Home > Environment > Solar PV (now with grid backup)

Solar PV (now with grid backup)

August 13th, 2015

Following my previous post there was some discussion about the need for grid backup of solar PV to deal with extended periods of overcast weather. It’s obvious that storage will help with this to some extent, since batteries can store electricity from the grid as well as from distributed solar. I thought I would try to put some numbers on this (slightly changed from last time to simplify the numbers).

I’ll focus on 1 kW of solar PV generating an average of 4.8 kWh per day, with (as before) 2 kWh of storage. If there is zero solar generation and no demand management, the entire 4.8 kWh per day must be drawn from the grid. In the absence of storage, we might suppose that 1kW of backup capacity is needed to match the peak output of the solar PV system. But with storage, all that is needed is enough to supply 4.8 kWh over the course of a 24-hour day, that is, 0.2 kW.

The optimal backup choice is a fully dispatchable technology such as hydro or gas. Hydro resources are pretty much fixed, so I’ll focus on gas. According to the US Energy Information Agency, the capital cost of gas-fired power plants is around $1000/kw so our grid backup will have a capital of cost only $200 for each kW of distributed solar. There’s also the need to take account of fuel and distribution costs. Fuel costs will be low since the system is only used as backup, while distribution costs will be around 20 per cent of what would be need if peak loads were to be met by centralised generation.

To sum up, if battery storage becomes available at a sufficiently low price, there’s no obvious problem with a system in which over 80 per cent of capacity, and an even larger proportion of generation is distributed solar PV.

Categories: Environment Tags:
  1. August 13th, 2015 at 22:52 | #1

    Yes, the role of batteries in reducing the peak grid generation capacity is interesting.

  2. Ikonoclast
    August 13th, 2015 at 23:36 | #2

    “Following my previous post there was some discussion about the need for grid backup of solar PV to deal with extended periods of overcast weather.”

    At a household level, the feasibility of grid disconnection is becoming a walk in the park.

    I believe, after discussions with Ronald Brak, that in Brisbane a house needs no grid connection and could run well with 5.5 kW to 7.5 kW (nominal) of solar PV panels, solar hot water and 2 x 7 kWh Powerwalls. Whether this is currently (no pun intended) advisable financially is still doubtful. Technically, it would work fine and such a 4 person household would almost never run out of power provided they used relatively little air-con.

    In turn, air-con is no big deal. You insulate the house, install a thermal ballast system indoors (it could be rainwater tanks built into the core of the house), have extra solar PV to run the aircon by day only and the house stays cool at night with thermal ballast.

  3. August 14th, 2015 at 03:02 | #3

    “But with storage, all that is needed is enough to supply 4.8 kWh over the course of a 24-hour day, that is, 0.2 kW.” I don’t get this at all. The thought experiment suggests that demand is flat over the day. The entire problem results from the fact that it is not.

    Suppose (off-the-cuff but more realistically) that half the daily demand of a representative household comes in the evening, or 2.4 kwh. Absent domestic storage, all of this has to be met from the grid. In the limiting case when the only generation is solar, the grid would need 2.4 kwh of storage. Economically storage on this scale is the last resort, and the number comes down as you call on wind and dispatchables. Where grid storage wins over domestic is that the grid can average over many houses, and use the most effective technology, which isn’t likely to be Powerwall.

    I really do wish somebody would run scenarios for 90% or 95% renewable generation. As JQ points out, the cheap backup source is gas. Squeezing the last few percent of fossil electricity out will be expensive: massive pumped storage is the world’s driest continent, huge CSP plants, or jam-tomorrow EGS geothermal. It will probably deliver more carbon reductions for the buck to spend that money on electric transportation or biochar.

    My suggestion is therefore to campaign now for 95%, which is certainly affordable. In 20 years, something will turn up to get us to 100%.

  4. Will Boisvert
    August 14th, 2015 at 05:56 | #4

    I don’t get your analysis here either, John.

    Look, when an overcast day dawns, the batteries are dead—they have been drained to supply power over the previous night. Batteries are dead, PV panels are dead per your stipulation. That means the house is running entirely on instantaneous grid power—nothing else. So the gas “backup” needs to generate the household’s peak power consumption all by itself. And as James Wimberley notes, peak loads will be much higher than the average load you have budgeted here. (It’s a mistake to analyze issues of power reliability in a framework of average energy output.)

    The right way to analyze this scenario is to realize that, once an overcast day dawns, it’s as if the battery storage does not exist. In order for the battery storage to play a role, you would have to draw yet more grid power into topping up the battery that’s already been drained (which would also raise instantaneous load above what you’ve budgeted here). Given round-trip losses, that makes no sense, especially if you envision the supply coming from fast-ramping gas plants that don’t overproduce during low demand periods.

    There’s a simple rule of thumb for sizing the dispatchible “backup” generation that’s needed for wind and solar: you have to have enough to meet peak demand assuming that virtually all the intermittent generation goes away, which it will from time to time. Battery storage in any feasible amount doesn’t change that equation. So the current establishment of dispatchible generation will not shrink even with a large overbuild of intermittent generation and storage, and the dispatchible grid’s capital costs, maintenance costs and distribution overhead will persist along with substantial carbon emissions.

    –“distribution costs will be around 20 per cent of what would be need if peak loads were to be met by centralised generation.”

    I don’t understand. Electricity distribution costs are mainly the overhead cost of stringing power lines to houses (and associated gadgetry). Since every house under your scenario still sometimes needs power lines to connect it to gas generators for backup power, why will the distribution costs go down at all?

  5. John Quiggin
    August 14th, 2015 at 08:59 | #5

    In order for the battery storage to play a role, you would have to draw yet more grid power into topping up the battery that’s already been drained

    Exactly, and that’s what the calculation assumes. The grid power supplies, over 24 hours, all the household consumption needs. The battery solves the peak load problem. In fact, the analysis works without any distributed generation at all. But the economics of storage are a lot better with solar PV.

    I agree that I should correct for round-trip costs. On the other hand, the assumptions are very conservative in other respects – zero solar generation in all households connected to a given grid, no demand management, no interruptible industrial customers etc.

    On distribution costs, Australian experience has shown that they are largely determined by peak load, not, as previously assumed by the fixed cost of stringing lines to houses.

  6. Ikonoclast
    August 14th, 2015 at 09:33 | #6

    @John Quiggin

    “On the other hand, the assumptions are very conservative in other respects – zero solar generation in all households connected to a given grid…” – J.Q.

