Wind power can easily replace 40% of Coal, Oil, Gas and Nuclear for electricity generation in the UK. Wind expert David Milborrow shares his 30 years of experience in grid integration and penetration of wind resources and all variability associated with electricity grid supply and demand. David Millborrow in conversation with Matthew Wright of Beyond Zero Emissions.
Milborrow is an expert in large scale wind power integration and penetration on modern electricity grids. He started his work with the Central Electricity Generating Board (CEGB) in the United Kingdom where he worked on aerodynamic research, policy development and plans for UK’s first wind farm. His first foray into variability issues research was in 1988.
He is a mechanical engineer by training and has worked with the CEGB's research laboratory on advanced types of cooling towers. He then turned his thinking to fans for cooling towers and became interested in prospects for wind power in the CEGB when wave energy was the favoured renewable. He had a large wind tunnel to experiment with and was able to simulate large arrays of wind turbines and work out realistic spacing etc.
After seven years research he got involved in planning, technological and economic issues associated with real wind farms and was involved with the construction of the first commercial wind farms in the UK.
CEGB’s last transaction, before privatization, was an equity contribution to that first wind farm.
Milborrow became an independent consultant in 1992.
Under the privatization model, research is still done but is mostly funded on a collaborative basis and contracted out to independent operators.
What level of penetration has wind power achieved?
Milborrow referred to the myth about wind power that it fails when the wind stops blowing. This is not a problem. He indicated that electrical systems are used to coping with variable demand and uncertainty. Wind adds only a little more uncertainty. He offered a useful analogy to help us appreciate the additional uncertainty wind produces:
A pebble thrown into a still pond compared to a pebble thrown into a busy pool of swimmers where the perturbations are hardly noticed. It is the latter perturbation that wind produces.
Similarly the extra cost of dealing with large quantities of wind energy (up to about 40% of all energy provision) is quite modest. The extra cost to the consumer is approximately $11-12 (Aust) per MW hour ie. an additional 5-6% on domestic electricity bills.
There are periods when the available wind exceeds system demand and thus a modest amount of wind needs to be curtailed. Milborrow has been asked by WWF and NGOs to look into possible mitigation measures for “mopping up" surplus wind and reducing costs. He has identified ten mitigation measures including the take-up of electric cars (absorb surplus overnight) and better wind predictability (to reduce cost). (Wind is not, in fact, totally unpredictable. It varies in a manner that can be quantified).
The power of wind turbines can be easily controlled or “turned down”, for example, by pitching the blades.
Variability compared to intermittency
Nuclear, coal and gas-fired stations and conventional thermal generation is intermittent. The output from a nuclear power station can disappear instantaneously. Wind output, however, “swings around in a lazy manner” and the characteristics of the power swings can be quantified to a high level of precision. Consequently, wind power is variable.
Shifting fuel sources and the difference in delivery between base amount of power and peak
If for example, electric cars are charged at night, the load profile is smoother so the electricity system operates more efficiently. Additionally, because electric cars are more fuel efficient, there is also an overall win in fuel economy.
Smart meters could, if properly configured, attenuate the load in return for lower tariffs. This would aid the assimilation of renewables and improve the overall efficiency of the electricity network. Work has been done into the scope for means to reduce the cost of the spinning reserve and thus reduce the cost of variable renewables.
Deferring charging of, say, electric vehicles and space heating and cooling can make use of surplus power at night. Also, moving from gas to electric heating (and using electricity from renewable sources) and allowing for flexibility in terms of when the charge is taken, saves valuable fossil fuel sources.
We can achieve in the order of 40% penetration of wind on our grids!
In conclusion, Milborrow noted that in Australia we have an advantage over the United Kingdom in that our best wind resources line up with our population!
