The future of energy generation and storage

[Ann]

Pumped hydro built upon an assumption that new wind and solar is later built to do the pumping. It carries the risk that the future does turn out to be coal or that battery prices plunge. That said, in the event that the "problem" is cheaper batteries, the pumped hydro remains useful as such just unnecessarily expensive.

Thank you Smurf, your whole post was incredibly interesting.

Wind power may have serious limitations as a viable source of reliable energy with things such as El Nino events taking place....
From the last ASX Notice from IFN...Infigen Monthly Production - March 2019

 

[SirRumpole]

Pumped hydro built upon an assumption that new wind and solar is later built to do the pumping. It carries the risk that the future does turn out to be coal or that battery prices plunge. That said, in the event that the "problem" is cheaper batteries, the pumped hydro remains useful as such just unnecessarily expensive.

Just a thought from an amateur, give priority for pumped hydro in areas of high rainfall eg FNQ so that the storages will be filled to a greater degree from natural inflow rather than pumping ?

 

[rederob]

I like your details, but am picky as always:

The first, factual, one is that if coal is to remain the mainstay of generation then we need to build more capacity to replace that which is wearing out.

This is an assumption.
The fact is more capacity is being built.
What remains to be seen is how much capacity will be built in the absence of a national energy policy which continues to stymie large scale investment.

Whilst wind and solar are adding energy as many have pointed out, they are not adding much in terms of useful firm capacity.

This is true within the context of existing infrastructure. However, interconnector skeletons with a spine up the eastern seaboard, plus battery storage (and here I include small-scale pumped hydro like that being planned by AGL recently), has the capacity to negate the need for Snowy2.0.

 

[rederob]

Intermittency - light reading:
The solution is easy to follow
Some technical stuff:
Home solar PV & battery
Wind farm & battery
A case for flow batteries
Battery optimisation - wind

 

[Smurf1976]

I like your details, but am picky as always:
This is an assumption.
The fact is more capacity is being built.
What remains to be seen is how much capacity will be built in the absence of a national energy policy which continues to stymie large scale investment.

More than happy to be picked on.

Using AEMO data which is publicly available for the whole NEM (that is, Australia excluding WA, NT and remote areas), committed new capacity (including upgrades of existing plant) as follows:

Coal = 180 MW (upgrade of Loy Yang B (Vic) and Bayswater (NSW))
Gas = 210 MW (new - Barker Inlet power station (SA))
Biomass = 24 MW (Tabeland Mill expansion in Qld)
Battery Storage = 77 MW (several sites, 25 MW in SA, rest in Vic)
Wind = 3622 MW (numerous sites)
Solar = 3267 MW (numerous sites)

Announced closures:

Coal = 2000 MW (Liddell, NSW, 2022)
Gas = 480 MW (Torrens Island A, SA, 2019 - 21 in stages)
Diesel = 34 MW (Mackay OCGT, Qld, 2021)

Now where the complexity arises is with energy versus dispatchable power.

So far as energy is concerned, the wind farms and solar will generate around 15,000 - 16,000 GWh per annum give or take a bit (using typical values, I haven't calculated that using any detailed site specific factors). That's somewhat more than the actual output of Liddell, Torrens Island A and Mackay OCGT combined.

Add in the energy capabilities of Barker Inlet power station and the coal upgrades and biomass and all up it's roughly double the actual annual energy output of Liddell, Torrens Island A and Mackay OCGT in practice.

Where the problem exists however is with firm dispatchable power or in layman's terms generation that can be operated based on demand rather than the weather (wind sun) where the generating plant is located.

For that there's only 710 MW being added versus 2514 MW being closed.

Coal upgrades add 180 MW
Barker Inlet adds 210 MW
Biomass in Qld adds 24 MW
Batteries in Vic and SA add 77 MW subject to being managed appropriately (discharged at the right time)
Wind adds 219 MW based on AEMO's calculations
Solar adds zero since peak demand occurs just on sunset (summer) or when it's completely dark (winter).

There's the problem. Wind and solar is going gangbusters in terms of adding energy but it's not replacing the dispatchable capacity which is already problematically short in one state (Vic) and too close for comfort in two others (NSW and SA).

