I'd recommend the book The Grid as a primer on how stuff like this can happen. Energy is strange, in the sense that everything produced needs to be used more or less immediately. Maintaining the energy grid is a careful balancing act between preventing surges and blackouts.
There are a lot of grid-related engineering problems that come from increasing decentralization (e.g. wind, solar, hydro) because it increases unpredictability in supply.
I wonder if we will ever get appliances that respond to the grid, a freezer that drops its temperature a couple of degrees because energy is cheap or a washing machine that waits an hour because it is expensive.
I suspect there is quite a bit of personal demand that could be shifted easily.
Yup, pretty sure we will. Actually, they've been around for a couple of years now [1, 2]. I think there are smart-grid-enabled electrical heating systems as well.
"Smartening" appliances can lead to surprising security issues [3] as well. If you have the time (and interest) for an entertaining primer into the subject, watch any of Mikko Hyppönnen's recent Talks explaining "Hyppönnen's Law".
When I lived in Zambia, Africa in 2005 there were refrigerators made by South African companies for countries without 24 hour power. Similar idea to this, really.
My house has a smart power meter in it that reduces/shuts off AC during times of peak demand (as commanded by the power company, PEPCO). In return you get a certain amount off each bill.
Bought the book, but spoil a little for me if you would: I know that some utilities turn on pumped hydro storage to handle excess capacity, but what other ways are there to efficiently waste energy?
A long inclined track, and magnitudes of order more energy inputs in the form of steel and super heavy duty tracks and earth moving and maintenance than could ever be recovered.
Soneone here recently ran the numbers assuming you could move rocks unencumbered up and down earths gravity well and the mass required to generate enough electricity to power the world even for a brief movement. The numbers were sobering.
Yea rail storage is pretty niche and definitely can't power the whole world.
The only advantage is that there is a very low switching time on the load from charge to discharge compared to pumped hydro. This could be extra useful for load balancing a distributed grid.
Flywheel storage for wind energy. It's mostly used for grid balancing.
Beacon Power has a complete solution but they went bankrupt in 2012, then got bought by Rockland Capital who intend to rehire the staff and build a 20 MW plant in Pensylvania.
Also, here's an article of such an energy storage project in Ireland by Schwungrad Energie Ltd.
Right now and for the foreseeable future, lithium ion batteries are a better alternative to flywheels, though they will see some use for applications that require high power but little total stores energy.
The power:energy ratio is high, but has few applications on the grid. Even lithium ion's typical 4:1 MW:MWh is a bit high.
The really unexplored territory is seasonal storage. Electricity to hydrogen, methanol, etc. have huge round trip efficiency hits, like 30%-50%, but may make a ton of sense.
Ah finally something I am knowledgeable about.. building a software simulation system for a smart power-grid that can account for renewable energy sources was my grad school research project. Linked here if you wanna learn more: http://arunpn.com/projects/aces-co-simulator/
Perhaps I'm not fully understanding how this works, but could one conceivably run a business whose only product is predicting when the price will go negative and "wasting" power?
Perhaps not just when prices are negative, but it's not an uncommon practice to 'store' power by pumping water uphill when it's cheap and demand is low (generally at night), and then using that potential energy to generate and sell it when electricity is more expensive.
I have absolutely no clue what it would take to make that profitable, but it's a strategy for handling the natural spikes of demand when using power generation methods like solar and wind that are unable to adjust quickly to changing demand.
>I have absolutely no clue what it would take to make that profitable
You need a large watershed with a wide range of useable head. It is an excellent system when the geography and hydrology allows for it.
Unfortunately, such areas are not common or evenly spread, and are not really possible to economically construct without a great deal of cooperation from the local geology.
Also where this geology is present that is usually far away from the places where it is economic to deploy wind-turbines, and often from where it is economic to deploy solar panels.
Also, what "greens" often ignore: these giant storages, are dams on flatter areas are often also devastating to the environment.
You also need one where you're given significant environmental carte blanche. These types of projects are similar to dam projects in the scope and scale of the amount of the environment that they destroy, times two as you need upper/lower ponds.
I honestly do not think there is anywhere in the world that has easy access to the ocean, a large potential hydraulic head, and a pre-existing elevated saline lake to work with. Maybe somewhere in the Atacama? Is there enough power demand in the region to make it worthwhile?
As the sibling comment shows, you can build an artificial lake large enough to be feasible. The Okinawa plant seems to have been decommissioned for lack of need (at its given cost). It was probably a victim of the current oil/gas extraction boom and artificially deflated oil prices. Still, I wonder about their ability to control costs due to corrosion--salt water is horrible for this reason.
I couldn't find much more about the site in English, but considering how expensive the lining must have been for what was a pretty small storage pool, in addition to needing expensive materials like stainless steel and assorted specialty plastics to deal with corrosion...