Originally published 2009-10-09 13:18:20 +1000
Matthew Wright: On today's program we'll be talking to David Milborrow, an expert in large scale wind power integration and penetration on modern electricity grids. David has had more than 30 years experience in renewable energy and that's longer than I've been on the planet. David started out at the central electricity generating board, which managed the UK's electricity sector post World War II until privatisation where he worked on aerodynamic research, policy development and plans for the UK's first wind farms. His first study was into variability issues in 1988 and since then David has produced a number of studies on this issue. Most recently, one for Greenpeace and of course he's done some for the National grid. So you're joining us on the line late in the evening UK time is David. Good morning, well I should say, good evening, David.
David Milburow: Good morning!
Matthew Wright: And thanks very much for staying up for us, what time have you got there.
David Milborrow: It's not too late, just quarter and a half past ten.
Matthew Wright: Ok, alright so just before bed, a great way to spend your evening and to have a good night's sleep is to talk to us. And I guess to introduce yourself a bit more in your words, so can you tell us how you got involved in the electricity sector, and in particular, having such knowledge on wind power and variability issues?
David Milborrow: Well, I'm a mechanical engineer by training. And in the central electricity generating board, I was at their research laboratories initially working on research into advanced types of cooling tower. I then turned my area of thinking about things for cooling tower literally sort of inside out and got interested in the prospects for wind tower in the CEGB.
Now at that time, wave energy was the favored renewable but suffice to say that there was a small wind power program and we identified two key issues: one of which was the way in which wind turbines might be brought together in the so-called wind farm and to what extent they would interfere with eachother, the power losses and so on. And a related issue, they were all wind powered resourced. Fortunately we had at our disposal a very large wind tunnel and so we were able to simulate larger rays of wind turbine in that wind tunnel, work out realistics, spacings and the factors which influenced the energy losses and so on.
And I went on after about 6-7 years of research up to the head quarters of the organisation and that's when I got involved in the fanning issues and indeed the technical and economical issues associated with real wind farms. And indeed I was involved in the work that led to the construction of the first commercial wind farm in the UK. The last transaction of the central electricity Board, the Central Electricity Generating Board, was an equity contribution to that first wind farm. And after privatisation the industry was reorganised and the activities were completely reorganised and I became an independent consultant in 1992. Since when I, to a certain extent, traded all my technical knowledge about aerodynamic and technical issues and indeed done a lot of the work that you mind in your introduction. So that's probably enough of an introduction.
Matthew Wright: And interestingly that Central Electricity Generating Board had science laboratories that, I'm not sure where they were, but that's something that really hasn't carried forward in many countries that have gone for the privatisation model.
David Milborrow: Yes, that's a perfect exact point. They still do do research, but it's mostly funded often on a collaborative basis and contracted out to independent operators.
Matthew Wright: Now, in terms of your research, we became aware of you when we were looking into large scale penetration of wind into modern electricity grids and there's a lot of, we found a lot of myths out there, there's a lot of opinions flying around, a lot of it coming from a fair amount of ignorance and also an emotional basis and so I guess what we wanted to get down to is what was the value of wind as a renewable and how does that integrate, what sort of levels of penetration can that get. So can you give us a bit of a background on wind power and how it meets the demands of modern electricity consumers and the way grids work?
David Milborrow: Yes, as you say there are a lot of myths out there. "What happens when the wind stops blowing?" is the favorite saloon-bar pundent's question. The implication being that all sorts of dire things are likely to happen. But in a nutshell it's simply not true because electricity systems are use to dealing with the variable demand imposed on their systems by us consumers who put all sorts of strange habits like getting up in the morning and going to work and in most cases that imposes quite a severe strain on the network, the demand in Great Britain, for example, increases by several thousand megawatts between 7am in the morning and 10am. And although our habits are predictable to a certain extent, they're not totally predictable. Our television programs, for example, may be more or less popular than the control engineers predict and what we do during commercial breaks in television programs or at the end of television programs, we all go and make a cup of tea, and the demand shoots up. So electricity systems are quite use to coping both with quite rapid changes in demand and with uncertainty.