    I agree. In daytime, in a sub-tropical city like Brisbane, absolute or very near zero power generation from a solar panel array of even a single home is very rare. I have seen my 5.5 kW (nominal) array pushing out 1.0 kW with light grey total cloud cover and drizzle. Many of our stormy days might feature an hour or two of deep gloom but then the rest of the day features bright sun and clouds for a lensing effect often generating in the range 4.2 kW to 5.2 kW on my system.

    A week of rain will feature various lighting conditions over each day in summer in Brisbane even if cloud cover remains total. There is considerable difference between light grey cover and dark grey cover. The assumption that rainy day light conditions equal a black night for solar panels is not warranted. Eight hours of 0.5 kW average production from a 5.5 kW nominal system is quite possible on a very gloomy, rainy day in summer. Since that’s 4 kWh and more than half a charge for 7 kWh Powerwall while someone is at work maybe that is not to be dismissed.

    Anti-PV campaigners are basically claiming an average rainy day is like a black night. This is simply not true in sub-tropical and tropical areas. 33% of the world’s population lives in the tropics.IIRC at least another 33% live in the sub-tropics. Thus 2/3 of the world’s population live where sunshine is very plentiful and where even rainy days (partial or complete) produce significant solar power. Not everyone lives in Buffalo or Duluth. Cool temperate and frigid zone dwellers are not all that representative for solar PV demonstration purposes when it comes right down to it.

  7. Collin Street
    August 14th, 2015 at 11:39 | #7

    I don’t understand.

    If someone is telling you something you don’t understand the go-to assumption should be that you are in error.

  8. August 14th, 2015 at 11:51 | #8

    Ikonoclast, I’ll let you know I don’t want you to go off grid. I’d much prefer it if you stayed on-grid and exported your surplus solar electricity and contibuted to the reduction of fossil fuel use for grid power.

    It may seem that electricity distributors are trying to drive you off the grid with the recent increase in daily supply charges in Queensland, but as soon as people start dropping off-grid they will probably remove all incentive to go off grid on account of how they would rather have some of your money than none of your money. Of course, for this to happen, some people are going to have to go off-grid first. But there seems to be plenty of people who say they are willing to do this around. (This is of course assuming that Queenslanders are not simply declared to be the property of electricity distributers and required to pay or do whatever they desire.)

    As soon as people start The reason why we have supply charges is to discourage energy efficiency. If Australia was serious about cutting greenhouse gas emissions there would be no supply charges and we’d just pay for the electricity we use.

  9. Will Boisvert
    August 14th, 2015 at 12:19 | #9

    @ John Quiggin,

    Yes, you’re right, batteries tied to the grid should in theory replace some gas peaking capacity.

    But batteries don’t really favor solar, and especially distributed solar, because

    1) Utility-scale storage is much cheaper than distributed household storage because of economies of scale and expertise. For example, utility-scale Tesla batteries cost about 40 percent less per kwh than the household version, not counting scale economies in installation, maintenance etc.

    2) Storage and arbitraging of electricity is much more valuable for baseload generators than for solar. Solar stores during periods of high daytime demand (provided the sun is out) to use during periods of somewhat higher evening peak demand (or very low nighttime demand.) A baseload generator can reliably overproduce and store during low-demand nightime hours and sell into peak demand at a much larger profit.

    So in a market for stored electricity, stored baseload power will undersell stored distributed solar.

  10. August 14th, 2015 at 12:24 | #10

    Ikonoclast, sorry, ignore the word salad final paragraph of my last comment. I forgot to edit that out.

  11. John Quiggin
    August 14th, 2015 at 13:00 | #11

    @James Wimberley

    I assumed 40 per cent or just under 2kWh of demand to be met in the evening. So, a 2kWh battery, fully recharged by evening does the trick. The battery runs down to zero by late evening when demand falls below the average level for the whole day, which is 0.2kW, then starts recharging again.

  12. Will Boisvert
    August 14th, 2015 at 14:43 | #12

    @ John Quiggin 11

    –“So, a 2kWh battery, fully recharged by evening does the trick. The battery runs down to zero by late evening when demand falls below the average level for the whole day, which is 0.2kW, then starts recharging again.”

    Hmm.

    So if the entire grid power feed of 0.2 kw is diverted to recharging the 2 kwh battery, with nothing left over for any other load, it will take 10 hours to recharge the battery. If there’s residual household load of, say, 0.1 kw–maybe a fan or two?—then it would take 20 hours to recharge the battery. (What if there’s a morning rush with high loads?) If there’s a persistent household load of 0.15 kw it will take 40 hours to recharge the battery, too late for the evening discharge, and into a permanent storage deficit.

    Maybe it’ll work out, but I’m kind of skeptical. I’m not sure you’re being realistic about power demands in this scenario. It doesn’t feel like a “conservative” modeling exercise.

  13. John Quiggin
    August 14th, 2015 at 16:08 | #13

    If there’s a persistent household load of 0.15kW and a peak on top of that, then you will use more than 5kWh per day, say 7.5kWh.

    So, the solar PV system needs to be 1.5 kWh, the battery 3 and the grid backup 0.3. It’s just a question of scale.

  14. Ikonoclast
    August 14th, 2015 at 16:39 | #14

    @Ronald Brak

    I will very likely do my sums when my private power pole needs replacing. My decision will depend on that cost, on the cost of Powerwalls or equivalent and on the charges and payments I get from being connected to the grid. It is really up to the utilities whether they drive me off-grid. I suspect if their business model was sensible and equitable to shareholders and customers alike (who are all stakeholders after all) they would be able to price to keep me on-grid. But let’s wait and see.

  15. August 14th, 2015 at 22:33 | #15

    @Collin Street
    “I don’t understand”: the Oxford convention is considerably less deferential.

  16. August 14th, 2015 at 22:55 | #16

    @Will Boisvert
    I confess I incline to Will’s side on this. The electricity supply system has a lot of moving parts, and very simplified thought experiments can only tell you so much. As a rule, large-scale scenarios for 100% renewables like Jacobson’s only see a modest role for storage, on cost grounds. If it gets cheap, then you have more options. Where Australia and Germany are heading is an economically inefficient overbuild of domestic behind-the-meter storage, because the incentives are all wrong. Will is right that utility storage should always be cheaper from a system POV.

