solar and other technologies that gives you the best mix because every renewable energy technology has a different characteristic." -- Dr Roger Dargaville




           



Dr Roger Dargaville
Dr Roger Dargaville

Dr Roger Dargaville works in the fields of climate change, carbon cycle and energy. One of his current research themes is modelling of energy systems, looking particularly at optimisation of renewable energy systems under Australian conditions. This work will help find ways to achieve the maximum reduction in carbon emissions at the lowest cost, and will reduce uncertainty in renewable energy output estimations.

Roger has a background in atmospheric science and currently studies the impacts of changes in stratospheric chemistry on the climate of the Southern Hemisphere, especially the Antarctic Ozone Hole.

Dr Dargaville is a research fellow with Federation Fellow Prof. David Karoly in the School of Earth Sciences. He completed his PhD at the University of Melbourne, and has worked at the University of Alaska, Fairbanks; National Center for Atmospheric Research in the USA; CNRS; UNESCO and the International Energy Agency in France.

Credits

Host: Dr Shane Huntington
Producers: Kelvin Param, Eric van Bemmel
Associate Producer: Dr Chrstine Bailey
Series Creators: Eric van Bemmel and Kelvin Param
Audio Engineer: Gavin Nebauer
Voiceover: Nerissa Hannink

View Tags  click a tag to find other episodes associated with it.

Download file Download mp3 (22.3 MB)

Getting it on the grid: Integrating renewable energy into our power supplies

VOICEOVER
Welcome to Up Close, the research, opinion and analysis podcast from the University of Melbourne, Australia.

SHANE HUNTINGTON
I’m Shane Huntington. Thanks for joining us. Meeting our growing energy needs from clean renewable sources is a major challenge, one that must be addressed in the reasonable timeframe if the worst predictions from climate models are to be avoided. Underpinning the problem are numerous and complex technological, environmental and economic factors and coming to grips with all of it requires contributions from a diverse range of disciplines. Our guest today on Up Close is Dr Roger Dargaville. Roger is an energy and climate change analyst from the school of earth sciences and the Melbourne Energy Institute at the University of Melbourne. He’s working on a complex model that he believes will make better and more efficient use of energy systems in Australia and elsewhere in the world. Welcome to Up Close Roger.

ROGER DARGAVILLE
Thanks Shane, good to be here.

SHANE HUNTINGTON
Firstly, let’s talk about the kind of reduction in renewable energy usage that will be required at this point in countries such as Australia but across the world in order for us to do our part in climate change mitigation.

ROGER DARGAVILLE
In Australia the carbon emissions, because our energy production is very carbon intensive, is probably one of the highest in the world so we’re running at about 20, 25 tonnes of carbon per person. The global average is five tonnes. We need to reduce that global average by about 50 per cent by about 2050 if we want to avoid that two degree climate warming predicted by IPCC. So to go to two-and-a half per person by 2050 means that in Australia we need to reduce our emissions by about 90 per cent per capita overall. So that’s a massive challenge. So 2050 sounds like a long way away but it’s only 40 years so we need to start reducing our emissions very quickly to have any chance of hitting this target.

SHANE HUNTINGTON
Now internationally I could imagine the energy that is required on a daily basis varies from country to country. Where is all this energy coming from and how do we go about comparing that use from one nation to another depending on the types of society that we find?

ROGER DARGAVILLE
Well in Australia the main supplier of energy is fossil fuel and mainly coal fired power stations. They have high inertia, they don’t change the amount of energy they produce very easily so we have a lot of excess energy at night time when consumers aren’t using it so that encourages a lot of heavy industry so by the nature of having coal fired generators we have a very energy intensive economy. So Australia by its nature is very carbon intensive.
If you say for example go to a developing country they will have very unreliable power supplies so their per capita emissions are very low. Of course they are further back on the development curve. Then if you look at somewhere like France which is a highly developed country obviously but they have 70 per cent of their electricity coming from nuclear power. So they have a very low carbon intensive infrastructure because they don’t produce carbon from making their electricity and also their electricity is relatively expensive so they tend to be relatively frugal with their energy consumption. Whereas in Australia we have the cheapest electricity in the world virtually, it’s just a couple of cents per kilowatt to produce it from a coal fired power station and so that really encourages very energy intensive industry.