And what matters when it comes to assimilating wind on the system is the additional uncertainty. And that is in fact quite small. Even when you get to levels of 10, 15% on an energy basis. An analogy that is perhaps quite a good way of explaining it is if you got a pebble and you throw it into a calm swimming pool, the ripples will spread out and it will all be totally visible. If you throw that pebble into a swimming pool full of energetic swimmers, you will hardly notice the perturbations from the pebble. Now I'm sure you can see where I'm heading. The large swimming pool is a typical electricity system with all the perturbations caused by the consumers; the pebble is the additional perturbations caused by the wind. And it's the additional perturbations that matter. And that's really the nub of the whole issue.
Matthew Wright: Now there was this report that the UK were going to be having an all-energy target as part of some European Union targets of, and it was going to include space heating, water heating and transport fuels as well as electricity. And they were going to go for a serious penetration of renewable energies across all of those different energy types. And it was said that that possibly more of that move to renewable energy would happen in the electricity sector than say in transport fuels and in space heating. So they talked about...
David Milborrow: That's right...
Matthew Wright: ...They talked about that wind may perhaps hit 32% and you basically had done some reporting on the fact that 30% and 40% were plausible and that didn't cause too much trouble. Can you tell us a little about that?
David Milborrow: Yes, that's absolutely right. I'd drawn on data from a number of system operators, that's the people who control the networks literally around the world, and in particular of course our own national grid. And in a nutshell, what you find is because you are looking, as I said earlier, at the additional perturbations, rather than the total perturbations of the wind, you find that the extra costs of dealing with large quantities of wind energy, up to say about 40%, are in fact quite modest. And we're talking about an extra cost to the consumer not exceeding, well I can't be totally dogmatic in these days of very uncertain fuel prices but I suggested that the total cost of variability at the 40% level won't be above about 7 pounds a megawatt-hour, now that's about 11 or $12 AUD a MW-hr. In this country, that represents an additional 5-6% on domestic electricity bills, no more.
Matthew Wright: That's really amazing.
David Milborrow: So that's the bottom line.
Matthew Wright: Yea, that's fantastic news. Now you did mention a small amount of curtailment at 40% and I'm wondering if you could explain what curtailment is to listeners...
David Milborrow: Yes, once you get an amount of wind energy on the system, equivalent to a penetration of about 30%, because you need, again let's use the figures I'm familiar with for Britain, 20% penetration would represent about 26,000 megawatts of wind. Our minimum demand is about 21-22 gigawatts. So there will be periods when the available wind will exceed the system demand. So it's under those circumstances that perhaps a modest amount of wind will need to be curtailed. And indeed I've estimated the levels in the report I did for the World Wildlife Fund, but and this is a big but, the WWF and the other NGOs asked me to look at possible mitigation measures that would deal with those, shall we say, mopping up the surplus wind and reducing additional cost. And I actually identified 10 mitigation levels. I suppose the most popular one in terms of the discussion that are going around the world at the moment is the hook up of electric cars, they would be able to absorb surplus wind during the night. Let me just mention one other one, that's better wind predictability. There's an awful lot of work going on around the world about predicting wind variation better. And the anticipation is that this also will reduce the additional cost. I would add possibly at this point, something that I should have mentioned earlier, that wind is not totally unpredictable as perhaps some of the pundents would have us believe. It does vary in a manner that can be quantified and that's the sort of information that enables the grid operators to come up with those estimates of the additional costs.
Matthew Wright: Just before I ask the next question. Curtailment...so how do you actually curtail the wind from wind turbines.
David Milborrow: That's quite straight forward. There would have to be some mechanism by which the system operators would instruct the wind farm developers, please turn down the power. And you do that literally by pitching the blades, much as happens in an aircraft propellor, by unloading the blades, reducing the angle of attack, you can in fact reduce the power and wind turbine power outputs are in fact, quite well controllable.
Matthew Wright: So that can be done very smoothly, obviously.
David Milborrow: Oh yes, yes. So the jargon is "set point". If you've got a wind turbine with a rated output of say 3 megawatts and the instruction from the controllers is such that they only want 2/3 of the output then the set point will be set down to 2 megawatts, the pitch of the blade will be changed and each wind turbine would then generate a level of power not exceeding 2 megawatts. So it's nothing untoward as far as wind turbine technology is concerned.