    One of the Fraunhofer institutes built a 100% renewables scenario for Germany. IIRC their worst planning case was 2 weeks in November with negligible wind and sun. Their deus ex machina was P2G. True that nowhere in Australia ever gets as little sun as Hamburg on a grey winter day.

  17. BilB
    August 15th, 2015 at 04:01 | #17

    Will Boisvert,

    Your idea that hgid storage is cheaper than Powerwall storage is not correct by my calculation. An information source:

    http://cleantechnica.com/2015/05/09/tesla-powerwall-
    powerblocks-per-kwh-lifetime-prices-vs-aquion-energy-eos-energy-imergy/

    The return on battery storage is the retail price of electricity minus the price paid by the grid supplier all times the energy life of the powerwall and minus the purchase cost of the powerwall. The other important cost is that of the daily charge or line connection charge.

    In the case of my system I pay 26 cents yo buy a unit from the grid. I get a credit of 6 cents for for each unit I “park” in the grid. Where I am storing 10 units per day plus buying another 10 to top up my Winter consumption andy daily connection fee is $1.2 each unit I buy costs me 31cents and each unit I store costs me 25 cents. A 10 Kwhr Powerwall has a primary life (to 80%) of 41,000 units and costs (simplified for ease of calc) $4100, then the cost of Powerwall storage is 10 cents per unit against the grid’s 25 cents
    Game over grid.

    Powerwalls don’t just drop dead at the the 80% stage (considered to be their commercial life) at that point they are still holding 8Kw hours which beside its new replacement gives a household storage of 18 Kwhrs thereby reducing the cost of storage to 6 cents per unit.

  18. Ikonoclast
    August 15th, 2015 at 06:47 | #18

    In theory bulk storage of almost anything should be cheaper than micro-storage due to economies of scale. If electricity consumers find it cheaper to store energy in home battery systems, then I would draw the conclusion that the utility is over-charging for power and power storage and under-paying for power fed back into the grid. Of course, this is what private monopolies do. They overcharge and gouge a captive market.

    The value of distributed storage and distributed generation might consist of a number of factors;

    (1) It creates some sort of competitive pressure on the private monopoly utility perhaps forcing it to drop charges a bit and raise payments a bit.

    (2) The grid can be made more robust and balanced by distributed storage and distributed generation.

    (3) Long distance transmission costs might be reduced by more local and regional generation.

  19. Will Boisvert
    August 15th, 2015 at 09:28 | #19

    BilB, you are right, in places with extemely high retail prices, like Hawaii and Australia, home-stored electricity might underbid grid-stored electricity. Most places aren’t like that.

  20. Ikonoclast
    August 15th, 2015 at 09:41 | #20

    Notwithstanding what I said above, utilities too need to be generating renewable energy and not energy from dangerous, biosphere-wrecking fossil and fission fuels.

  21. Will Boisvert
    August 15th, 2015 at 10:56 | #21

    JQ @ 13, 11

    –“If there’s a persistent household load of 0.15kW and a peak on top of that, then you will use more than 5kWh per day, say 7.5kWh.”

    Huh? If the house uses 0.15 kw for 23 hours that still leaves room for a peak load of 1.35 kw for an hour while staying under your 4.8 kwh daily limit. If that peak comes, say, 7 hours after the battery drained to zero per your stipulation—Early morning rush? Top up the EV before a trip?—then the grid feed would have to supply 1 kw instantaneous power to meet it. So under plausible conditions that comply with your average usage constraints, gas backup capacity five times larger than the 0.2 kw you have budgeted could be needed (and perhaps more).

    I agree with your underlying point that home solar battery storage can be reconceived as a kind of replacement for gas peaking power (although I think that battery-displacing-gas only makes practical sense as a utility-scale complement to (preferably low-carbon) baseload generation). But your model here doesn’t really show it, probably because 1) your method of deriving peak capacity requirements from average daily usage is unsound, and 2) there are technical issues, like charging times and the size of the surplus power margin, that seriously effect the issue but haven’t been carefully considered.

  22. Collin Street
    August 15th, 2015 at 11:27 | #22

    In theory bulk storage of almost anything should be cheaper than micro-storage due to economies of scale.

    No, not really. This is the problem that the Space Shuttle program ran into: most of the cost was in engineering design, not materials, and disposable rockets could be mass-produced. [and disposable rockets could use order-of-magnitude cheaper materials because the task and the loads were much, much better defined and understood, but that’s a different story.]

    Neither batteries nor solar panels have significant economies of scale. There’s no non-linear effects like you get with a thermal plant, where bigger is definitely better up to the limits of our technology: two solar panels produce twice as much electricity as one and has twice the demand for management equipment. Same with batteries. Concentrated utility-level plants have to be custom-designed, custom-produced, and the bigger they are the more work required: household distributed plants can be mass-produced by the literal millions and barely need to be designed at all, and even industrial plants are fairly straightforward. Or small effects like land usage: you can use waste space in domestic housing for a distributed system, but utility plants require acquisition of dedicated land.

    I mean, no guarantees, but on the face of it it certainly wouldn’t be absurd if the distributed net were cheaper, and what evidence we have suggests that it in fact is.

  23. BilB
    August 15th, 2015 at 12:04 | #23

    Will Boisvert,

    Would you care to prove your “most places aren’t like that” claim?

    I’m pretty sure that the above is true for all but a bunch of less populous US midland states

    http://www.eia.gov/electricity/monthly/epm_table_grapher.cfm?t=epmt_5_6_a

    Look at the residential rates.

    In the UK the rate is 11 pence which is 23 cents Australian plus the connection fee at 50 cents per day. I think that when you do the conversion to Aus dollars you will find that what I have said holds true most everywhere.

    So the argument that grid storage is cheaper than PV solar fed Powerwall battery storage is debunked.

  24. Will Boisvert
    August 15th, 2015 at 12:56 | #24

    BilB, I confess I don’t follow your accounting. By the way, 10 kwh powerwalls cost about $7,000 installed, can only cycle weekly, are warranteed for 10 years –Musk claims 15 years of life–so a low-ball figure for the cost of storage might be 50 cents per kwh. It’s for backup during a blackout, not for daily storage. The 7 kwh Powerwall has a much cheaper installed storage cost, about 14 cents per kwh, which alone is higher than average US retail electricity prices.