SHANE HUNTINGTON
You mentioned a couple of cents per kilowatt. What kind of overall energy requirement is there for a country the size of Australia of approximately 21 million people compared to some of the other ones you mentioned like France and some of the developing countries?

ROGER DARGAVILLE
Energy requirements. Interesting question. There’s what you need and what you have. We waste a lot of energy because it’s inexpensive. The solutions are relatively expensive, so to double glaze your windows, to drive more efficiently, to use a more carbon efficient industrial process. These all cost money and so there’s not much incentive to actually invest that money when the cost of electricity is so low in the first place. So the amount of energy we actually need is probably 30 per cent less than what we actually use. The estimate is that we could save 30 per cent just through efficiencies. But then it’s a very philosophical question as to how much we need. A lot of people get by with very little electrical energy and a lot of us use a lot.

SHANE HUNTINGTON
We keep talking about reducing our emissions but presumably with the way our societies are developing but most of the countries in the world, in particularly the developing countries, will be increasing their energy requirements to quite a large amount. What sort of growth are we talking about in terms of energy requirements for a country like Australia or for the world as a whole?

ROGER DARGAVILLE
The question is how much can we reduce carbon emissions while we still provide enough energy for people? So it’s a decoupling of the carbon industry from the economy. We can still very happily produce enough energy for everybody even in a growing population but by moving away from heavily polluting fossil fuel technologies for producing our electricity we can actually still produce the energy without the carbon emissions.

SHANE HUNTINGTON
Let’s focus for a moment on the grid itself. Tell us a bit about how the grid actually functions because as you said there’s obviously a number of power stations that are running flat out but our energy use varies I’m assuming on a daily basis, on a time of year basis, even on a time of week basis I suspect.

ROGER DARGAVILLE
Absolutely. In Victoria, Melbourne is sort of in the centre of the state. The main power producing area is in the La Trobe valley several hundred kilometres to the east of the city so we have some major power lines coming into the city from there. We’re also connected to Tasmania, the island to the south of Australia which actually has a lot of hydro power so we actually receive power from there when demand is high. We’re also linked to South Australia which has significant wind farm capacity so we’re actually importing a fair amount of wind energy from South Australia. We’re also connected to the Snowy River Hydro Scheme which is to the north of the state.
It’s a fairly complicated link up with lots of different sources of power coming into the main population centre in Melbourne. We have very much a diurnal cycle. In the morning people wake up, they switch on lights, they switch on kettles and they start ramping up their power usage. In winter we hit a peak in the early morning and also in the early evening. And in summer on a hot afternoon our main demand for electricity happens in the middle of the afternoon because of air conditioning. In summer it’s much more variable because the weather’s much more variable. So if you get a 45 degree Celsius day in Melbourne which happened last year, we hit about 10 gigawatts of demand compared to four gigawatts of the base load. We have high demand during the week because of people running offices and shops and weekends tend to be quite quiet. Over the Christmas period we see a remarkable closing down of energy consumption while everyone goes away on holidays and it goes down to about two gigawatts for a couple of days.

SHANE HUNTINGTON
You mention these cycles but I assume there are some things that are just constantly on. There must be an energy demand that is required by certain industries that run 24/7 that we need to meet. What percentage of a city like Melbourne’s energy just goes into those constant industries - hospitals, aluminium smelters, things of that nature?

ROGER DARGAVILLE
About 25 per cent of the energy, electrical energy, is used in the domestic sector and the other 75 per cent is used in the commercial and industrial and manufacturing. There’s a diurnal cycle in the commercial side, very much so in the domestic side. But on the manufacturing and industry about half of our energy is pretty much used around the clock. That’s because of this synergy that we’re running coal fired power stations that they have that power available at night. Because it takes about 24 hours to shut down a coal fired power station. There’s no benefit to shutting down the power, it’s there overnight, it’s extremely cheap so people use it.

SHANE HUNTINGTON
I’m Shane Huntington, your host for this episode of Up Close coming to you from the University of Melbourne Australia. Our guest today is Dr Roger Dargaville speaking about energy and the usage of the grid.
Roger, when we talk about the variations in demand and how these power stations that we currently have in many countries are producing the base load - so that’s the minimum we need for consumption. How do all the renewable type energy producing areas like wind and solar and so forth fit into that game?