Matthew Wright: Now you did talk about the meteorlogical work going into predicting and reducing some of that variability and you did use the term variability. Can you tell us the difference between variability and intermittency, because intermittency is obviously often thrown around in regards to wind power.
David Milborrow: Yes, that's a good point. I got told off by the American Wind Energy Association some years back for using the term intermittency in the context of wind. And I've gone around ever since preaching the gospel of variability. Nuclear, coal and gas-fired power stations are intermittent, by which I mean, and that is literally the dictionary definition of the word, every now and again, usually due to instrumentation faults or something like that, the output from a nuclear power station, say, will instantaneously trip (to use the jargon) offline and in this country the biggest nuclear power station on the system is 1200 MW so that can disappear literally instantaneously. It doesn't happen that often, but bearing in mind that we've got a large number of generating units on our system. And at one time gas turbine units were tripping out about once a month. This sort of thing is not unusual.
Wind power, on the other hand, simply isn't like that. Particularly as the grid controllers will be looking at the aggregated output from all the wind turbines in the country. And if you look at the data, and there are not data we can look at from western Denmark, the output from wind swings around in a fairly lazy manner. And as I indicated earlier, the characteristics of those power swings can in fact be quantified to quite a high level of precision. So wind power is variable. Conventional thermal generation is intermittent.
Matthew Wright: For listeners, you're with the Beyond Zero program on 3CR radio, 855 on your AM dial. My name is Matthew and today we're speaking to David Milborrow and he's an expert in large scale wind power integration and penetration on modern electricity grids.
Now David, you were just talking about how smooth the total of all the wind turbines output is when it's added together vs. other big generators and I understand that, we've certainly got them here in Australia, there's some big connections on the grid. Some of the nuclear plants have very big turbines and they can trip off the entire turbine in one instant and also the connection to France. Can you tell us a bit about just how big they are and how that compares to wind?
David Milborrow: Yes, that's right. I think I've given you the key facts. The British system is probably fairly typical of other electricity systems around the world. Clearly you couldn't have a 1200 MW nuclear power station in Ireland, for example, the system is that much smaller. So most electricity systems, the maximum unit size, there isn't a fixed number, but it's probably no more than about 1 or 2% of the maximum demand. The system simply couldn't stand the strain. So that is one of the key factors, not having too big a unit on the off system. Have I answered the question?
Matthew Wright: Yes, I think so. In that case, we were talking about 40% penetration of wind as a possibility in the UK grid and often all the the studies we hear, size it from very small geographical areas, in comparison to Australia. We've got our own continent here so we talk about studies from Ireland, from the UK, from west Denmark or Denmark whole. How would geographical diversity affect the output of wind and reduce that curtailment amount? Given that, for instance, our electricity grid runs thousands of kilometers up and down the eastern seaboard.
David Milborrow: Yes, I understand and that's a good point. The fact that Australia is that much bigger than Great Britain which in turn is that much bigger than western Denmark means that the power swings in Australia would probably be somewhat less than they are in Great Britain. Once you get to a certain size, the indications are that the additional benefits are small. The modelling work that's been done in this country suggests that the additional smoothing in Britain compared to Denmark, for example, is relatively small, but it's still worth having. And as you say, in somewhere like Australia which is even bigger, then the power swings are going to be smoothed even more. So big is beautiful in this context.
Matthew Wright: Another thing about electricity grids, that the listeners may or may not be aware of, is that the ratio of what you tend to have as your general minimum output vs. where the peaks go and in many countries it's sort of like your extreme is twice the amount of power that you need to deliver during the peak times than offpeak times. And what I was wondering is if you could tell us a bit about that but also if you start switching fuel sources, so fuel sources that use to be say liquid fuels or coal or things like that for space heating and cooling and water heating and transport fuels where we're talking about electrification of motor vehicles and rail electrification, if they start to move onto the electricity grid, do they reduce the difference between the base amount of power delivered and the peaks?
