Storage can replace gas in our electricity networks and boost renewables

The Conversation

Renewables could benefit from more energy storage capacity in the electricity network. reupa/flickr, CC BY-NC
Renewables could benefit from more energy storage capacity in the electricity network. reupa/flickr, CC BY-NC

Energy storage could replace peak gas in our electricity network. That’s the finding of a study that my colleagues and I recently published in the Journal of Applied Energy.

Energy storage is often considered the holy grail of the electricity sector. Tesla’s Powerwall home battery system, for instance, allows households to store energy from solar panels, to be used when the sun isn’t shining. It is seen as a vital piece of the puzzle in a future with more renewable energy.

Storage is great for households, but could also be as important in the wider electricity network. Here’s how it could work.

Volatile prices

Generators or power stations sell their electricity on a wholesale market (in eastern Australia this is the National Electricity Market or NEM). From there it is passed onto households and businesses by retailers at retail prices. The wholesale price is a significant factor in the cost of electricity (other factors include poles and wires).

The wholesale price varies throughout the day – sometimes quite considerably, as you can see in the chart below from Queensland. In times of peak demand, prices can skyrocket to 300-400 times the average price.

Half-hourly wholesale electricity prices in Queensland, at the beginning of this year. The average price for the full 2014-2015 financial year was about $50/MWh. (Author provided, data from AEMO)

This volatility is largely a result of physics: generators have to match demand instantaneously, because electrical energy can’t directly be stored.

People don’t use electricity equally throughout the day. Usually electricity use is concentrated at the end of the day, or on the very hottest day of summer when people fire up their air conditioners.

Electricity networks are typically set up to meet the maximum possible peak demand. They meet this demand with flexible generators such as open cycle gas turbines (which are quick to fire up and shut down, unlike generators such as coal-fired power stations). Such “peak” gas generators are typically used less than 5% of the time.

Load duration curve for the National Electricity Market in the 2008-09 financial year. Curve illustrates the percentage of time that the system is at or above a particular demand level. A large amount of capacity is required for small time periods throughout the year. Author provided, data from AEMO

These rapid variations in energy demand, along with outages of generators or transmission lines and generator bidding behaviour on the market, can result in highly volatile prices. This is where storage can play a role.

Energy can be stored as chemical energy (in the case of batteries), or in other ways such as gravitational potential energy (in the case of pumped hydro), to be used later to generate electricity when convenient.

These electricity storage technologies can also provide peak capacity. In our paper, we found that this was the main value of energy storage. In fact, peak capacity potential may turn out to offer greater value than other options for meeting peak demand.

Surprisingly, we found this value wasn’t affected by energy losses involved in storage (not all energy is recovered when released from storage).

Powering up with storage and renewables

Due to its high flexibility, gas is often considered to be an ideal partner for renewable energy, because it can pick up the slack when the sun isn’t shining or the wind isn’t blowing.

But as the share of renewable energy continues to expand, large-scale electricity storage offers a promising alternative to gas.

In fact, a study by the Australian Energy Market Operator suggested that significant energy storage was crucial to a 100% renewable energy system, in order to minimise costs while maintaining reliability and security standards.

Our research found that storage actually has a competitive advantage over gas when it comes to meeting peak demand. While both can provide peak capacity, storage can also gain extra revenue by taking advantage of smaller price differences that occur on a more frequent (such as daily) basis. When taking this into account, storage may already be cheaper than gas in meeting peak demand. New reports from the US estimate batteries could replace gas in 3-5 years.

Relative costs of providing capacity from an open cycle gas turbine (OCGT) and pumped hydro electric storage (PHES). The right most bar shows the cost of capacity when the revenue from daily arbitrage is taken into account. (Author provided)

Australia’s electricity system is currently oversupplied with capacity to generate electricity – by around 37%. As such, there appears to be no need for new capacity for the foreseeable future.

However, there may be demand for new storage capacity if older generators are withdrawn from the electricity network. Alternatively, the outlook for storage may improve as renewable energy generation is increased to meet mandated targets.

Increasing penetration of variable renewable energy will increase revenues for storage. In times of high generation output there will be more opportunities for storage owners to shop around for lower prices. This fluctuation between prices is already happening in South Australia.

In this way, storage and renewables may prove mutually beneficial.