    How do the economics of Tesla storage work out for grid vs. Household in the US? Well, the 100kw utility-scale Powerpack sells for about 40 percent less per kwh than the household 7 kwh version; assuming similar scale economies of installation and maintenance, count on a grid storage cost about 40 percent lower than household storage, or about 5.6 cents per kwh lower. Wholesale grid power during nighttime trough sells for 3-4 cents per kwh, generating costs for distributed solar are likely triple that or more on average. Stored household solar has some advantage in transmission and distribution costs if used at home, none if exported to the grid. But much of the cost of the grid is overhead, those costs aren’t avoided unless household storage users go completely off grid, which is not feasible. So the transmission and distribution cost advantage of home storage over grid storage is much less than meets the eye. Add up all the costs and grid storage comes out much cheaper than home storage. See here for details. http://thebreakthrough.org/index.php/issues/renewables/the-grid-will-not-be-disrupted

    If home storage really is cheaper than grid power then people will flock to it. So far they haven’t.

  25. Will Boisvert
    August 15th, 2015 at 13:08 | #25

    Also Bilb, in the CleanTechnica article you cited on storage costs at 17, did you notice that their estimates of household battery storage costs averaged 24 cents per kwh, (range 10 cents to 50 cents) and that their estimates for grid battery storage costs averaged just 4.8 cents per kwh (range 2 to 9 cents)?

    According to your own source, household battery storage is going on 5 times more expensive than grid battery storage!

  26. Ikonoclast
    August 15th, 2015 at 15:33 | #26

    @Collin Street

    You might be right. That is why I used the “in theory” caveat. Of course, if practice disproves theory then theory is wrong, incomplete or inapplicable to the case in question.

    Putting up solar panels on house roofs might well be cost-efficient because apart from the money used to buy and install the panels there really are no other significant opportunity costs. The land? Already paid for to build the house on. The roof? Already there. The roof surface? Not being used for anything else except maybe rain water collection which the panels do not impede in any case. Roof more likely to collapse or fly off from the weight or windage of solar panels? It seems not. My insurance company thinks not and adds nothing to its premium while still providing cover for the solar panels just like anything else attached to the house.

    With energy storage it might well be the case that certain forms of bulk storage do indeed show economies of scale. Pumped hydro and molten salt heat storage come to mind. The latter is usually combined with CST (Concentrating Solar Thermal) arrays.

  27. rog
    August 15th, 2015 at 18:29 | #27

    However, the economic argument over the pros and cons of domestic solar has been somewhat muted by the offer of free instal plus reduced rates.

    How can they do this?

    http://goo.gl/p06F4D

  28. BilB
    August 15th, 2015 at 19:01 | #28

    Will Boisvert,

    As you say the 10 Kw unit is a different battery chemistry and suitable only for a slow drain service life. That had not come up before. But your installed price is a worst case retail price, and not representative of the market expectation. The article I linked demonstrates 4 pricing options depending on a range of distribution methods. Storage power at 10 cents per unit is entirely practical and probable especially in the “national interest” distribution model, though there are others with long term commercial histories.

    That you cannot understand my simple maths is not surprising considering the general trend of the articles you write which are inflationary and clearly intended to support interests such as the Rockefeller’s (you know the guy who moved heaven, Earth and the Temperance Movement to kill Henry Fords dream of biofuel powered cars in favour of his (Rockefeller’s) fossil fuel). To be more correct I think for you it is not a matter of cannot follow, but definitely do not want to understand as this gives you plausible deniability that there really are far cheaper options than you write about.

    But the fact is that your article, the one you linked to, is clearly written for the tired old Nuclear Lobby and is intended to befuddle those who would not bother checking the facts. Howlers such as this opener

    “On cost, the average residential retail electricity prices in the US are $0.12 per-kWh, while electricity from Tesla’s Powerwall on paired rooftop solar would cost 30 c/kWh or more”

    are what I would call less than honest, especially where you flip flop between grid solar and residential solar arguments. The population unit rate is actually 14.03 cents per unit. That is precisely where Australia’s electricity prices were before our grid operators took advantage of measures intended to promote the reduction of CO2 emissions to rort the public locking in higher prices with an insane transmission infrastructure spending spree that forces wheeling prices from the once 33% to the present 50% of the retail price. Meaning? Australia’s prices are not coming down to where they were.

    I have gone to a fair bit of effort to demonstrate that there are workable complete off grid solutions, and I am in the process of testing those designs in physical hardware. The key differences are the use of PVT panels and backup power generation, neither of which feature in your incomplete ruminations.

    What Elon Musk has done is throw out the challenge to battery manufacturers world wide to match or better his pricing and performance model. He has claimed the high ground and global profile along with naming rights with his bold move. But that is just the kickoff. Already the product specifications have improved, and will certainly continue to do so. Global competition is energetic and will certainly mobilise to take advantage of the immense amount of expectation that Tesla have created.

  29. BilB
    August 15th, 2015 at 19:13 | #29

    What that is about Rog is that solar energy that is produced and used locally does not incur the wheeling 50% of retail cost, at least an electricity distributor it does not. So Origin here, I believe, are able to use their buying power to obtain PV hardware at a below commercial rate and profit from the free (no fuel cost) solar energy while at the same time reducing their exposure to eventual carbon pricing. It is a good deal for the environment WIN, power utility WIN, and consumer win.

  30. August 15th, 2015 at 21:50 | #30

    John, your figure for the capital costs of gas plants appears to be for combined cycle ones, but since in reality they will be rarely used, they are more likely to be cheaper open cycle gas turbines. These can cost as little as $300 Australian a kilowatt, although more efficient and reliable models can easily cost more than twice that. For a back of the envelope calculation, a figure of $500 Australian a kilowatt seem suitably conservative to me, although it could be refined further if one wanted to go into more detail. This halves the capital cost of gas generation in your estimate.

    At the current international gas price of around $3.88 Australian it would cost about 4.2 cents in fuel to generate a kilowatt-hour of electricity from gas with an open cycle turbine. The exact amount will depend on its efficiency. Other operating costs are not particularly high. We can and do operate gas turbines by remote control with no workers on site.

  31. August 16th, 2015 at 01:02 | #31

    @BilB
    I think Will’s and my argument on storage is that from a system point of view central storage will be no more expensive than domestic (at worst constant returns to scale with banks of Powerwalls) and far more probably much cheaper (pumped hydro, where plants go to GW scale, CSP, huge flow batteries).