The realistic solutions to that, other than blackouts or forced restrictions, are (in no particular order):

Shift demand away from the peaks. That is, convince end users to use less electricity during peak times (generally by means of using more at some other time).

Install new generation which takes some sort of primary energy resource (fossil fuels, biomass, hydro, nuclear) and turns it into electricity. That is, build what most people mean when they refer to a power station.

Install any system which is able to provide dispatchable capacity when required, but which is "charged" by means of drawing from the grid at some other time. In practice that's either pumped hydro or batteries (or both).

From a technical perspective it really makes no difference so long as the end result is that the megawatts can be injected to the grid when demand requires it, regardless of what the wind or sun happens to be doing at the time. That said, a storage based solution also gets around the problem of what to do with surplus wind and solar generation when it occurs. That's a problem already in SA, presently it just goes to waste, and is a looming problem in Vic and WA (the latter being a separate grid).

If nothing gets built and the closures go ahead? It'll work at all times when the available supply from remaining conventional generation + wind and solar exceeds current demand but the lights really do go out the moment it doesn't. Given that it's straightforward to turn the power off to an area (eg suburbs or a whole town) the consequences are in practice inconvenience plus whatever economic and political fallout ensues. The grid as such can work with loads being turned on and off - consumers probably won't see it so calmly however.

This is true within the context of existing infrastructure. However, interconnector skeletons with a spine up the eastern seaboard, plus battery storage (and here I include small-scale pumped hydro like that being planned by AGL recently), has the capacity to negate the need for Snowy2.0.

In a purely technical sense, so long as it's built big enough and operated in a coordinated manner, then sure it's an alternative.

In practice though, well I won't claim to know how I'd go about getting it done since it requires a pretty big shift in politics and current thinking.

At the investment and construction level, it needs the politicians out of the way.

At the operational level, it needs commercial rivalry put to the side and operate the storages based on technical factors not who bid the best price for this 30 minutes.

That commercial aspect was an actual problem this past summer in Victoria and really did put consumers in the dark. The owners of batteries decided to discharge them at a time when other companies still had spare capacity available (within Vic and also from SA) in order to profit from the high spot price. In doing so they ran the batteries completely flat by the time they were actually needed. Same thing happened twice - first one shut down some industrial load and the second one blacked out homes and small businesses with that second incident making the mainstream news with its consequences (but not the cause).

I acknowledge that Snowy 2.0 and Battery Of The Nation may not be the best possible solutions to the overall problem. What they have in their favour however are that they are big projects with a single operator of each, thus substantially getting around the commercial rivalry issue, and they are actual proposals not hypothetical.

 

[rederob]

Now where the complexity arises is with energy versus dispatchable power.

Yes, that's been covered many times now.

Install new generation which takes some sort of primary energy resource (fossil fuels, biomass, hydro, nuclear) and turns it into electricity. That is, build what most people mean when they refer to a power station.

NO!
That's old thinking.
You answered it with:

Install any system which is able to provide dispatchable capacity when required, but which is "charged" by means of drawing from the grid at some other time. In practice that's either pumped hydro or batteries (or both)

That's the point I made previously.
You also noted that some wind energy is now being "wasted." Which means that there is already an incremental capacity that can be channeled into flow batteries (or small pumped hydro).
Your subsequent comments note the issues that need to be resolved to make this happen.
Snowy2.0 is now owned by the federal government. I cannot see how it would ever be a viable proposition in the private sector.
I have not seen you show how the economics would stack up, let alone explain how the project becomes practical without about 20% greater equivalent additional electricity capacity being available over and above Snowy2.0's generation, so that the dam is topped up after each energy dispatch.
I am not having a go at you here, but am pointing out that you seem to keep overlooking this factor which must be in place concurrent with Snowy2.0 coming online, and not afterwards. In other words, there is no point in building Snowy2.0 unless you also build no less than an equivalent amount of generating capacity. If there is a logic to that approach, then I am lost.