Source: The Conversation. Reproduced with permission.The Conversation

 

Comments

3 responses to “Storage can replace gas in our electricity networks and boost renewables”

  1. James Hilden-Minton Avatar
    James Hilden-Minton

    The conclusion here is rather anticlimactic. The thought that because there is an oversupply of fossil generator batteries will just have to wait until these retire is rather depressing and misses the economic disruption at stake.

    Batteries continue to get cheaper perhaps declining 10% per year. So within just a few years batteries displace all new gas or oil peaking plants. But go out a few more years and new batteries are even cheaper than old peaking plants. Thus, there is a growing opportunity cost of waiting for old plants to age out.

    The capacity markets that keep peaking plants on life support need to be opened up to competition to batteries. Batteries can serve multiple purposes. So allowing batteries to participate in capacity markets need only cover part of the cost of the battery. Thus, batteries can come into this market much cheaper than what single purpose analysis might suggest. So once this capacity market is opened to batteried, the capacity payment can come down quickly. This will pull the plug on many peaking plants that actually provide very little value to the grid.

    I’d also point out that in many places utilization of peaking plants is very poor. For example, in the US CT NG plants realize a capacity factor of 4.8%, about 70 minutes per average day. Under 10% utilization such a plant may have an LCOE of 185$/MWh, but if the utilization drops to 5%, then LCOE explodes to 305$/MWh. So already batteries are quite competitive given a condition of oversupply/underutilization. SolarCity has a PPA at 145$/MWh for a dispatchable solar facility with 17 MW of solar and 52 MWh of storage.

    So the conclusion I draw from this is that batteries are already competitive with peaking power plants and the gap will continue to widen. Moreover there is already an oversupply of capacity manifesting very poor economics due to underutilization. Thus, these assets are already over valued and increasingly obsolete. The solution is to impair the poorest performing assets and make space to new technologies to compete. Delaying entry of batteries into this market only serves to protect incumbent asset owners at the expense of all other grid participants. This is an asset bubble, and situation will not get better until write downs happen.

  2. Ian Avatar
    Ian

    Looking at the load duration curve: Energy would be the area under the curve or the product of load x time. The amount of fossil energy saved by getting rid of the peaking gas generators is minuscule. Is substituting these with batteries really worth doing when the base load coal stations keep chugging along supplying most of the average load? I guess what the author is saying is why not finance battery storage with those very lucrative spikes in demand. Batteries can serve a three fold purpose, peak shaving, bulk storage and instantaneous, high capacity discharge to supply high energy spikes. Those interested in renewables should be lobbying to allow behind the meter battery storage to participate in supplying high capacity peaking services in an active role or, at the very least, force utilities to drop connection fees and allow them to charge a fee according to peak capacity demand. This would encourage people with large variations in their electricity demand to install batteries and peak demand limiting devices at the meter. At present Ergon is charging a connection fee of about $1.50 per day. This should be scrapped and a peak demand fee of roughly the same value applied to the average home user’s peak demand. Elderly pensioners will pay next to nothing if they limit their peak demand and the occasional holiday home owner or aircon user will pay a lot for having electricity capacity available. Those with batteries and load limiting devices can manage their demand so that no peaks occur. Instead of the $ 550/ year connection fee discouraging solar and storage, An amount approaching that for peak usage can offset the cost of solar and battery storage. If the networks do not agree to this change in tariff structure then people will leave the grid completely and the possibility will be lost of utilising behind the meter storage for high capacity peaking ( the large spikes in demand in the article’s graph) .

    Put another way the potential of 500 000 houses installing 3KWH batteries each with a peaking capacity of 6 kW for 1/2 an hour could smooth the worst of network demand peaks very effectively at very little cost to the networks and electricity retailers. Any old fool can pump out electricity at 5 or 10c a KWH constantly, but providing load levelling and peaking capacity that’s another matter entirely!

    1. Ian Avatar
      Ian

      Another thought, Western Australia is saddled with capacity payments to gas generators, many of whom have never run their turbines. The government can manipulate its tariff structures by encouraging solar ,wind and battery storage development to provide a secure supply to its people, then shut down and mothball the coal fired plants, this will force these gas peaking operators to generate power at much lower and less expensive peaks. Running a gas peaking plant at full capacity 24 /7 at 5 or 10 c/KWH will soon burn out the generator or send its owners bankrupt!

Get up to 3 quotes from pre-vetted solar (and battery) installers.