    However, it’s difficult to imagine a payment system which would harmonize the private and the system costs. It might work if householders paid the wholesale spot rate (the order of magnitude in most places is 5c/kwh) plus a very large fixed charge for grid connection. Nobody is proposing this revolution, which would drive many offgrid. So as I said we will end up with an inefficient overbuild of domestic storage, which will look entirely rational to the consumer.

    I’ve commented elsewhere against Will’s views on baseload, nuclear, and solar. We are not soulmates.

  32. August 16th, 2015 at 01:04 | #32

    @BilB
    I think Will’s and my argument on storage is that from a system point of view central storage will be no more expensive than domestic (at worst constant returns to scale with banks of Powerwalls) and far more probably much cheaper (pumped hydro, where plants go to GW scale, CSP, huge flow batteries).

    However, it’s difficult to imagine a payment system which would harmonize the private and the system costs. It might work if householders paid the wholesale spot rate (the order of magnitude in most places is 5c/kwh) plus a very large fixed charge for grid connection. Nobody is proposing this revolution, which would drive many offgrid. So as I said we will end up with an inefficient overbuild of domestic storage, which will look entirely rational to the consumer.

    I’ve commented elsewhere against Will’s views on baseload, nuclear, and solar. We are not soulmates.

    @Ronald Brak

  33. August 16th, 2015 at 01:09 | #33

    @Ronald Brak
    In support, the EIA publish capacity factor data for different generating technologies in the USA. The “baseload” combined cycle gas plants come in at around 40%, only a little below coal. The straight gas turbines, built as cheap peakers, come in at under 10%. (http://www.eia.gov/todayinenergy/detail.cfm?id=14611)

  34. Will Boisvert
    August 16th, 2015 at 03:34 | #34

    BilB 17:

    –” But your installed price is a worst case retail price, and not representative of the market expectation. The article I linked demonstrates 4 pricing options depending on a range of distribution methods. Storage power at 10 cents per unit is entirely practical .”

    Not for household battery storage, not according to the CleanTechnica article you cited. Their lowest listed prices of 10 to 12 cents per kwh were for wholesale or retail *purchases* of batteries and didn’t include installation costs. Unless you are a licensed electrician, you’re going to have to pay someone a lot of money to install them and that raises the per kwh price (50 to 100 percent, to judge by other data in the article). If anything, my figure of 14 cents per kwh lowballs the cost of home battery storage, at least according to your own source.

    No way around it–utility-scale battery storage is a lot cheaper than home battery storage, which means batteries *increase* the economic advantage of grid baseload generators over distributed solar.

  35. August 16th, 2015 at 03:37 | #35

    James, if you regard the system as, “storing energy”, whether as pumped storage, in flow batteries, or chemical energy in the form of an explosive fluid ready to be shoved in an open cycle turbine, then utility scale storage is, at this point in time, going to be much cheaper than home and business storage.

    But if you regard the system as being, “Australians getting electricity to use”, or more logically, “Australians getting the services that electricity use provides” then in Australia at least, it looks as though the total system cost will be lower if home and business storage is used rather than utility scale storage. That is, Australians who use electricity will end up paying less on average per kilowatt-hour with home and business storage rather than utility scale storage. By definition, the lowest cost method must be the one under which Australians pay the least.

    And I’ll mention that any time you have a fixed charge as a component of grid electricity bills it has the effect of discouraging efficiency which results in more kilowatt-hours being sold, higher peak demand, greater transmission costs, and a higher price being paid by Australians for “the services that electricity provides”. If our daily supply charges were eliminated and the marginal cost of electricity increased by enough to make up for the lost revenue, we would see an immediate decrease in electricity use as people economise more, and we would have an immediate decrease in greenhouse gas emissions. And this effect would grow over time as people purchase more efficient appliances and lead less grid electricity intensive lifestyles. So getting rid of supply charges entirely would be a good start, and from an efficient allocation of grid resources and transmission infrastructure point of view it might be a good idea to charge a multiple of the spot price. In Australia this would be about 10 times the current spot price if electricity distributers were to take in as much dosh as they do now, but with this system we wouldn’t have to build any new transmission infrastructure throughout almost all of Australia for a very long time, and so all else equal, the multiple should decrease over time.

    Just to be clear I’m not saying that charging a straight multiple of the spot price would be the best way of doing things, just that it would be a clear improvement over what we do now.

  36. Will Boisvert
    August 16th, 2015 at 03:55 | #36

    Sorry, for BilB 17, that should be BilB 28. (In my comment at 32.)

  37. ZM
    August 16th, 2015 at 08:52 | #37

    I agree with Will Boisvert about the grid. Retaining a grid is important, as it allows for the energy system to be managed with the best interests of the whole public in mind. For instance, without baseline energy, energy supply is likely to be somewhat more variable. So at times of low energy generation, it will be better to have a grid and the energy regulator’s office can advise that energy shouldn’t be used for less important purposes, and ensure that the more important uses have sufficient energy for their purposes.

  38. Collin Street
    August 16th, 2015 at 09:26 | #38

    Sure. But neither funding the grid through connect charges or a per-kw tariff works, as people have set out above. There’s no market mechanism that’ll get you what you want: if you want a grid you have to pay for it explicitly through taxes.

  39. ZM
    August 16th, 2015 at 10:13 | #39

    I agree in theory that the State government here in Victoria should have kept the utilities in public ownership. I am not quite sure in practice how we can return the utilities to private ownership in the near term, unless everyone bought utilities bonds to fund a public buy back.

  40. Ikonoclast
    August 16th, 2015 at 10:36 | #40

    @ZM

    Easy. Nationalise them. There would be some fine detail to nationalisation which might well entail full compensation for small investors and appropriation of excess wealth from rich people. Mind you, if you took wealth off the US oligarchy they would send an invasion force so you would have to keep an eye on realpolitik.

  41. ZM
    August 16th, 2015 at 10:58 | #41

    If you wanted to nationalise them you would have to come up with a plan to make it acceptable. Since the companies paid for the utilities they should at least get the price the government charged back, as you can’t have governments selling things and then nationalising them again at a profit.

  42. BilB
    August 16th, 2015 at 11:09 | #42

    ZM,

    The grid will be maintained. However, 30 years from now it will be substantially different. City centers and industry as we have built them need to be maintained. The grid will maintain its stability be reconfiguring those loads. It is not the huge drama that the anti renewable lobbyists are imagining.