 

[rederob]

Lazard has detailed storage options, presented from a commercial perspective.
Rather than begin at page 1, you can jump to pages 21 onwards to see real world value snapshots.
The bit that is most enlightening is that flow batteries are expected to decline in cost by between 38 and 45% over the next 5 years.
Tesla's Hornsdale battery was an effective solution to a very small fraction of the overall problem. In 2022 terms it will likely be viewed as a comparatively expensive option.
In fairness, it's a bit like buying most things electronic nowadays. In a few years time you will get significantly better performance for the same price, or less.

 

[moXJO]

Lazard has detailed storage options, presented from a commercial perspective.
Rather than begin at page 1, you can jump to pages 21 onwards to see real world value snapshots.
The bit that is most enlightening is that flow batteries are expected to decline in cost by between 38 and 45% over the next 5 years.
Tesla's Hornsdale battery was an effective solution to a very small fraction of the overall problem. In 2022 terms it will likely be viewed as a comparatively expensive option.
In fairness, it's a bit like buying most things electronic nowadays. In a few years time you will get significantly better performance for the same price, or less.

Do you work in the industry rob?

Flow batteries do look interesting. Whats the lifespan?

 

[rederob]

Do you work in the industry rob?
Flow batteries do look interesting. Whats the lifespan?

No, I don't work in the industry.
A few years back flow batteries were not in the ballpark for residential use. Nowadays they are, and are cheaper on a life cycle basis than Tesla type batteries given there is minimal storage capacity degradation after 10000 cycles (or 25 years of total daily discharge, which is unlikely), and up to 20000 cycles is possible.
Sydney University's Gelion batteries may prove to be even cheaper once scaled into commercial production.

The bit I did not add to smurf's points was that AEMO has views on home battery storage which, if combined with microgrid technologies, could also potentially negate the need for Snowy2.0.
With the rate of decline in price of flow storage batteries it may well be the case that most homes by 2030 which presently have PV arrays will also have an appropriately scaled battery. Indeed, it may be that homes add extra PV panels when installing their batteries depending on the opposite trajectories of price with cost of electricity.

 

[moXJO]

No, I don't work in the industry.
A few years back flow batteries were not in the ballpark for residential use. Nowadays they are, and are cheaper on a life cycle basis than Tesla type batteries given there is minimal storage capacity degradation after 10000 cycles (or 25 years of total daily discharge, which is unlikely), and up to 20000 cycles is possible.
Sydney University's Gelion batteries may prove to be even cheaper once scaled into commercial production.

The bit I did not add to smurf's points was that AEMO has views on home battery storage which, if combined with microgrid technologies, could also potentially negate the need for Snowy2.0.
With the rate of decline in price of flow storage batteries it may well be the case that most homes by 2030 which presently have PV arrays will also have an appropriately scaled battery. Indeed, it may be that homes add extra PV panels when installing their batteries depending on the opposite trajectories of price with cost of electricity.

There are a huge range of different batteries vying for mass adoption.
I do hope these get up.

I get sick of the promise of something then the failure to materialize in a way the consumer can take advantage of.

 

[rederob]

There are a huge range of different batteries vying for mass adoption. I do hope these get up.
I get sick of the promise of something then the failure to materialize in a way the consumer can take advantage of.

Labor has a policy to subsidise 100000 home batteries initially, with a target of 1 million by 2025.
Let's pretend for a moment it happens and, for ease of maths, they add a 10KW battery. That becomes 10GW capacity through household batteries by 2025. Now look at these numbers, and compare them to generation capacity numbers in some of smurfs posts above: Snowy2.0 becomes a stranded white elephant.

 

[moXJO]

Labor has a policy to subsidise 100000 home batteries initially, with a target of 1 million by 2025.
Let's pretend for a moment it happens and, for ease of maths, they add a 10KW battery. That becomes 10GW capacity through household batteries by 2025. Now look at these numbers, and compare them to generation capacity numbers in some of smurfs posts above: Snowy2.0 becomes a stranded white elephant.

But what batteries?
Will there be pollution caused when the battery life ends?
And in the example of the roof batts we had cheap sht that contained chemicals coming in from china.
Can we trust the government and consumers looking for lowest costs on something like that?

 

[basilio]

But what batteries?
Will there be pollution caused when the battery life ends?
And in the example of the roof batts we had cheap sht that contained chemicals coming in from china.
Can we trust the government and consumers looking for lowest costs on something like that?