David Milborrow: Yes, the thinking is that they would. Because let's take electric cars, it would be ideal to charge them during the night. Now let's answer the other part of your question. The demand in Great Britain on a typical winter's day falls from a peak of about 59,000 MW at 5:30 in the evening to something round about 42,000 MW by midnight or thereabouts. So all that plant has to be turned down and if it could be used to charge electric cars, not only would it smooth the load profile as it's called, which would mean that the electricity system would operate more efficiently, but also, because electric cars are more efficient, more fuel efficient than cars propelled by conventional petrol or diesel, you've got a winner as far as overall fuel economy and fuel efficiency is concerned. So there is a lot of interest in electric cars, and there's lots of interest in other so-called demand-side management mechanisms as well. Would I be prepared to have my evening meal at about 11:00 at night instead of 6:00? Well I don't think so, but there probably is a certain amount of discretionary load, discretionary demand, strictly speaking in most households that the so-called "smart meters" that are much talked about at the moment could, if they were correctly configured and the consumer was happy, attenuate that load perhaps in return for a lower tariff. So there are lots of ideas like these around, that could not only aid the assimilation of renewables, but also improve the overall efficiency of the electricity networks.
Matthew Wright: So also the time that we're investing money in upgrading and moving toward renewable energy, if we also implement smart meters, does that also have an effect on that sort of spinning reserve we have, that sort of back up power that just in case the wind comes off or in case that giant nuclear plant trips off, that spinning reserve, you don't have to pay for as much as that with smart meters.
David Milborrow: Absolutely, dynamic demand is, forgive me for introducing the jargon, it's very simple to explain, is a mechanism that would probably give you cheaper spinning reserve, effectively, by simply having a small device fitted in the plug of your refrigerator and when the system frequency fell, it would inhibit the fridge from coming on. When the system frequency was too high, the fridge would cut in and you probably would need a slightly wider deadband on your fridge temperature but the science has been done and there is scope for ideas like this, reducing the cost of the spinning reserve and so reducing the extra costs of variable renewables.
Matthew Wright: And deferring the charge of electric vehicles would act in the same way?
David Milborrow: Oh yes, yes, quite. Electric vehicles is probably a little bit simpler in that most electric vehicles would be sitting in the garage at night and so could make use of the power that tends to be suplus in any network. But again there would probably be a proportion plugged in during the day and you would need reversible connections that they could be used in much the same way. There's even talk of having charging points at supermarkets and so on, but I personally think that's probably a little bit further into the future.
Matthew Wright: Yep and another area that I read about that you may or may not be aware, and I don't know if they're talking about on a per domestic dwelling basis, but space heating or space cooling and being able to somewhow defer that.
David Milborrow: Yes, you're right. And indeed the move from gas to electric heating is again discussed in the context of the sort of measures we're talking about, particularly as off-beat storage heaters have advanced considerably since they were first introduced some 50 or 60 years ago and again they are very flexible as far as when they take in their charge. I don't think you could send the charge back the other way but certainly in terms of absorbing the electricity, again, that is an idea that is being discussed. And if you use electricity instead of gas, and that electricity comes from renewable sources, then you're saving valuable fossil fuel sources.
Matthew Wright: Look we've basically run out of time and it's been fantastic speaking to you, David. Thanks very much for coming on and also very intersesting to hear that we can achieve on the order of 40% of wind on our grids and I'm sure you're aware that Australia has pretty decent wind resource. So thanks for joining us and we hope to get you on again.
David Milborrow: Ok, thank you. Yes I did take the precaution of taking a look at your wind resources and they're good. Oh and finally I've just got a second to say that you've got an advantage to us. Your best wind resources line up with your population. In a lot of countries that's not the case.
Matthew Wright: Yes, that's certainly a helpful situation. And of course we've got solar thermal resource with storage which in the UK they've been looking at from Africa but we can get it sort of up the road.
David Milborrow: Yes, that's true.
Matthew Wright: Alright thanks David.
David Milborrow: Thank you.
Transcript by Nicole Caruso