    The hardware for residential solar compatibility will be substantially different. I am just oing through the evaluation process now as I face the challenge of retaining all of my solar energy for my own use. Heating and cooling is an issue. I was about to instal a twin head split reverse cycle air conditioner to soak up daytime electricity surplus but when I go through all of the issues it looks like what I need to install is twin tank water system with a heat pump between 2 insulated water tanks which will be 16 C and 34 C. The heat pump has then a constant work load moving heat through 18 C , but that could be adjusted for economy peak season. Then the task is to use the respective reservoirs for managing air temperature within the dwelling. This system then allows the reservoirs to be contributed to by other systems such as the PVT thermal panels, hot water from a log fire in winter. Summer cooling of the hot tank will require a fan and radiator, perhaps with a wet feature in extreme heat. The tanks don’t have to be particularly large, I have to work out how big the thermal reservoirs need to be for various house sizes. In this system the reservoirs add to the energy storage capacity of the system complementing the Powerwall substantially.

    To put some numbers on it a 180 litre water cylinder at 45 C temperature differential contains 9.4 Kwhrs of energy. Times 6 million that is 56 Gwhrs of storage. So if a property had two 180 litre tanks with a temperature differential of 40 C there straight away is another 54 Gwhrs of storage and 9 Kwhrs per energy storage per property .

    So you see Will Boisvert, where your approach is to tear down renewables component by component, when the system as a whole is considered, a complete working unit in the way a car is a multiplicity of technologies working to provide the complex function of comfortable transport, the whole exercise becomes efficient and affordable.

    I am trying to keep the physical size of the hardware package to a modest size, perhaps a tall twin door refrigerator size. This will house all of the components of the energy system: inverter; powerwall; backup generator; 3.5 kw heat pump; fluid circulation pumps; primary thermal ballast tanks (perhaps 60 litres each with a phase change medium if there is a suitable one for these temperatures); and heat booster module for household hot water. For anyone who went to the Sydney boat show they might have seen much of this hardware as it is applied to yachts and power boats.

    It took 2 guys 6 hours to fit my new 4 Kw system (PV’s not PVT’s), including rewiring the power board. It does not take as long as some might imagine. When the complete system hardware arrives in one module as described above, it might take another 6 hours including fitting room radiators and doing the plumbing. I am thinking that I need to expand my modular design out to 6.5 Kwhrs to cover most features including charging one electric car in winter and 2 in summer. But we will see.

  43. Collin Street
    August 16th, 2015 at 11:41 | #43

    Since the companies paid for the utilities they should at least get the price the government charged back, as you can’t have governments selling things and then nationalising them again at a profit.

    There’s no legal or moral impediment per-se; the commonwealth constitution only provides for nationalisation on just terms, which has been repeatedly held — and was almost certainly intended — to mean “reasonable in the totality of the circumstances” rather than market value or what-they-paid-for-it.

    [and the states are under no limitations at all except for implicit guarantees of procedural fairness and natural justice: much the same end result.]

  44. Ikonoclast
    August 16th, 2015 at 11:42 | #44

    @BilB

    We don’t have to worry. Renewable energy will now win hands down by standard physical, technological and economic processes. Debating renewables is already a post hoc pastime.

    This is not to say we have nothing to worry about. In-built climate change, death of the oceans and species extinctions are still big concerns to name some examples.

  45. John Quiggin
    August 16th, 2015 at 11:48 | #45

    Replying to Will and James, I’m not claiming that solar PV with distributed storage and modest grid backup is the optimal solution, just that it’s:

    (a) technologically feasible, in particular not precluded by EROEI concerns
    (b) economically affordable, even if not least-cost
    (c) very close to zero emissions

    That is, we have a solution to the problem of decarbonizing electricity supply (and therefore also, most kinds of motor transport). If other solutions turn out to be cheaper, or superior in other ways, that’s great.

  46. August 16th, 2015 at 12:16 | #46

    In my previous comment I appeared to be saying that it could be practical for households to pay a straight multiple of the spot price for grid gelectricity. It is certainly not practical at the moment, though it would definitely reduce the need for new transmission infrastructure. And bankrupt a lot of people who neglected to turn the mains power off during a critical peak. It would only be practical for people with smart home energy storage, or once there is enough energy storage to completely eliminate periods of very high spot prices, and even then it certainly may not be the best thing to do. Of course we could introduce a system now that results in people paying nothing for electricity when its spot price is zero and more when the spot price is high, which is much less extreme than a straight multiple and would will still reduce the need to build new transmission infrastructure.

  47. BilB
    August 16th, 2015 at 13:48 | #47

    Ikonoclast,

    I agree, renewables are locked in. But I have made a lot of claims about the subject over the years, so now I am in the process of verifying whether I was correct, or to what degree I was wrong. If you read the second para at #42 this is what I am planning fight now. I am grappling with developing a real solution for keeping my house both warm and cool within my solar panel ouput. I’m pretty comfortable with what I have come up with, just this morning. I’d rather have an absorptive chiller, but that is not happening yet so it will be a rankin heat pump and I have to not waste any energy. That is why it is a two tank thermal transfer and ballast method much as you were explaining up page.

    I would love to be able to just buy the complete thing “turnkey”, but we are just not quite there yet.

  48. Will Boisvert
    August 16th, 2015 at 16:29 | #48

    JQ 45

    Well, yes, what you’re proposing is an energy system where we have natural gas generation, capable of running the whole grid as “backup” and then build more and more intermittent capacity to try to abate as much gas generation and CO2 as possible. That is a feasible system, and it may be “affordable” depending on how much residual carbon emissions you’re willing to tolerate.

    But let’s be clear that the notion of comprehensively decarbonizing the grid with intermittents has now gone out the window. I’m not sure that’s good. World electricity usage will soar in the future and even a 20 percent residual fossil component over time will approach unsustainable absolute levels.

    You’ve rather breezily pronounced that we can reach “80 percent capacity” with solar, which is surely true, and that that will translate into an even higher fraction of generation, which is surely false. Remember the capacity factor, at best 20 percent with distributed solar, in most of the world much less (10 percent in Germany). Your 80 percent solar capacity will contribute perhaps 16-20 percent of yearly generation. Germany right now has solar capacity of almost 50 percent of the country’s peak demand, but it contributes only about 5 percent of actual generation.

    So overbuilding of intermittent capacity that runs into a wall of diminishing returns is a necessity for high intermittent penetration. Batteries are supposed to help with that, but the most they can do theoretically is to ensure that very low intermittent capacity factors are not further eroded by curtailment of peak production. And they can only do that if battery capacity itself is grossly overbuilt to bridge the very long periods of surge and slump.