Fair comment. Lets take Rederobs analysis of the potential of adding 100,000 10kw home batteries to the national grid by 2025.
Then look at the lifecycle values of flow batteries and in particular cheaper low pollution options like zinc bromine technology from Redflow.

If a government chose to establish a set of parameters for battery storage that they would subsidise that encompassed long life and little pollution, they could encourage development in these particular areas.
The partial subsidy of the batteries would ensure a more economic outcome than just throwing many billions of dollars at a Snowy 2 scheme.

https://reneweconomy.com.au/redflow-seeks-18-million-scale-flow-battery-production-62976/

 

[SirRumpole]

Labor has a policy to subsidise 100000 home batteries initially, with a target of 1 million by 2025.
Let's pretend for a moment it happens and, for ease of maths, they add a 10KW battery. That becomes 10GW capacity through household batteries by 2025. Now look at these numbers, and compare them to generation capacity numbers in some of smurfs posts above: Snowy2.0 becomes a stranded white elephant.

I don't get why excess capacity is a bad thing.

Sell it off cheap to start up businesses to encourage more industry into the country.

 

[rederob]

I don't get why excess capacity is a bad thing.
Sell it off cheap to start up businesses to encourage more industry into the country.

How do you sell something which is already in oversupply, and available cheaper by other means?
That's exactly what would happen if 1 million 10KW household batteries were installed before Snowy2.0 was finished.
Here's Snowy2.0's maths in basic form:
1MW electricity generation came from a 1.15MW input into refilling the top dam.

 

[moXJO]

Fair comment. Lets take Rederobs analysis of the potential of adding 100,000 10kw home batteries to the national grid by 2025.
Then look at the lifecycle values of flow batteries and in particular cheaper low pollution options like zinc bromine technology from Redflow.

If a government chose to establish a set of parameters for battery storage that they would subsidise that encompassed long life and little pollution, they could encourage development in these particular areas.
The partial subsidy of the batteries would ensure a more economic outcome than just throwing many billions of dollars at a Snowy 2 scheme.

https://reneweconomy.com.au/redflow-seeks-18-million-scale-flow-battery-production-62976/

I like the product pitch, but they seem a long way out from decent production capacity.

 

[moXJO]

How do you sell something which is already in oversupply, and available cheaper by other means?
That's exactly what would happen if 1 million 10KW household batteries were installed before Snowy2.0 was finished.
Here's Snowy2.0's maths in basic form:
1MW electricity generation came from a 1.15MW input into refilling the top dam.

I'm still not convinced we can get to 1 million batteries in a reasonable time frame?

Maybe mixed but not solely flow batteries. And if it ended up being left to the market, I can guarantee we will end up with terrible products.

 

[rederob]

I'm still not convinced we can get to 1 million batteries in a reasonable time frame?
Maybe mixed but not solely flow batteries. And if it ended up being left to the market, I can guarantee we will end up with terrible products.

I too find 1 million batteries ambitious, but 100,000 would mean the base case for a Snowy2.0 is tenuous, and that's not including the multi-megawatt batteries that each installed wind turbine could top up free of cost (and yes, I have mischievously excluded opportunity cost in order to make a point). This product in a wind farm as pictured could top up a 100MW battery bank (via 3 hours at maximum capacity) each day and still contribute significant additional power to the grid.
And your point about flow batteries is true to the extent that there are few producers for residential applications. This link is a good present state of play locally.

 

[Smurf1976]

I am not having a go at you here, but am pointing out that you seem to keep overlooking this factor which must be in place concurrent with Snowy2.0 coming online, and not afterwards. In other words, there is no point in building Snowy2.0 unless you also build no less than an equivalent amount of generating capacity. If there is a logic to that approach, then I am lost.

Happy to answer the question etc.

Only concern there is being mindful that ASF is a stock forum not an engineering one although that said this subject does have relevance to multiple listed (and other) companies.

Starting from the basics, the key point is that electricity (by which I mean AC power in the grid) cannot itself be stored at all. We can store other things (chemicals, heat, water etc) which can be used to produce AC power but we cannot store AC power itself.