ROGER DARGAVILLE
In Australia we have maybe one per cent of our power coming from wind power. That one per cent just disappears in the noise. If it’s a windy day and they’re producing a lot of power the coal fired power stations and gas fired power stations just ramp back a fraction. It’s barely noticeable. But then if you go to a country like Denmark which has a very large percentage of their power coming from wind power, they have an issue that when the wind power drops off they’ve actually got to take quite serious action to shore up supply and they tend to import electricity from other countries. To the north, Scandinavian countries have quite a lot of hydro power so they’ll buy hydro power when they need it. They have the luxury that their neighbours can actually supply power, they can buy nuclear power from France if required through the European grid.
The way that the system works, you generally speaking have an auction house, a clearing house that manages the market. All the different suppliers will put in a bid for how much power they can supply at a price and then depending on demand, the market operator will tell which operators to provide power. If you’re a wind operator you don’t have a choice, if you’ve got it you provide it. If you’re a solar the same. Then the fossil fuel powered industries - or the industries that you can control like hydro as well will vary their supply in response.

SHANE HUNTINGTON
You’re working on a model that will helps us to better develop and sustainably secure our energy needs. Tell us a bit about the specific aim of this model overall?

ROGER DARGAVILLE
Okay, so the model is called the Melbourne University Renewable Infrastructure Laboratory or MUREIL. The goal of MUREIL is to look at how to design a system of renewable energy made up of wind, solar and other technologies that gives you the best mix because every renewable energy technology has a different characteristic. Solar will produce energy during the day, wind will produce it during windy periods, geothermal for example can produce base load, tidal is far more predictable than wind but tends to be more expensive.
So you have all these different characteristics and also the geographical placement of your energy technology stations greatly impacts when and how much power they can produce. The logic is that we’re trying to work out what is the optimal mix of wind, solar, where do you put them? How does that interact with demand? Because as we’ve mentioned we have our highest demand during the day, solar is able to produce highest supply during the day so that’s an obvious marriage. But then solar can’t produce power at night. Wind does but of course it’s not always blowing. So, the question is what mix of those two is optimal? How much back up do you need? Can you supply the back up through hydro or geothermal? How much fossil fuel back up do you need to have a reliable system?

SHANE HUNTINGTON
This must be an incredibly complex model because my understanding is you bring together weather, the grid demands and also some of the economics. Let’s start with weather. How does the weather element of the data go into the model to give you an idea of what the grid will need to supply?

ROGER DARGAVILLE
We actually have a weather forecasting model. It’s a very sophisticated physical model. It produces very realistic correlations between windiness and sunniness. We really believe that it’s very close to reality. Now we don’t model specific situations, we’re more running sort of in climate so it’s more modelling general situations. We can get out of that if it’s windy at one end of the state we can get a good idea of what the wind will be like at the other and importantly whether it will be sunny in the spots where we’re hoping to put solar power stations.
We look at the model and what it predicts for the power output for all the different technologies and then we select where we think the optimal sites are; where they’ll actually produce a reasonable amount of energy. But we don’t only produce the best sites, we actually look at the what we call suboptimal sites, like a wind location or a location that will produce wind energy, maybe not as much as another site but is actually able to produce energy when the windier site is unable to.

SHANE HUNTINGTON
So is this a form of risk mitigation so you have a scenario where it’s either really, really windy or not windy at all in some of these what we would call optimum sites whereas some of the other sites are never really windy but always average sort of levels of wind available. Is that how that works?

ROGER DARGAVILLE
More or less. Because of the way weather systems move across the country and the state, if you have a windy cold front or a low pressure system which tends to be windy, as it moves across the state you’ll have a windy episode in the west and then that will progress across the state. So if you have wind farms only in the west which is where we’re tending to put them at the moment because this is the windiest part of the state then you’ll get a peak of power coming onto the grid when that’s windy and then after the cold front’s passed it’ll drop off. Now if you have more wind farms in the east of the state they’ll pick up that windy episode as it passes over half a day later say. So they may not be as windy, they won’t give you as much average output but they complement the windier sites.

