Mike Jonas,
Australia’s retail electricity price has gone through the roof and is still climbing. This is a proposal for how to bring it down again. Something like this proposal already operates in many other countries, so this is not a step into the unknown. Actually, I would claim that over the last few years the Australian system has been forced further and further into a dark hole, and this proposal is for a large step into the light. Although large, the step is simple, rational, and inexpensive.
The Basic Principle
The basic principle behind this proposal is:
The purpose of the electricity grid is to meet demand at minimal cost.
This is very important, and the present disastrous situation is in large part because this basic principle has been discarded. Note that there is no mention in the basic principle of any need to pander to the needs, wishes and vagaries of electricity suppliers. The suppliers’ only purpose is to meet demand profitably, and the proposed system’s only purpose is to enable that at minimal overall cost.
The Present System
The present system uses an auction mechanism to match supply to demand. Each supplier submits bids to supply certain amounts of electricity to given regions for upcoming 5-minute trading intervals. The Australian Electricity Market Operator (AEMO) runs a dispatch auction every 5 minutes, and accepts bids cheapest first until they have enough to match expected demand. From this, they set a spot price for all accepted suppliers and all users. This system is known as the National Electricity Market (NEM).
As currently operated, bids can be submitted day-ahead (or earlier) but remain re-biddable up to dispatch. This fuels volatility, because effectively there is just one short-term 5-minute-based system. It has been observed that: “… regional electricity markets in Australia are characterized by relatively low levels of annual, weekly and intra-daily seasonal patterns, but are by far the most volatile markets in this study.” [1].
The Proposed System
The proposed system is based on the fact that most demand is fairly well known in advance – AEMO know roughly what tomorrow’s demand will be, they just don’t know exactly how it will pan out minute by minute. So if AEMO can cater for most of the demand in advance, then they only need to be flexible for the relatively small part of demand that varies. The proposal is: There should be a binding day-ahead auction, by region, for specified quantities of electricity for specified periods of the following day. The specified periods would range from an hour to the full 24 hours. In other words, AEMO can work out its expected requirement by region for each part of the following day, and lock supply in for most of that. The next day’s 5-minute auctions then only need to deal with variations in demand, not the whole of demand.
Variations of this proposed Day Ahead Market (DAM) system are used by many countries [2], and significantly reduce volatility – “We also find that electricity markets organized as day-ahead markets exhibit a significantly lower overall price variation compared to markets with real-time trading.” [1]. For suppliers, a day-ahead market also increases certainty. Businesses can plan better with certainty, and this factor alone will, in time, reduce overall cost [3].
The present system already has mechanisms for handling situations like a successful supply bidder failing to deliver, expected demand failing to materialise, and organisations trying to game the system by eg. cartel operation or withholding supply. All of these mechanisms can continue. Some new provisions might be needed for the day-ahead market, such as deviation charges [4] and ‘gross’ bidding [5]. The present system also allows suppliers to have backup arrangements with each other, and these can continue too. So, for example, a solar company could put in a bid for several hours’ supply to a nominated region for the next day, and protect themselves by having a backup agreement with a gas company in case there are more clouds than expected.
The point of this proposed system is that it matches the market mechanisms more closely to the patterns of demand. In other words, it aligns the market with the basic principle as stated above. No kind of supply is favoured or penalised. Wind and solar companies, for example, whose supply is unpredictable, can still participate fully by getting backup agreements. The main difference from the present system is that suppliers become more responsible for variations in supply, allowing the market operator to concentrate more on meeting demand.
The processing logic for the day-ahead auction is far from simple, but it isn’t new – it has already been done. The USA’s PJM, for example, optimises diverse bids into a least-cost 24-hour plan [6], and typically clears about 95% of the next day’s demand. This forward-clearing approach has delivered estimated 10-20 % system-cost savings through better planning and lower volatility.
Oh, and one more thing: there are no government subsidies or mandates in the proposed system. Removal of those, in time, cuts overall cost too.
Implementation
It really should not take long to get the proposed system up and running.