That is in the same way as we cannot store sound or light. A record, CD or MP3 file plus a player, amplifier and speakers is a means of reproducing sound, with some imperfections from the original, but does not store actual sound as such. That is somewhat pedantic but of relevance to the point.

Electricity demand in the NEM (National Electricity Market - basically the whole country except WA, NT and remote areas including Mt Isa) is 21,013 MW.

In order for it to work, everyone's lights stay on, generation output needs to match load in real time and at all times. The margin for error there is trivial and if, for whatever reason, generation cannot or does not match load then that's a huge problem.

If generation is insufficient then, for a known problem, a human directs load shedding. Someone in charge directs that x MW be switched off and that's passed to the network operators to do it as per a pre-determined plan (the details of which are confidential in most states but are public in SA).

If the problem is not foreseen, eg because something suddenly fails, then Under Frequency Load Shedding (UFLS) should come to the rescue and automatically disconnect loads so as to restore the supply and demand balance. At that point it's involuntary so far as humans are concerned - load is going off and there's no choice in that.

The prospect of UFLS operating tends to send shivers down spines for good reason in that it works fine in theory but not always so well in practice, the usual failing being that far more load ends up being disconnected than really ought to be.

The complete blackout ("System Black") in SA was an example of that. Transmission failure caused a loss of generation which ultimately ended up tripping all generation and supply from Victoria in a matter of seconds. Not the intended result but anyone who tells you that isn't a risk when UFLS operates is I would suggest a tad too confident in their engineering. It can go spectacularly wrong yes and that most certainly isn't the only incident to ever have occurred and it's not even the most recent.

On 25 August 2018 the mainland (ex-Tasmania) NEM unintentionally split into three, that being Qld and SA both disconnecting from NSW and Vic respectively. The cause was a lightning strike affecting Qld - NSW transmission causing that to trip and everything else became a consequence of that. SA - Vic power flow became unstable and tripped, frequency slid in NSW and Vic which tripped load in both states, supply from Tas > Vic went to its absolute limit and in doing so tripped some industrial load in Tas.

The responses were intended but in saying that, they have a bit in common with BHP intentionally derailing their runaway freight train or a pilot choosing to intentionally land a plane on water. Intended responses only in the context of being the least bad outcome from a situation which has already gone horribly wrong.

If the above all sounds like a near miss that could have ended seriously badly then that's a fair summary of it yes. There's an AEMO report available publicly (online) for those interested and suffice to say that changes are afoot to prevent a recurrence. Changes of the sort which have broader implications politically and financially but that's another story.

So operating any sort of automated load shedding is something you do only if it's unexpected. If it's expected then a human directs action before the critical point arrives. Action as in blackouts.

If the reverse problem occurs, too much generation, then so long as you have control of the generation then ultimately it can be stopped. Not always without difficulties, particularly when it comes to restarting and the time that takes, but it can be stopped as such.

The exception there is distributed generation (primarily rooftop solar) the overwhelming majority of which is not under any form of central control. Too much power? Can't stop it other than by means of deliberately raising voltage and/or frequency which has its own problems. That's not currently an actual problem but it could become one if enough such systems are installed (and AEMO forecasts this point to be reached by the mid-2020's in SA with Vic coming next).

So to summarise on that point, AC power going into the grid must always equal consumption + losses in real time.

Now the limitation of wind and solar is that operation relies on an energy source, sun or wind, which is not stored and which arrives intermittently at the location of the generating plant. Or in layman's terms, output is highly variable on any shorter timeframe - hour to hour or day to day. Only in the long term, at least monthly, is it anywhere approaching consistent.

Hence the issue that whilst wind and solar most certainly can generate electricity, they do not contribute in any meaningful way to the ability to match supply and demand in real time.

If wind and solar are generating then yes, that means less need at that moment for some other source of generation.

If wind and solar are not generating then there needs to be sufficient generating plant, of some other type, to meet demand in full at that point in time.

In that context coal does not itself produce AC power. Coal produces heat. Heat + water = steam. Steam drives the turbine which turns the alternator. It is the alternator which produces the AC power. Point being that no amount of coal, gas or anything else produces AC power without having the required equipment in place.