I can't know for sure, but regardless of whether they say it was due to weak demand or not, I bet it was the saltwater issue that sunk the project more than anything else.
I hadn't realized that the incentive structures could make it possible for an organization apart from the energy producer to run such a facility. An elegant application of market economics.
I don't think there are too many products that you can make money from buying. The few I can think of are things like bee hives (paid to remove it and paid for the bees), dirt/sand in abundance areas (I'm not sure they sell it), and other natural resources. Most of them you're getting paid for removal and not buying though.
Housing and real property can sometimes effectively work that way, where you are paid for buying because you are taking on responsibilities by purchasing the property(say, for liens, cleanup, etc)
Aluminum smelters already do exactly that. They use power no matter what, of course, but are well-positioned to be able to absorb massive amounts of electricity when the cost structure makes it beneficial to do so.
The way you separate aluminum from alumina is by electrolysis. Like you guessed, it is highly parallel. It is as simple as having hundreds of pots, each with huge carbon electrodes, dumping megawatts of power into the molten matrix to extract the metal. And it consumes an obscene amount of power to do so.
This is also the primary reason that aluminum recycling is so beneficial. It is dirt cheap to remelt already refined aluminum vs. smelting fresh ore, even though aluminum itself is an exceptionally common element.
Yes, but aluminium smelting has certain constraints - namely, the molten metal cannot solidify[1]. This restricts your ability to quickly spin up and down smelters, but they are still potentially a very useful part of an energy smoothing strategy.
Huh, interesting. I read the whole thing and understood at least half of it ;).
Are the damages primarily thermal? I couldn't figure it out from that report, although it seemed to somewhat hint that they were. I.e. if you could keep the cells warm (with more insulation or otherwise), would they still get damaged by the shutdown?
Your assumption is correct. The vast majority of the damage comes from thermal stressing of the refractory materials that make up the crucible of the furnace. If there were better insulators available, so that the rate of heat loss through the furnace was minimized, it would be a huge improvement to the industry, both in energy saved during production, and greater amounts of time available after shutdown before thermal effects wreck the furnace.
I remember reading that in aluminum smelting, compared to other industrial processes, the energy cost is an especially large part of the overall costs. So when energy costs are abnormally low, smelting aluminum could go from bad to OK to wildly profitable.
Power traders do this all the time. Each power plant/wind farm/solar farm has a generation cost and a time span that is required for startup or shut down. Balancing these factors and meeting demand the cheapest way possible is their business.
Also keep in mind that the price to consumers will never be negative. This would be only the generation and distribution side of the house.
It is a signal to ramp down or if you have a storage facility store.
If the prices remain negative or low for long periods it is a indicator to build transmission to where prices are high. Or if yours are a consumer build you factor or data center in the zone with low prices.
Another concept is "dispatchable load". With fuel-driven generation, you have dispatchable power -- additinal generating capacity is brought online to meet anticipated demand.
With dispatchable load, some process which can be rapidly activated or deactivated with little loss in overall efficiency is utilised. Examples would be water pumping (simply for water storage, irrigation, or other uses, not pumped-hydro power storage), or aluminium smelting, or electrically-driven thermal or reduction processes.
An equivalent concept is "make hay while the sun shines". The Sun's power isn't dispatchable (though it's fairly predictable), and if you're doing something that relies on it, you're best advised to make use of it whilst you can.
You could even go so far as to have a bank of batteries, be paid to take in power, then earn money selling it back to the grid later when prices are positive.
You'd probably need more batteries than would be economically feasible in that precise setup. But a similar scheme is used with electric water pumps and hydroelectric turbines. It's called pumped hydro:
There is also a Tesla/SolarCity solar+battery plant in Hawaii that opened earlier this year and is currently operational. It is 13 GWh and 52 GW, thus it provides 13 GW of stored solar energy for 4 hours after the sun goes away.
While the utility of electricity for consumers may become zero, transport of it via the grid still costs money. So for producers the electricity price becomes negative.
If you own a large building or campus, you can get an ice maker to integrate into your HVAC system, and exploit electricity price dips. Thenwhen juice is pricy, you're just running air past your stored ice instead of running the AC.
The one catch is if you want this, you essentially have to let the utility control your system over TCP/IP
I'm thinking that'd be a stretch. Anything that would sink enough power would have high capital costs, and negative prices (in my experience in US power markets) don't occur all that often. If you tried this as your primary business model, you'd wind up with a lot of money sunk in assets that may or may not pay off (and if they do pay off, will happen indeterminately in the future).
It would be idle most of the year though. Perhaps a big bank of resistors would be cheap enough to soak up some of that negative priced power and profit.
And it has been so sunny in California that electricity prices have turned negative (late morning, temperate spring weather with low load, high hydro output due to a wet spring)
The Internet and open data movement are just incredible. I am still amazed that regular people can get such insight, and fairly well presented anytime online.