AEMO already uses linear programming (LP) for security-constrained economic dispatch (SCED). Extending it to a day-ahead market should not be all that difficult. In any case, equivalent commercial software – like GE’s PROBE or Hitachi’s Network Manager EMS – is readily available and licensed to other operators. At $A20-30M, the cost pales into insignificance beside Snowy 2.0’s overruns. But it could make sense for one state or region to pilot it first before national rollout.
With the greater understanding that should come from using a day-ahead system, it should be possible to improve the long-term planning (“Capacity Market”). In future, will a nuclear plant be cheaper in total than more renewables, for example? Certainly, Australia should use much more of its own gas and coal [2].
Acknowledgement. Grok helped a lot with finding information, reviewing the article, and suggesting improvements. Please note that any errors are mine alone – I still have to check everything.
References and Notes
[1] Electricity markets around the world – Klaus Mayer & Stefan Trück (2018), Journal of Commodity Markets.
[2] Countries using a version of DAM include Canada, Norway, Finland and USA (‘PJM’ states). Their electricity prices are mostly lower than Australia’s, and less volatile. Norway and Finland do have the advantage of plentiful hydro, but Australia has the advantage of cheap domestic gas and coal. It is noteworthy that Finland and the PJM states have some nuclear power, which is competitive in a DAM. See also [3].
[3] Some countries that use a DAM have higher electricity prices than Australia, so some other things have to be in place (or absent) as well. Grok pointed out that these countries’ high prices typically stem from external factors like high taxes, heavy renewables subsidies/mandates, and gas import dependence.
[4] Imbalance settlement price – This example is from France. A price is applied to deviations from agreed supply. Note that there can be rewards as well as penalties.
[5] PJM Operating Agreement §1.10.01(b) (FERC Docket ER24-1234, filed Jan. 30, 2024: https://elibrary.ferc.gov/eLibrary/document/20240130/5073). “Each Generation Owner shall offer into the Day-Ahead Energy Market all of the output of each of its Generation Capacity Resources up to the Economic Maximum for such Generation Capacity Resource for each hour of the Operating Day, unless such Generation Capacity Resource is on a planned outage or derated for maintenance or other reasons.“. NB. I am assured that this citation and quote are correct, but I couldn’t check them because I couldn’t open the link. I could find Energy Market Must Offer Requirements for Generation Capacity Resources which does show that suppliers must bid all their available capacity. This is called ‘gross’ bidding.
[6] PJM Fact Sheet – Understanding PJM’s Markets (2024)
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Bravo..! Words from the Wise..!
Why not extend the proposed system and have the bids FIT to a typical 24Hr demand curve. For example, a low overnight supply, a morning peak, a low during the middle of the day and again an evening peak.
Only those that will supply against that curve can bid, everyone else can drop off the market since they are clearly parasitic.
Bids should be for 24 hours of supply. And when that improves things, let’s get back to bidding for weeks at a time, then months and years. Industry doesn’t want 5 minute power pricing, they want annual, or even life of plant security.
Only a *bureaucrat would promote 5 minute pricing, (or maybe the owner of an intermittent generator).
*bureaucrat = a post turtle, promoted well above their capabilities.
These are spot prices. Generators and large users are free to negotiate in large time periods for what ever MWh they like, and do. But almost nobody buys 100% power that way as there’s a margin that they have less or more demand. They buy more for that say 5% or sell their excess if any on the spot market
I think this proposal would make life difficult for solar and wind based suppliers to bid accurately one day ahead. That could be a good thing.
Is this what drives the 5 minute method?
Shift to five-minute settlement (post-October 2021): The rule was changed to align settlement with dispatch intervals, using a five-minute period. This was a direct response to the growing share of wind and solar energy, which has more variability.
Humour?
The Devil makes me do it.
Thankfully Mike Jonas acknowledges at the end that his idea is merely a possible palliate disguising the fundamental systematic problem viz. for efficient long-term base-load electricity generation wind and solar do not work.