That is like saying petrol doesn't move the car. The engine moves the car. Petrol is just stuff you need to make the engine work but petrol of itself cannot move the car. Take the engine away and the petrol is useless by itself.

Which in the context of electricity means that we need a set of non-wind and non-solar generating plant capable of meeting the entire demand for electricity less any "firm" amount we determine that we can count on wind + solar not going below. Trouble is, that "firm" amount is only a few % of capacity for wind and it's literally zero for solar.

Following are some charts for the entire NEM for the past 7 days showing only wind and solar generation:

Spot the problem?

At times wind + solar are close to a third of the total load but at other times production falls in a heap. Now consider that demand is generally high just as the sun sets and solar stops generating and on a daily basis the problem is obvious - wind and solar generate energy but they don't generate AC power on a firm basis as and when it's needed.

They have a lot in common with your neighbour who on some days offers you a free ride into town and back since they travel at the same time you do. That's nice but since they don't do it every day, it doesn't change the fact that you need to own a car. It just means you drive it less.

Now to clarify a point about what is, and isn't, a generating plant.

In a practical operational sense, if it puts AC power into the grid then it is a generating plant.

What it costs, what resource it is using, how much CO2 comes out and so on don't change the fact that it puts AC power into the grid.

What about storage?

At the time of generation, a battery or pumped hydro is a generating plant like any other. It puts AC power into the grid right now. That it can only do so after having previously taken AC power out of the same grid means it is not a net source of generation, since it uses more than it returns due to losses, but if we had 21,013 MW of pumped hydro and batteries then right now they could run the entire grid so long as they're not empty. As a means of balancing supply and demand in real time it works.

Batteries and pumped hydro thus add firm generating capacity, that is they can put AC power into the grid to match demand in real time, despite being a net loss so far as energy is concerned.

In that context they have something in common with a loan. You give me AC power right now, on demand, and I'll pay you back with interest. That in simple terms is the deal.

On the other hand solar and wind intermittently put AC power into the grid but cannot be depended upon to do so at any particular time. They are an intermittent source of energy but are not a reliable source of dispatchable generation. Refer to the chart.

Put the two together and so long as everything's big enough they do indeed cover all aspects. Solar and wind supply the energy with the first priority being supplying load on the grid and the surplus pumps water and charges batteries. When solar and wind fall short then the pumped hydro and batteries supply the required dispatchable capacity, all the way up to well over 90% of load at times. So long as everything is sized adequately, and that's really just a combination or meteorological data + engineering, then it works.

Now here's a chart showing daily generation from wind and solar across the NEM for the past 12 months:

It's all pretty easy in spring and summer to be charging batteries or filling up small pumped hydro schemes perhaps not every day but certainly every few days. That high demand is somewhat correlated with high solar production during those seasons helps in that regard.

I draw to everyone's attention however what happened in the second half of Autumn and on various occasions during Winter last year. It wasn't an isolated occurrence, same thing has happened previously, and there's the problem.

Now about this idea of storing relatively small amounts of energy in batteries or pumped hydro and charging them on a daily basis.

How do we charge them during those weeks?

The answer realistically, unless we oversize wind and solar to huge extent, involves some "big" source of generation which can be turned on and run 24 hours per day, thus reducing load on the batteries and small pumped hydro during the peaks and providing some opportunity for recharging when demand falls off.

The options for that in practice:

Coal, oil or gas could do it so long as we have coal, oil or gas-fired power stations and a stockpile of fuel ready to go.

Existing conventional hydro (Snowy, Tas, to a lesser extent AGL) can do it within the capacity limits of those schemes. Hydro Tas is certainly well aware of that one and already doing serious work, and by that I mean drilling holes in the ground not just something in an office, with a view to improving the ability to run the system intermittently.

Large scale pumped hydro capable of sustained high output can do it so long as the water was pumped up at some previous time.

Beyond that it's really about economics, environment, politics, business trying to make a profit, etc.

In practice though, you won't find too much opposition to what Snowy or Hydro Tas want to do from competitors and there's a reason for that. Scratch beneath the surface, pick any large player in the industry and investigate all their financial deals and you'll probably find Snowy and or HT in there somewhere via a hedging arrangement or some other deal.