Oversupply is easy to burn up -- as long as there's robust and quick mechanisms for signalling when oversupply is going on.
There's plenty of loads that can be quickly spun up -- transferring workload from one datacentre to another is possible, as is running refrigeration systems to generate ice (or some other phase-change material) that can be used up during peak demand.
Anything remaining can be taken care of by bespoke energy storage systems, like batteries or pumped-storage.
So then pump water up a hill during times of excess and then use that stored energy to spin turbine generators during times of need. just because it's supplying "too much" energy at one time doesn't mean it's not something that could be used effectively with the right infrastructure.
Ideas like this unfortunately end up not being very economical because of the fixed costs of the infrastructure needed and the relatively low duty cycle of use.
And hopefully to be ideologically consistent, these detractors have the same problem with coal and nuclear where it's more expensive for them to shut down and restart than it is for them to keep running and pay the negative price.
I have two qualms about this claim: The first is purely economical. Does this not simply indicate that there is insufficient supply of electricity storage (ie batteries)?
The second is... wow, if excess supply of energy is a reason there are detractors to wind and solar, then maybe treating electricity as a market is totally misaligned with perpetuating sustainable human activity. Maybe we shouldn't do that.
It's not the market that creates that problem for wind and solar--it's physics. Excess electricity supply has to closely match demand or bad things happen to the grid, market or no market.
I don't think it goes far enough, I don't have any incentive as an individual to change my consumption patterns. If my price was continuously variable I would almost certainly change my behaviour to lower my energy costs.
Oversupply is only a temporary problem. When long distance transmission lines to other countries have been built, energy can be transferred across the continent to where demand or available storage is.
This just means inefficient pricing. If prices were adjusted in real time, one could schedule charging car batteries, heating water in the electric water heater, etc., when prices are cheap.
That's why electrification of transport will be good for the grid.
Cars can charge at night, or any other time when energy supply is abundant and/or demand is low, and export back into the grid at peak times with V2G. Storage batteries on wheels!
People charge their electric cars overnight. Water in the hot water tank will stay hot for a couple days, so it makes sense to heat that at night, too.
In hot weather, you can also run the energy-hungry A/C at night to cool a tank of water or a block of masonry, then use that during the day to cool the house.
The reason nobody bothers doing that is because the price of electricity is the same day and night.
Why would it be inefficient to charge your car battery at night? And storing coolness in a masonry slab doesn't sound inefficient to me. Pumping water uphill and running it back through a turbine is pretty inefficient.
I was referring to the cooling part, not the car charging. My totally unsubstantiated gut feeling is that a piece of concrete or a tank of water wouldn't stay cold long enough to be useful. With my local company, electricity during the day is something like 20%-100% more expensive (less of a differential in winter, more in the summer)
These problems are fundamentally about the loss of what's called "spinning reserve".
In a traditional steam turbine generator system, a number of generators are run at well below fully capacity with a full head of steam in the boiler but essentially ticking over.
When there is a sudden requirement for extra power signalled by a relatively small drop in voltage, the inertia in the spinning turbine and alternator take up the load as the governer opens the steam valve releasing as much of the stored pressurised heated steam in the boiler to take up the load, as it is the reduction of temperature of the steam that is the primary source of power.
The loss of so many steam turbine powered systems is causing the problem.
If these issues are to be solved it is about insufficient energy storage systems such as inertial storage systems or batteries to take up the function of spinning reserve.
In certain parts of Australia Frequency is deliberately maintained throughout a wide range of load demands which means it is by other means (magnetic amplifiers) that voltage is used as the signal for power.
It is not until voltage signally cannot maintain the desired frequency that the under frequency load shedding kicks in.
In fact the obsession with frequency is so high that over time beyond any 24 hour period (and thus even years) that an electric synchronous clock is usually always within at most a few 100th of a second out, as the system frequency is adjusted up and down to correct for accumulated errors.
As a result frequency control is spot on but voltage control is generally pretty awful.
Use of load shedding or generation shedding to balance load and generation on a grid are a different mechanism than the continuous adjustment of generator output by the governor in response to changes in frequency that occur as mismatches in real power demand and generation result in changes in the kinetic energy of every synchronous machine.
Load shedding is triggered by underfrequency and undervoltage elements, generation shedding is triggered by over frequency elements.
I am aware that in New Zealand the distribution utilities (especially in Christchurch) perform 'peak shaving' to reduce the peak demand on some substations by turning off people's hot water tanks using a ripple signal injected on to the power system, but that is to alter the load profile not perform frequency regulation.
Every large system operator attempts to balance the positive and negative frequency errors from 60Hz to zero over a 24 hour period so that clocks are correct, that is not unique to Australia.