Yes the baseload is surprisingly high . Sure some variation for night and day and weekend and the morning and evening peaks are the killers as that sizes the whole system
Actually, it is more away of establishing the extent to which wind and solar can work. The beauty of Linear Programming is that it is a perfect fit to electricity supply and it can optimise things that look completely different and/or incompatible to a human mind. In the day-ahead market that I envisage, wind and solar suppliers – not the grid operator – have to manage their variability. So, for example, coal, gas and hydro suppliers can put forward a wide range of supply patterns and prices for the whole 24 hours. A solar company that made bids just for the daylight hours could struggle because the algorithm optimises the entire 24 hour period. That is very different to the 5-minute system, where each 5 minutes is looked at separately. Basically, the wind and solar companies are going to have to come up with bids that cover the whole 24 hours or find themselves bidding just for the 5% of demand that has not been met in the day-ahead market. If they can do that, that is fine by me. What it will deliver is the lowest possible cost over the whole 24 hours.
Over the longer term, the government has to get to grips with what mix of supply is needed, and the patterns emerging from a day-ahead market will help them to do that – if they want to be helped.
The mix of supply needed? Gas, oil, coal, hydro and nuclear. These are the only renewable and reliable energy know to Human Race. Wind and solar can’t even produce enough electricity to sustain their own operation, much less supply cheap and abundant electricity to a society.
This should be extremely lucrative for wind and solar. I mean, they’re both free, so it’s all profit for them.
They are free, aren’t they, Nick? Nick…..?
Actually it could be highly costly depending on the variability of the area they’re in and non-supply penalties! A Day Ahead Market requires reliability, and wind and solar are unreliable sources of energy.
Zig Zag was being sarcastic.
There’s a model somewhere that says exactly that.
From its beginnings I classed the present AEMO bidding system as ratshit designed by an overeducated underworked academic on a personal mission. I am all for change of the type suggested. Why should we taxpaying users of electricity be punished by an algorithm from a person not even elected? Geoff S
DUMP the entire intermittent renewable energy system. It was a blunder from the start and will not get better. Build nuclear fission, develop your NG and coal. Stop blowing up power plants.
Yes, I like this in principle. There’s a catch, how do they calculate forward estimates of electricity demand 24-hours ahead … if they’re relying on temperature forecasts from the BoM, they’d be buying too much electricity every day and the account holders will pay for that.
The elephant in the room is that it’s Spot market only pricing. Smart generators and large clients already buy via direct negotiation well ahead of time and only use spot pricing for a fraction of their transactions. On top of that hedging is done to cover unexpected rises or fall in pricing
My thoughts are that only dispatchable, synchronous suppliers should be allowed onto the grid supply.
Wind and solar industrial estates would have to PROVIDE THEIR OWN BACKUP, and feed through a large, regulated, rotating mass that connects to a generator that provides AC to the grid.
How they keep that mass turning and providing dispatchable electricity is up to them… gas, coal, unicorn farts, …. whatever…
It should NOT be up to the rest of the grid to provide back-up for intermittent suppliers.
Yep level playing field on behalf of consumers at the stroke of a pen to curb the dumping. All tenderers of electrons to the communal grid can only supply those they can reasonably guarantee 24/7/365 (ie apart from unforeseen breakdown) along with FCAS or keep them for themselves. That way fickles have to provide storage to up their average tender or partner with dispatchables and pay them their just insurance premiums for that. Take your pick with crushing the great lie that fickles are cheaper dumpers.
Electrons are DC transmission. AC lines used energy field ‘around’ the lines alternating 50 or 60 cycles per sec. Electric potential
The market price for wind and solar are usually negative when they are exported as in Europe.
Simple enough. Have the suppliers pay to dispose of their energy. That will fix it.
You know how the woke climate changers are always banging on about how it falls more on the weaker sex etc? Well it seems there is a way for equality of outcomes with car insurance premiums by ticking the woke box for fellers-
Australian insurance companies offering cheaper deals for non-binary customers: ‘You’re going to get a better deal’ | Daily Mail Online
You know all important equity makes cents in the long run lefties 😉
First step? Remove all socialist morons from ALL governmental positions.
Mike, this is great article. Pity that Blackout Bowen won’t read it.