Advanced battery storage for UK’s National Grid20 April 2018
Renewables have less inertia than the large turbines used by coal, gas and nuclear, meaning grid frequency changes faster when demand and supply are out of balance. Advancements in battery storage technology and increased competition could be the answer to the UK’s increasingly inflexible and intermittent electricity supply, as James Lawson reports.
With more than 25% of UK generation capacity now coming from renewables, electricity supply is less flexible and more intermittent than ever before. That makes balancing the grid challenging, strains local distribution networks and, worse, impedes further renewable deployment. Batteries increasingly look like the solution to all these problems.
National Grid is the UK’s transmission system operator (TSO), responsible for maintaining voltage and frequency within tightly defined limits. To do that, it needs to access flexible generation in near real time, contracting in advance for more than 20 different ancillary services, including short-term reserve and voltage control.
“The bulk of power is still traded in long-term contracts, but ancillary services that deal with power quality are becoming more meaningful and important,” says Aris Karcanias, senior managing director at FTI Consulting.
Because solar and wind have far less inertia than the large turbines used by coal, gas and nuclear, grid frequency changes faster when demand and supply are out of balance. Too rapid, and that trips protection relays at the generation plant. Really fast frequency changes have the potential to uncontrollably disconnect large chunks of generation, causing parts of the grid to shut down.
Enhanced frequency response
Balancing today’s grid requires new services to cope. The first of these services to appear, enhanced frequency response (EFR), rapidly pumps large quantities of energy into the grid in order to maintain 50Hz.
“You want less than 500ms response time for EFR,” says Karcanias. “Lithium ion (Li-ion) batteries can provide a 20ms response and can also participate in bilateral deals for capacity markets.”
Published in mid-2016, the EFR tender’s attractive rates have encouraged investors to sign up for storage projects in droves. From a standing start a few years ago, the UK’s utility-scale storage now totals over 100MW, with gigawatts in the pipeline. Though hybrid flywheels, demand-side response and other innovative technologies were in evidence, 888MW of the 1,137MW that pre-qualified for the EFR tender was for Li-ion battery projects.
E.ON’s 10MW battery located at Sheffield’s Blackburn Meadows biomass plant became the first EFR project site to go live this October. Others include the 20MW Broxburn and 35MW Port of Tyne batteries, both built by RES.
The latter system is currently the UK’s largest, and these projects graphically demonstrate how the EFR tender has brought UK battery storage into the mainstream. RES brought its experience of US utility-scale deployments to the party, along with its proprietary Resolve control system.
“We’ve worked with National Grid since 2014,” says Tim French, development director at RES Group. “We shared our expertise from the US and helped them figure out what the new EFR service was going to look like.”
The UK’s 14 distribution network operators (DNOs) are the other big potential infrastructure storage customers. Shifting use patterns means local networks designed decades ago no longer satisfy today’s electricity generation and consumption requirements. Normally, DNOs are forced to put in extremely expensive new lines and equipment to cope.
“If you have a constraint locally, you can put in a battery instead, and avoid spending millions of pounds in network upgrades,” says French.
The 33/11kV Leighton Buzzard primary substation is a good example, needing extra capacity to cope with peak winter demand. Instead of installing a third transformer and approximately 20km of underground cable, the Smarter Network Storage demonstration project installed a 6/10MWh grid battery that successfully provided peak shaving support for up to 1.5 hours at a time.
Batteries are also today’s default choice for behind-the-meter storage, where generation and consumption are co-located. Businesses large and small, as well as consumers, can time-shift electricity for their own use, sell it back to the grid and participate in energy markets as merchants, using techniques like arbitrage: storing cheap electricity and then selling it at higher prices during peak demand hours.
Time-shifting is what Cornwall’s Olde House project was built to do. Storage vendor Red T worked with Centrica to attach a 1MWh flow device to an existing 350kW solar array. Solar generation and cheap electricity charges the battery, which then maintains power 24/7 to the houses on-site. Because they only import at off-peak times, it also lessens the load on Cornwall’s weak grid.
“They can sell excess power to the grid too, but with today’s low feed-in tariff, it’s usually better value to store and use it themselves,” says Joe Worthington, Red T’s communications manager. “They save a third on imports from the grid, increased their solar utilisation by 1,800% and now get a 10% return on their asset investment.”
Pairing legacy solar with battery storage is becoming common. Anesco’s new 10MW Clayhill solar farm shares an existing grid connection and uses a 6MW battery to support grid services and arbitrage, making the project profitable without fresh subsidy. In contrast, there’s little storage activity in UK wind as yet.
Where solar production has predictable nightly troughs, a turbine might generate for a week straight and then produce nothing for three days. That requires a larger, more expensive battery that then often goes unused.“As wind is much more variable, it makes sizing the storage quite challenging,” explains Worthington. “That can make a project’s economics difficult.”