Can you provide any links to how voltage is used to regulate the frequency in certain parts of Australia? A magnetic amplifier is not a term I am immediately familiar with. It almost seems to me like a normal synchronous generator, where a current in the rotor causes a rotating magnetic field, which then induces a voltage in the stator that is much large (magnetic amplifier).
I am an EE and program power plants including turbine governors that regulate the system frequency so I am curious about any methods that differ from what we are doing in North America.
The Magnetic Amplifier is between the pilot exciter (a permanent magnet generator) and the main exciter and sends a "feed forward" to the pilot oil governer so it reacts before the speed of the turbo alternator slows.
The basics seems simple enough - the system was long (too much supply) during this period above and beyond what is contractually ordered between generators and electricity suppliers (eg, to consumers).
Having a negative price would seem to make sense if your goal is to keep consumption balanced with generation on an open market.
Theoretically yes, though I guess the details are a little more complicated (depending on who exactly bought the energy and under what terms). e.g. energy company x might have bought all the capacity from wind farm y for the next 10 years, so then the fact that it's producing excess energy at that point in time becomes x's problem rather than y's. Presumably x would factor stuff like this in to their purchasing decision, so as long as the average price remains above some level they will still make a profit.
Or bitcoins will simply be devalued, and lose buying power until electrical demand changes or all of the last possible bitcoins are finally mined into existence.
So if you operate a windmill in your backyard while prices are negative the power company will send you a bill? Why couldn't the power company just break the circuit if they couldn't handle the incoming power? The windmill would still spin but not generate any power, like a gas generator with nothing plugged in. Seems better than sending the windmill operator a bill so the power company can generate a bunch of steam 100s of miles away...
Not realy. There are TWO markets for electricity. One is the Day Ahead system, where producers and consumers agree on who runs what generators and who consumes how much, in 24 hours.
The real time market, meanwhile, covers who it is who has to adjust his production or consumption to cover the difference between what everyone thought would happen 24 hours ago and what happens now.
The real time price hits negative, but it's only the delta between real time volume and day-ahead volume that's traded at the negative price.
And it's the people who are turning knobs, being paid by those who are sticking to what they planned to do at the Day Ahead market.
Here's how it works: Generators put out bids for how much they can produce and the price. Consumers pull in the bids ordered by price until they have enough power.
After that they pay everyone the highest price for the power.
So making your power negative just means you want to be first on the list.
You don't actually get paid negative money. It's just a tactic.
If demand is low then the highest price can be negative.
When prices are negative Generators have to pay for the power they produced, for the privilege of not shutting down their process be it nuclear, coal etc. presumably hydro and natural gas would shut down in response to negative prices.
The article is about this just happening in the UK. For 5 consecutive hours the price was minus $25.
It also happens in Ontario, and maybe elsewhere but I don't know as much about how other markets deal with oversupply.
Wind generators generally have fixed price contracts for as much as they can produce so they carry on generating even if the price is negative and they get the same positive price. Or if they do have to shut down for system reliability they will get paid even though they didn't generate any power. That is called being constrained off.
If the price is negative it is because generators would rather pay than go offline. The nuclear plants in Ontario will bid negative because they have real costs if they have to alter their output.
I maintain that the price is negative because the highest selected bid was negative, and any generators producing during those 5 hours had to pay $25 for every MWH they produced.
are you in the power business in the UK?
I used to work on software that implemented the market settlement rules in Ontario in order to calculate a distribution utility's cost of power every day, and I've also spoken at length with an economist who wrote market rules, and examined the behavior of the market participants for compliance with the rules, but it is possible that the uk and Ontario markets differ in this regard, although the article indicates the price was negative.
if you had an extension cord running from your gas generator to your neighbors house and he doesn't want anymore power, couldn't he just unplug on his end? What kind of maintenance costs are we talking about? I can understand if a tornado knocks down your power lines, but how does the amount of power on those lines increase costs? And if the person on the other end unplugs isn't there no electricity running on those lines anyways?
If your extension cord had the amount of copper a mains line had it would have cost thousands of dollars. Now imagine if you ran these extension cords through to every neighbor in a city. And then there is a storm that takes some of those extension cords out. Then you have to have a truck with a cherry picker to get up there and fix them. Well then you need to hire people to do that. Then you need insurance to cover them. Etc.
I don't understand how negative power prices can exist? Why wouldn't everyone just dump the power into the ground and make a profit? Surely 0 has to be a floor.
You could think of the problem in reverse: there's an excess of energy and they need to pay people to get rid of it (because using that much energy isn't as easy as dumping it in the ground).
If you could find a good way of getting rid of a lot of excess energy (worlds biggest kettle?) then you could potentially make a bit of a profit! Until someone comes along and finds something useful to do with the energy...
There are a lot of grid-related engineering problems that come from increasing decentralization (e.g. wind, solar, hydro) because it increases unpredictability in supply.
https://www.amazon.com/Grid-Fraying-Between-Americans-Energy...