It's a great big tangled web in that regard. AGL's power stations do indeed burn gas produced by Origin and others. Alinta burns coal mined by AGL. There's hedging arrangements all over the place. Origin supplies half the gas used by Engie at Pelican Point and takes half the power from it. AGL Hallett is actually owned by Energy Australia. And so on. Things aren't entirely as they seem.

So what about the original question and Snowy 2.0 specifically?

With the closure of multiple existing power stations over the coming years there is a requirement to build substantial new dispatchable generating capacity. Without that, demand can't be supplied in real time and load shedding (blackouts) is the inevitable result.

From a purely technical perspective it makes no difference what sort of firm dispatchable capacity that is, so long as we're talking about capacity that works regardless of short term weather.

So long as there is either unused capacity at fossil fuel power stations on occasion, or there is surplus wind or solar, then building new dispatchable generation which is based on storage, charged by drawing AC power from the grid at a previous time, is a technically viable option.

The other technically viable option is building new conventional generating plant of whatever type.

In the context of storage based options, if the long term intent is to rely primarily on fossil fuels (or nuclear) then the large storage capacity of Snowy 2.0 would be pointless in practice. It would make sense only if it were the cheapest option on a peak capacity basis which it isn't.

If the intent is to rely primarily on wind and solar, gradually phasing out fossil fuels, then Snowy 2.0 or an equivalent is highly useful in achieving that due to it's ability to operate during lulls in solar and wind generation without recharging on a daily basis.

Snowy 2.0 is of itself nowhere close to being large enough. All present hydro both government owned and private in all NEM states, plus Snowy 2.0, plus Hydro Tasmania's projects, collectively meet less than 40% of current system peak demand. As such, and noting that plant closures in Vic and NSW alone over the next 20 years are likely to be more than twice the size of Snowy 2.0 and Battery Of the Nation combined, there is a very substantial opportunity for others wishing to develop storage projects or other new firm dispatchable generation.

On the question of coal I will note that all options are a fossil fuel option in the short term.

Build new capacity based on coal, gas or diesel = that's a fossil fuel option.

Build batteries or any pumped hydro = ultimately in the short term that's a fossil fuel option with added losses on top.

The difference is that the former locks in fossil fuels indefinitely whereas the latter is based upon the notion that fossil fuel based generation will be phased out well within the life of the project such that, for most of its life, its a non-fossil option even though on day one it certainly isn't.

Note that what I haven't said here is whether or not there ought to be a bias toward fossil fuels / nuclear or toward wind and solar. I've very deliberately avoided that due to the controversy surrounding it although I will make the observation that if you look at those who are significant players then none of them are actually working on the basis that coal is the future.

As for the politics of it all, I will simply note that the time for implementing an optimal solution has already passed. It's simply too late now to avoid either spending $$$ more than would have been necessary with earlier action or alternatively accepting supply shortfalls and load shedding. What actually happens there - anyone's guess really.

 

[Smurf1976]

Some more charts for those interested. This one for battery operation in all NEM states over the past 7 days. In practice it is primarily the Hornsdale Power Reserve (aka the "big battery") in SA but does include others in SA and Vic.

As you can see, it's used for very short term "fine tuning" in practice, not bulk energy storage, and does that extremely well.

Next chart shows past week's hydro generation and pumping loads. Generation from this source is low at this time of year due to lower than average electricity demand - a look at the same data in mid-winter or during hot summer weather would show peaks more than double those seen here.

Note that the volatility is intentional, tracking the combined effects of varying electricity demand as well as varying wind and solar generation. Note the peak around 6pm.

Something else I'll add to all this is about the future and changes in demand.

Electric vehicles are one that will likely happen.

The other one I'll note is simply this. A typical Winter day has ~600 GWh of load across all NEM states and ~250 GWh of gas space heating in Victoria. That's only gas heating, not including gas used in industry or for heating water, is only for Victoria, and doesn't include other fuels such as wood.

There will be changes in the means of generation and in end uses but electricity as such isn't going away anytime soon. Especially so if we're to see a move away from the direct use of fossil fuels in engines and heating applications.