In 2016, Red T ran a trial on the Scottish island of Gigha with its 1.68MWh flow battery taking a feed from the three community-owned turbines. This stabilised the islanders’ own supply and allowed export when the locally constrained grid could accept it. Vattenfall, another EFR winner, will complete its 22MW Pen y Cymoedd battery in February 2018. To lower capital costs, it again shares its grid connection with the 228MW wind farm next door but the battery is not tied to the farm’s output.
The 2MW battery that Dong Energy is installing at its 90MW Burbo Bank wind farm is the only current example of UK wind storage. Delivering EFR services is its main purpose, but the battery and wind farm will work together in delivering those services, in addition to supplying better quality power to the grid.
The capacity market
From peak shaving to black start and demand reduction, there are many markets and ancillary services that storage can cater for. That’s just as well. With so much new battery capacity in the pipeline, the EFR market is getting crowded and rates look likely to fall.
UK developers will need to look elsewhere for income and the capacity market is the obvious one. The Port of Tyne battery is a good example, servicing a 12-year capacity market contract in addition to EFR.
“At the moment, most projects are frequency response but as you see more batteries being built, that market is finite,” says French. “There’s over 3GW in the planning system. They will want to participate in merchant services, ancillary services, and the balancing and peaking markets, using the battery for different services at different times of day.”
To handle the differing demands of these multiple services, the batteries need to be versatile enough to cope. Control systems like Resolve are part of the answer − but what about the batteries themselves?
Stress events in the capacity market may last up to four hours, considerably longer than the 30 minutes to an hour discharge time that is all most UK batteries can handle. Because of this, the UK’s Department for Business, Energy & Industrial Strategy (BEIS) is currently considering ‘derating’ smaller batteries.
The subsequent drop in revenue is worrying developers such as Anesco, which is reportedly investigating moving to flow battery technology to service the capacity market.
Redox flow batteries consist of two tanks of vanadium solution. When pumped into a reactor, the two solutions flow adjacent to each other past a membrane, and generate a current as electrons move back and forth during charging and discharging. They are more bulky than Li-ion: 250kW of storage approximates to one shipping container.
“Technologies like flywheels, capacitors and lithium are high power, they deliver energy very quickly,” explains Worthington. “Flow machines store a lot of energy and you can discharge them all day. They are close to cost parity with lithium and, if you take lifetime into account, ours are significantly cheaper over 20 years.”
Li-ion’s tendency to degrade when subjected to high cycles and repeated deep discharge was another concern for the BEIS. By contrast, vanadium flow batteries can be left completely discharged for long periods with no ill effects and have a ‘semi-infinite’ lifetime.
“Longer-duration batteries may be more effective for certain solutions,” notes French. “It depends on the revenue model. But where you need black start capability or back-up, you need high energy density. That’s what we’ve building in the US and Li-ion is very good for that.”
Red T is currently developing a hybrid battery that combines lithium and flow technologies. Vanadium makes up 80% of the capacity while 20% lithium provides high power for short durations. “With occasional managed use, the lithium lasts a lot longer,” says Worthington.
As renewable penetration continues to increase, more batteries to support the grid and smooth intermittency look likely. But although there have been recent encouraging moves, like continuing renewable subsidy support for batteries co-located with solar, the way ahead isn’t entirely clear.
Many batteries are too small to participate economically in energy markets. DNOs will take years to connect the gigawatts of proposed storage across many distributed sites. Policy and regulation that would allow tactics like aggregation are still absent, while research funding is overly focused on vehicle batteries.
With that degree of market immaturity, it’s hard to know which technologies will still be here in a decade’s time. But with Li-ion batteries now so powerful, cheap and compact, don’t bet against them still ruling the roost in 2027.
“Lithium is the main technology choice, because it’s established and proven,” says French. “Like PV, Li-ion will continually evolve to be faster, and have higher energy density and greater longevity.”
UK moves to low carbon power in 2017
From 21 June to 22 September 2017, almost 52% of UK electricity needs were met by low-carbon sources, including solar, wind and nuclear. In comparison, the same period in 2013 saw just 35% of energy needs coming from low-carbon generation.
“It’s been an exciting year managing the many network firsts – from a day where we operated the system with zero coal power, to one where over half of the UK’s energy demand was met by renewable generation,” says Duncan Burt, director at National Grid.
Tesla on track to build world’s biggest battery
Tesla’s “big battery” in South Australia was recently brought online. CEO Elon Musk committed to building the 100MW/129MW/h battery storage project next to the Hornsdale wind farm in 100 days earlier this year.
Part of the capacity will be contracted to the South Australian Government for grid security needs while developer Neoen will employ the rest – around 30MW/90MWh – as an arbitrage supplier for the wholesale electricity market.
“We expect this project to lay the groundwork for many similar projects, but at an even larger scale, in the years ahead,” Musk said.