    If we want high penetrations of intermittents, we either have to grossly overbuild redundant capacity, or grossly overbuild redundant storage (or, realistically, do both). And as you have now conceded, we still won’t be able to completely decarbonize the grid or sever the dependence of renewables on fossil-fuel backup.

    Batteries simply cannot solve the intractable problems of wind and solar.

  49. Ikonoclast
    August 16th, 2015 at 16:31 | #49

    @BilB

    As far as I go, I am neither a physicist, nor an economist, nor am I schooled in the technological arts, nor a manufacturing man, nor a manager, nor now even a working man. Other than that, I know everything. I just want you to know that. 😉

    (Apologies to H. Norman Schwarzkopf.)

    Having said the above, I think you are on the right track. I can’t say which particular energy generation and storage methods will be best, energy-wise and finance-wise for an off-grid property, but I think in Australia utilising solar power and the principles and effects of thermal ballast, thermal gradients, insulation and passive design will all be key features.

    It seems pretty clear to me that household energy storage will not be all about battery (chemical potential) storage. Only about 50%, very roughly, of domestic energy storage will be by battery. Another 25% of energy storage will be solar heated hot water stored as (surprise! surprise!) hot water. Another 25% of stored energy will be stored in thermal ballast for household heating or this final 25% of energy will be used to set up an insulated thermal gradient to keep a house cool. The latter we can colloquially term “stored cold”.

    These percentages are very rough BTW and apply to a tropical and sub-tropical country like Australia. Since 2/3 of the world’s population live in the tropical and sub-tropical zones these are the solutions which will work for the majority of the world’s population.

  50. August 16th, 2015 at 16:38 | #50

    Thanks for the EIA link, James. One interesting thing is we are actually appear to be reducing the number of open cycle turbines in Australia because we’re overstocked with them at the moment. New combined cycle gas plants were modified to operate in peak mode instead of load following as the expected demand they were built for did not materialise, and rooftop solar now consistantly supplies a significant portion of electricity use during periods of peak demand, which are in the afternoon during summer heat waves. In a couple of years, if home and business energy storage takes off, we might have even less need for open cycle gas/liquid fuel turbines.

  51. Ikonoclast
    August 16th, 2015 at 16:41 | #51

    @Will Boisvert

    Sorry Will, you are wrong on all counts. There are studies which show over-building redundant renewable capacity is indeed fully energetically and economically feasible.

    These guys from Stanford have done a lot of work in this field.

    Graeme R.G. Hoste
    Michael J. Dvorak
    Mark Z. Jacobson

    This is not their only paper on the general topic.

    http://web.stanford.edu/group/efmh/jacobson/Articles/I/CombiningRenew/HosteFinalDraft

    Also look up “Stanford engineers develop state-by-state plan to convert U.S. to 100% clean, renewable energy by 2050”.

  52. rog
    August 16th, 2015 at 19:01 | #52

    It’s a wonderful thing that despite all the claptrap from conservatives over climate change and negativity from so called experts over domestic solar so many people have voluntarily put their hands in their pockets and invested in a roof top system.

    And look at just how many roofs are sporting an array.

    Meantime the PM is tying himself in knots over moral policy.

  53. ZM
    August 16th, 2015 at 19:44 | #53

    Will Boisvert,

    “Well, yes, what you’re proposing is an energy system where we have natural gas generation, capable of running the whole grid as “backup” and then build more and more intermittent capacity to try to abate as much gas generation and CO2 as possible.”

    The idea is not to use natural gas but to use biogas for back up energy, as we have plenty of materials we can convert to biogas in Australia.

    There are at least three reports on 100% renewable energy for Australia where they use modelling of weather over a year to see how their ideas work.

    The only one I’ve read is by Mark Diesendorf in NSW who proposes to replace the current baseload system with an energy system with a first level of variable electricity provided by solar p.v. and wind turbines, and to even this out a second level fluctuating electricity provided by concentrated solar thermal which can go in there desert (but you don’t want too much as it will spoil the desert ecologies) and also biogas turbines in long grey spells in Winter.

    Beyond Zero Emissions here in Victoria have a slightly different plan, with more of an emphasis on concentrated solar thermal.

    I have never heard a discussion on why the different reports have made their different choices and which is better, but I guess having 3 or more will give the government more choices when they begin to act properly and more rapidly to mitigate climate change.

  54. ZM
    August 16th, 2015 at 20:04 | #54

    “But let’s be clear that the notion of comprehensively decarbonizing the grid with intermittents has now gone out the window. I’m not sure that’s good. World electricity usage will soar in the future and even a 20 percent residual fossil component over time will approach unsustainable absolute levels.”

    No this is not correct. As I said there are at least 3 reports for Australia showing 100% renewable is possible.

    There needs to be similar work done in other countries too, for instance someone told me they were in Vietnam and there is a big data gap on the renewable energy potential there. But even if as they are tropical they have a lower renewable energy potential, as they are part of Eurasia all of Eurasia can share an energy grid so as to accomodate the needs of tropical countries. In return tropical countries can preserve and increase their rainforests to draw down carbon.

    Here is a report you can read that is for the least cost energy grid so you should be pleased with it

    Least cost 100% renewable electricity scenarios in the Australian National Electricity Market
    By Ben Ellistona,?, Iain MacGilla,b, Mark Diesendorf

    http://www.ies.unsw.edu.au/sites/all/files/profile_file_attachments/LeastCostElectricityScenariosInPress2013.pdf

    Mark Diesendorf also writes as well as energy efficiency we will need energy conservation, and has written a couple of papers with Lawrence Delina on A Wartime Mobilisation For Rapid Mitigation Of Climate Change , as it is going too slow at present

    Is wartime mobilisation a suitable policy model for rapid national climate mitigation?
    By Laurence L. Delina and Mark Diesendorf

    http://www.ies.unsw.edu.au/sites/all/files/DelinaDiesendorf_EnergyPolicy-v2%5B1%5D.pdf

  55. August 17th, 2015 at 00:14 | #55

    @John Quiggin
    OK.
    I dimly remember from looking at linear programming as a graduate student that the practical way you solve a real-life optimisation problem is to establish a solution that works, and then fiddle around in the neighbourhood to improve on it. That was the way Soviet central planners did it. Their system worked, in the sense that the reason it fell apart was not failures of coordination but of quality and innovation. If the system isn’t conveniently convex, you may approach a local optimum that isn’t a global one. The substitution relationship between domestic and grid storage doesn’t a priori look eccentric in this way.

