Uncertain market

Filling in for wind and solar over days of calm and cloudy weather is not among the requirements that storage could potentially help meet

…

Understanding how electricity storage can competitively add value to power system management is key to assessing if and where there is a market for it as the transition to renewable energy progresses. Storage can potentially increase the range of options for meeting any or all of four principal requirements for reliable supply: provision of bulk power to make up for deficits in the variable output of solar and wind; system services (frequency response, reserves, voltage support) to provide increased flexibility in power system operation; management of variability to reduce price peaks and thus the cost of matching supply and demand; and easement of congestion on the network to reduce curtailment of green power production and defer spending on grid expansion. See a table of types, capabilities and principal uses here

Each principal use for storage can be further subdivided and within the four main types up to 15 separate applications can be identified. In all cases, options other than storage can meet the required need and that will not change much when transitioning to 100% renewable energy.

….

Essentially, there are six kinds of storage to potentially call upon: pumped hydro (versatile but geographically limited), compressed air (complex), various battery types (mostly very short-term only), flywheels (technically limited), heat/thermal storage (poor conversion efficiency) and power-to-gas (also poor conversion efficiency). To date, only pumped hydro is used at grid scale to regularly store and discharge large volumes of electricity.

… … … …

BULK POWER SUPPLY

No technology exists that can be widely deployed across all power systems to affordably store volumes of electricity sufficient to meet demand over consecutive days devoid of wind or sun. Should wind and solar on a power system reliant on their supply be insufficient to meet demand over days, a mixture of various alternatives would need to fill the gap. Among these, stepping up generation from other renewables, such as biomass or geothermal, buying green power from a neighbouring system, and triggering agreements to reduce demand are among the main options. Calling on some non-renewable thermal generation, most likely gas, might be necessary.

Some forms of storage may also be brought into play should cheaper alternatives run short. Types of storage with the right technical attributes for bulk power supply are pumped hydro, compressed air energy storage (CAES), flow batteries, power-to-gas, and heat stored as hot water, hot rocks, or in another medium, for conversion back to electricity using a steam turbine. Given sufficient land area, a large number of lithium-ion batteries could potentially also meet grid demand for an hour or two. Their falling price make them close to being suitable for this purpose.

Meantime, pumped hydro, well established for years, is the only storage technology capable of storing large amounts of electricity at a cost that makes it viable for several applications. Even so, it has had difficulty finding an economic footing in markets with growing volumes of renewable energy. In Germany, demand for reserve capacity has fallen as the proportion of renewable energy rises, reducing the market for pumped hydro and lowering the cost of managing variability. A number of local pumped hydro facilities in Germany, built in the expectation of a growing need for their services, have quietly closed. The German example is not unique. Texas is also noting a fall in spending on system services as its renewable energy supply increases.

To a lesser extent CAES can supply bulk power, but has technical and economic challenges to contend with and is not in widespread use (read more about CAES). Converting heat back into electricity is not an efficient energy cycle, which probably accounts for why it has not been widely adopted. Meantime, the versatility offered by flow batteries means they could make a contribution to bulk power supply, should their technical challenges be overcome.

……

MANAGEMENT OF VARIABILITY

Historically, storage has been perceived as having a role in “firming” the supply of wind and solar power on an electricity network that has significant proportions of both. The purpose is to reduce the variability of output from wind or PV installations. Such firming can take place either at the individual wind farm or at the level of the power system. Pumped hydro, CAES, thermal storage and power-to-gas all have the technical ability to provide firming, batteries less so, if at all.

The question, however, is whether so-called firm power is essential for reliable electricity supply, or even desirable. Making wind and solar behave in the same way as traditional generation serves no obvious purpose and may not be economically achievable.

Seen from the perspective of the grid, no storage system is capable of entirely firming the supply of a variable renewable energy resource. A real world example from Denmark demonstrates the point. On February 9, 2016 at 19:00, wind output from 3800 MW of capacity dropped below 1000 MW. It remained below 1000 MW for four days. When the calm period began, consumer demand would have been falling, obviating the need to boost production from the system’s gas and coal plant. To cover the next day’s morning and evening peaks, more power would likely have been needed, a pattern repeated over the next three days.

If storage had been used to boost production, assuming sufficient storage capacity, it would only have been for a few hours each day. As a capital-intensive technology, storage needs to be used intensively to pay for itself. The incremental cost of supplying electricity for just a few hours soon becomes unaffordable, as well as uncompetitive compared with using the existing fossil fuel capacity. For the green energy transition to be both fast and affordable, managing variability by occasional use, for short periods, of some of the considerable volume of thermal generation already in place can be a compromise worth making.

… …

TOO MUCH OF A GOOD THING

When renewable generators are producing more electricity than there is immediate demand for, storing the power rather than curtailing production is an obvious option. But only a limited amount of the energy from a long period of strong winds can be accommodated in a store. Curtailment of output is likely to be cheaper than paying for more storage, even when compensation is paid to the curtailed generator for lost revenue. Power system operators have for years curtailed all kinds of generation to help manage supply and demand variability. They are comfortable with compensating generators and it is part of the reliability of service that consumers pay for. Increasing the cost of that service by storing power for which there may be no or little need is not the foundation for a future market.

Storage at the individual wind farm level may in some circumstances be worthwhile, but the added value could be small, as demonstrated by the prices achieved in power auctions in the UK conducted by the Non Fossil Fuel Purchasing Agency. The difference in prices between firm power (such as landfill gas) and variable power (such as wind power) rarely exceeds £5/MWh, a long way short of the €40/MWh break-even point for stored power.

In some locations capacity firming to support the electricity network may be so useful that it has enough value to make storage commercial. The United States Department of Energy Storage Database lists over 200 projects where this takes place, often in remote areas. The value to the network lies in the system having access to the storage capacity and the stored power. Going off grid does neither grid customers nor individual consumers any favours, as pointed out in a 2014 report from Britain’s Imperial College, Can Storage Help Reduce the Cost of a Future UK Electricity System? “Distributed storage at the household level with no interaction with the network is neither the most economically attractive solution for end users, nor most beneficial to the network,” it states. •

TEXT Lyn Harrison & David Milborrow

This article is part two of our five-part special report on grid-scale electricity storage. Find parts one-five linked below or get the key takeaways at a glance