SHANE HUNTINGTON
You’re really talking there about very long term predictions because we’re actually building infrastructure and the idea would be that we would build some of that infrastructure based on this model. How sure are you that these sort of weather predictions in that longer term are going to give you those right answers?

ROGER DARGAVILLE
We don’t try and predict for any one particular day how windy it’s going to be. If we ruin the model out for a couple of years and then take averages and say this distribution of stations is the best mix to meet demand and take advantage of the natural variability that occurs in the system - of course with climate change it’s possible that the weather systems will change altogether. So there is a risk that it won’t be appropriate in the future. But part of what we do is do a scenario analysis where we look at different climate regimes and see what the weather patterns look like then and see if it makes a big difference to the network that you would build and if it does then we try and build it so that it is compatible suitable to a range of different scenarios.

SHANE HUNTINGTON
The other big aspect of the model of course is the economics elements of what systems you’re building and how they’re supplied. I mean this seems to me to be incredibly complex. Are you able to factor in things like a mythical carbon price that most countries do not have? Is that the level of economics modelling you’re doing?

ROGER DARGAVILLE
We have two levels of economics modelling. There is the market mechanisms, the government policies such as a carbon tax whether it’s an emission trading scheme or just a straight tax. That's actually relatively straightforward. That just adds cost onto producers who produce carbon. The thing which is tricky is actually modelling the supply and demand model.
In Australia we have what’s called the Australian Energy Market Operator and it’s a free market. On half hourly basis as I mentioned, the different operators will bid in for what they will produce at what price. So the price actually varies a lot and at the peak demand times, especially if there’s a failure in a power line somewhere or a failure in a supplier electricity prices can hit $10,000 a megawatt above the average of about $40 a megawatt. So that’s extreme. Overnight when there’s more supply than can be actually soaked up by the grid prices can hit zero. In fact they can even go negative which is a fairly bizarre episode where a consumer can get paid to take up electricity off the grid.
So modelling that kind of variability is really key. Because if a solar power station is producing power during the day during the peak that’s advantageous because that will be when the power prices are higher. You wouldn’t just take the average electricity price and apply that to a solar power station. Likewise with a wind power station, if it’s producing power at night then it’s actually not worth very much. It’s a complicated relationship that has to be taken into account.

SHANE HUNTINGTON
You’re listening to Up Close, coming to you from the University of Melbourne Australia. Our guest today is Dr Roger Dargaville speaking about energy and grid optimisation. Roger, the details of this model - how will they be applied? What sort of audience are you hoping to have for these details so that they’ll actually get these sorts of changes that we’re hoping will happen?

ROGER DARGAVILLE
A couple of different audiences. I guess government policy makers would be the main one and trying to get subsidies or other incentives to develop certain technologies is very important and we see the large scale solar thermal power as being one of the key parts of solving the renewable energy problem. The other market is the solar and wind energy sector themselves to encourage the wind farm operators to start building wind farms where it will actually optimise the system. You have to talk to both the policy makers and the wind farm operators at the same time because if there aren’t the correct incentives in place they won’t build the wind farms in the right place and if the wind farm operators aren’t interested in building where they won’t get as many dollars per megawatt hour but will shore up the system then the policy makers won’t be interested in putting the policies in place.

SHANE HUNTINGTON
We’ve talked a fair bit about some of the data you’re using of course being locally here in Australia and particularly in Victoria and Melbourne but I assume that these models are widely applicable across the world to a variety of systems? Can you tell us what areas you’ll be going into next in terms of the applications of the models once they’re complete internationally where, as a relatively small country Australia has a relatively small impact but I can imagine the international scope for this would be quite substantial?

ROGER DARGAVILLE
Australia is a country which has plentiful renewable resources. It’s actually a very easy first case for us to tackle. If you go to other places around the world, for example North America or European countries they have a significantly higher challenge because, especially if you look at Great Britain, seasonality in winter time they have very limited potential to produce power from solar energy so their solution will look very different. But there’s nothing in the model that stops you from using in any environment in any country. If you just plug into it the information about the different technologies that are available and what they cost in that area, the model will find the optimal solution. So in the UK it might be that a back-up system of gas fired power stations is more important than it would be in Australia so they can plan their infrastructure accordingly.