  56. Ikonoclast
    August 17th, 2015 at 08:18 | #56

    This article touches on a number of debates in this thread.

    http://cleanenergytransmission.org/stanford-study-u-s-can-move-to-100-renewable-energy-much-sooner-than-you-think-transmission-infrastructure-is-critical/

    The study apparently shows that in a fully renewables grid;

    ” – More than 90 percent of renewable generating capacity is utility-scale – including a large majority of solar PV; virtually all generation is connected to the grid.

    – Transmission and non-battery storage balance the natural variability of wind and solar to provide 24/7/365 power to everyone, everywhere.

    – Electrifying everything, including transportation, together with energy efficiency, demand response, and distributed generation make are essential to the ultimate goal.”

    Also;

    “Utility scale photovoltaics (PV) produce power at half the cost of rooftop installations. Wind power is only economical at large scales. The overwhelming majority of solar and wind power produced in the U.S. today comes from grid-connected, large scale facilities owned by or under contract with utilities.

    Jacobson’s vision of the future is no different, with large scale renewables providing about 93 percent of the power: 50 percent wind (30.9% onshore; 19.1% offshore); 30.7% utility-scale (PV), 7.3% concentrated solar power (CSP) with storage, 7.2% rooftop PV, 1.25% geothermal power, 0.37% wave power, 0.14% tidal power, and 3.01% hydroelectric power.”

    “ow does Jacobson turn myriad variable renewable generators into smooth and reliable 24/7/365 power in all 50 states? Transmission and storage – batteries not included. Transmission lines slash natural variability by blending diverse wind and solar resources over large regions:

    “. . . while the study bases each state’s installed capacity on the state’s annual demand, it allows interstate transmission of power as needed to ensure that supply and demand balance every hour in every state. We also roughly estimate the additional cost of transmission lines needed for this hourly balancing.”

    More transmission, Jacobson notes, would make it even easier and cheaper to achieve 100% renewable energy, by allowing the best quality, least cost resources to serve more customers in more states:

    “ . . . if we relax our assumption that each state’s capacity match its annual demand, and instead allow states with especially good solar or wind resources to have enough capacity to supply larger regions, then the average levelized cost of electricity will be lower than we estimate because of the higher average capacity factors in states with the best WWS resources.”

    Storage and demand response take care of the remaining variability – but not batteries – which are exclusively reserved for their higher value use in transportation:

    “Solutions to the grid integration problem are obtained by prioritizing storage for excess heat (in soil and water) and electricity (in ice, water, phase-change material tied to CSP, pumped hydro, and hydrogen); using hydroelectric only as a last resort; and using demand response to shave periods of excess demand over supply. No batteries (except in electric vehicles), biomass, nuclear power, or natural gas are needed.”

    In Summary.

    According to Jacobson et. al.,

    (1) utilities will remain the mainstay of power generation even with renewables;
    (2) battery storage of power will not needed except in vehicles.
    (3) distributed generation and state interconnectors will play a big role;
    (4) most grid energy storage will be as heat reconvertable to power.

    Look at the graphic in the article for percentages of contribution. I agree with the idea of avoiding biomass burning as a contribution to power though niche applications could still happen like sugar mills burning bagasse. I have my doubts that off-shore wind would be economically efficient or necessary at such a high percentage rate. I would question CSP (concentrating solar power) being so low. I would question no mention of solar updraft towers which solve the storage problem by generating power 24/7.

    One final point which many people forget. A fully electrical economy in the transport sector is four times more energy efficient than an oil economy. Thus it will need 1/4 of the net energy (EROEI in “gathering” energy) to do the same transport work. Electric motor energy efficiency approximately equals 80%. Internal combustion motor energy efficiency approximately equals 20%.

  57. John Quiggin
    August 17th, 2015 at 08:42 | #57

    @Will Boisvert

    Your argument about capacity is still stuck in “baseload” mode.

    In the setup I’ve described, solar PV will be first in the order of merit, followed by stored power from batteries, followed by gas-fired grid backup, used only during long periods of overcast weather. So, the 80 per cent solar PV capacity will translate into more than 80 per cent solar generation (either directly or via battery).

  58. James
    August 17th, 2015 at 10:50 | #58

    Siemens has a neat product which users renewables to create hydrogen, which is then used to fuel gas power plants. Once scaled, solar pv could supply 100% of loads. This system is economic when retrofitted to existing gas plant.

  59. ZM
    August 17th, 2015 at 13:55 | #59

    Will Boisvert,

    ““But let’s be clear that the notion of comprehensively decarbonizing the grid with intermittents has now gone out the window. I’m not sure that’s good. World electricity usage will soar in the future and even a 20 percent residual fossil component over time will approach unsustainable absolute levels.”

    No this is not correct. As John Quiggin said you are thinking of the old style energy system thinking.

    As I said there are at least 3 reports for Australia showing 100% renewable is possible using models of real weather to make sure it is feasible.

    There needs to be similar work done in other countries too, for instance someone told me they were in Vietnam and there is a big data gap on the renewable energy potential there. But even if as they are tropical they have a lower renewable energy potential, as they are part of Eurasia all of Eurasia can share an energy grid so as to accomodate the needs of tropical countries. In return tropical countries can preserve and increase their rainforests to draw down carbon.

    There is no reason that energy grids need to stop at arbitrary national boundaries and not be shared.

    Here is a report you can read that is for the least cost energy grid so you should be pleased with it

    Least cost 100% renewable electricity scenarios in the Australian National Electricity Market
    By Ben Ellistona,?, Iain MacGilla,b, Mark Diesendorf

    http://www.ies.unsw.edu.au/sites/all/files/profile_file_attachments/LeastCostElectricityScenariosInPress2013.pdf

  60. ZM
    August 17th, 2015 at 13:56 | #60

    Mark Diesendorf also writes that as well as energy efficiency we will need energy conservation, and he has written a couple of papers with Lawrence Delina on A Wartime Mobilisation For Rapid Mitigation Of Climate Change , as it is going too slow at present

    Is wartime mobilisation a suitable policy model for rapid national climate mitigation?
    By Laurence L. Delina and Mark Diesendorf

    http://www.ies.unsw.edu.au/sites/all/files/DelinaDiesendorf_EnergyPolicy-v2%5B1%5D.pdf

Comments are closed.