SHANE HUNTINGTON
Tell us a little bit just specifically about the model itself. How is it constructed? How does that work?

ROGER DARGAVILLE
We break it into different aspects of what we call a cost function. A cost function is literally a cost. So we’ll say there’s a cost of building some infrastructure, there’s a cost to using fuel, there’s a cost to meeting supply and the cost of not meeting supply is very high. The model will do everything it can to avoid failing to meet supply. You have obviously the economic cost or the economic value of selling your power at different costs and then you have transmission costs. It adds all these costs together and then it uses a search function to look for the best combination of everything that you’ve told it to find the least cost, the smallest cost. If you say okay, you’ve got a great wind resource over here but it’s going to cost $1 billion to transport the power it will say if you don’t have a lot of power coming down that power line it’s not worth building. So it looks at each aspect, rationalises it, justifies it and then comes up with the best solution.

SHANE HUNTINGTON
Now this must have changed your - as you’re working on this model must have changed your intuition as to what were the best outcomes here. I mean what were some of the things that you sort of changed your mind set on. For example put the wind farm in the windiest always being the best. Some of these things you noticed the early data coming out must have been modified in your thinking.

ROGER DARGAVILLE
Certainly. The initial results from the model - I should say our pilot study is looking at a fairly small domain, just the state of Victoria so several hundred kilometres across - so what I would expect the model to do would be to put a large amount of wind powered infrastructure in the windiest spots and then some back up further away. But what it actually does is it doesn’t really mind that much what the quality of wind resources at any one spot. It just wants to have them as far apart as possible to take advantage of that geographical distance. Then with solar I thought it would try and spread them around a bit but actually it says just put all your solar in the best possible spot that you’ve got, the sunniest spot. There’s actually little advantage to spreading your solar power stations around. So that was a slightly not intuitive result.

SHANE HUNTINGTON
Is it likely that the technology known as solar thermal will have a major impact on the way these grids are working given that it removes that diurnal cycle of only having power production when it is sunny to a scenario where it can be 24 hours a day?

ROGER DARGAVILLE
The technology you’re alluding to is the solar thermal with storage. Solar thermal on its own - a large scale concentrated solar thermal when you have an array of mirrors focussing sunlight onto a central point and then turning that heat into electricity via a steam turbine or whatever, that’s reasonably established. The new technology is solar with storage and what you do is have your array of mirrors as per normal but you heat up a solution of liquid salt which can then be stored in a storage tank. It can be stored very effectively with very small losses of heat over periods of many hours which means that you are no longer required to produce that power only when the sun is shining so it’s a storage system. It’s a storage system that’s embedded in the technology itself. You don’t have to have an external storage system such as battery banks or something somewhere on the grid.
It’s a revolution in the way that renewable are able to be attached to the grid and the technology that actually turns the heat in the liquid salt into electricity is very easy to ramp up and ramp down so it can respond very quickly in the variability in wind farms which are also attached to the grid. So the synergy between solar thermal with storage and wind farms is terrific. It makes both of them viable because the solar with storage but it has this fantastic advantage that it can offset what’s going on in the wind farms in terms of variability.

SHANE HUNTINGTON
Roger, I congratulate you on the work you’re doing. Thank you for being our guest today on Up Close.

ROGER DARGAVILLE
It’s my pleasure, thank you.

SHANE HUNTINGTON
Dr Roger Dargaville, energy and climate change analyst from the school of earth sciences at the University of Melbourne. Relevant links, a full transcript and more info on this episode can be found at our website at upclose.unimelb.edu.au. Up Close is brought you by Marketing & Communications of the University of Melbourne Australia. This interview was recorded on 4 August 2010. Our producers for this episode were Kelvin Param and Eric van Bemmel. Audio engineering by Gavin Nebauer. Background research for this episode conducted by Dr Christine Bailey. Up Close is created by Eric van Bemmel and Kelvin Param.
I’m Shane Huntington, until next time, goodbye.

VOICEOVER
You've been listening to Up Close.  For more information visit upclose.unimelb.edu.au. Copyright 2010, the University of Melbourne.


show transcript | print transcript | download pdf