blog | enspired trading

Longer-duration battery storage

Written by enspired | Sep 17, 2024 12:51:49 PM

How do we categorize BESS duration?

Duration refers to how long the asset can supply power uninterruptedly before it requires recharging. The energy market is observing a progression toward longer-duration battery storage, specifically 4-hour systems. Today, most operational systems are 1-2 hours, and this developed in line with the market demand for short-duration assets driven by the need for fast-response frequency restoration services. These battery assets react quickly to signals from the market and are only required to respond for short periods of time. Read more about current BESS (battery energy storage system) sizing conventions here.

 

How do we distinguish between long and short durations?

An industry consensus has yet to be reached, but anything under 2 hours is generally considered short, while anything above 6 hours is long. So-called longer-duration systems are everything in between. Duration depends on a battery’s ratio of MW to MWh, and the market is currently gravitating toward the 4-hour solution. The sample configurations below both equate to a 4-hour duration:

Battery 1
Power rating: 20 MW
Energy capacity: 80 MWh
Duration: 80 MWh / 20 MW = 4 hours

Battery 2
Power rating: 50 MW
Energy capacity: 200 MWh
Duration: 200 MWh / 50 MW = 4 hours

De-rating
Batteries originally designed as 2-hour systems can be de-rated to meet 4-hour requirements. De-rating intentionally reduces the asset’s power output while maintaining the total energy capacity. For example, a battery with 20MW/40MWh (= 2-hr duration) can be de-rated to 10MW/40MWh to achieve a 4-hour duration.

 

What is driving the growing interest in longer-duration storage?

Both the UK and Germany were first movers in adopting energy storage for grid balancing; the UK had requirements for 15-30 mins maximum responses, whereas FCR (Frequency Containment Reserve) has long been the main ancillary service market for battery storage, requiring only 45 minutes of continuous output plus a 15-minute rest period despite operating in 4-hour blocks. Both markets have seen saturation, and, looking specifically at Germany, this sparked an interest in utilizing aFRR (automatic Frequency Restoration Reserve), which, due to its minimum 4-hour response time, was originally unavailable for BESS. The market structure then underwent a change in the summer of 2023 with the introduction of the energy product, allowing market participants to deliver in 15-minute windows. Those bidding into aFRR capacity (availability to deliver power – long-duration product) are obligated to also bid into aFRR energy (actual delivery of power – short-duration product). This enables them to market only the amount their BESS can supply instead of having to meet the minimum requirement of aFRR capacity with their asset. mFRR (manual Frequency Restoration Reserve) requires a minimum of 8 hours and is therefore not applicable for BESS (yet). aFRR favors short-duration systems, and ancillary service markets are generally geared to short-duration products. Once they saturate, prices get squeezed down, and the demand for short-duration systems lessens because they cannot capture enough value anymore.

As these ancillary markets are getting saturated, there is an increased need to look elsewhere for revenue. Wholesale markets were used to rebalance following ancillary service activation but are frequently being used for arbitrage (buying low, selling high) as BESS can time-shift energy to more profitable periods. As wholesale arbitrage is about moving volume, combining it with the optionality to target other markets (aFRR, mFRR, etc) makes a strong case for longer-duration systems limited only by system prices.

Building batteries, however, is becoming cheaper and cheaper (see more details in chapter 3.2 of IEA’s World Energy Outlook Report). Manufacturing is one influencing factor; the others are renewable penetration and profitability in the power markets. Cell prices of $50/kWh were not expected until 2030, yet we’ve already seen lower prices in 2024. Whilst this is for EV cell prices, the trend applies to the wider battery industry and, naturally, fuels investments into the technology.

 

Source: World Energy Outlook Report by IEA

 

Wholesale revenue potential for longer-duration storage

Many markets are now moving away from ancillary services and more toward wholesale. As both the UK and Germany have seen with the increasing build-out of BESS, the primary revenue streams (ancillary services) became saturated, and clearing prices declined. Wholesale markets offer additional value for BESS assets in periods when ancillary services may be less lucrative and serve as an alternative revenue source if the assets do not win any contracts.

Demand profiles usually have a morning and evening peak (high prices) as well as a midday and overnight trough (low prices). Occasionally, the market will experience low demand in the morning, eradicating the early peak. Still, a 2-cycle opportunity will typically exist for storage.

  • Low demand during day and night drives prices down, allowing storage assets to charge cheaply either to perform ancillary services at the same time or later in the day or to simply charge and then discharge over the next available peak in prices.
  • High demand typically occurs in the morning between 8-9 am, due to the macro effect of almost an entire population doing the same things at the same time. A similar scenario happens later in the day, during the evening peak around 5-6 pm.

With this regular cyclical nature defining the demand profile, wholesale markets can be optimized around opportunities to trade and perform necessary grid services. To explore wholesale markets and the opportunities in more depth, it is essential to understand the change in landscape as more renewables enter the grid, and how this influences prices and ancillary requirements across the day.

 

Renewable generation profiles and market locality

To contextualize the demand and supply challenges, let us look at some countries with contrasting weather and market conditions. The UK is the perfect example of why a 2-hour system makes sense. Wholesale value can be captured in certain windows throughout the day, and in between those periods, short-duration activities like ancillary services can be leveraged to offset the battery’s downtime. Low prices and renewable availability pose a challenge for value capture. Solar ramps typically follow a normal profile – you know when the sun rises and when it peaks. In the UK, this means 7 hours of peak solar generation dropping the prices down, likely to zero if not lower. On either side of that, you have the so-called solar shoulders, where you have solar either ramping down or not producing at all. This coincides with the demand peak in the morning and evening, when people get up and when they come back from work, creating a mismatched demand and supply scenario. The local climate impacts the battery revenue potential severely, of course. In the UK, you can capitalize on overnight wind, but there is relatively little sunlight. Batteries are designed to compensate for such renewable deficits in the future.

In markets with heavy use of renewables, such as Spain, longer-duration storage is becoming increasingly interesting due to the large delta between demand and supply. Conventionally, demand has been met by generation – gas followed a generation profile matching the demand profile. Now that we are using so many renewables, gas is required less, coal is being removed from most systems, and nuclear is being phased out. This means we are losing more flexibility than our system can withstand. If these gaps are not filled, people’s power usage has to be restricted, and this is exactly the type of load shedding we saw play out in South Africa. When their coal plants went offline for maintenance, they contracted enough batteries to fill the gap, but the systems couldn’t be built fast due to a delayed tender process, which left the country in a situation where household power consumption had to be regulated to avoid total blackouts.

Scenarios like this make the conversation around longer-duration storage more prevalent. Problems might arise depending on how these assets are used. Batteries like to be cycled; they are not designed to serve as passive power reserves waiting on standby for occasional shortfalls. If you fully charge an 8-hour battery and then just let it sit idle before catching the evening peak with a 1-2-hour discharge, you are repeating this pattern 4 days in a row before recharging again overnight. That is a lot of stagnation – a battery that is used only 20% of the time is not a sound business case. For this reason, the capacity market mechanism seems like an attractive revenue stream, but availability payments alone are still not enough to underpin the investment case. In the UK, for example, you can secure 30-40k per MW annually with your 8-hour system, but to break even, you need to generate 70-80k per MW. If the cost is such that you would have to generate 100k to make a profit, that is a shortfall you will struggle to compensate for in the markets if you are reserving capacity to receive availability payments. Capacity markets limit your optimization potential because you cannot market the reserved capacity, even if it never gets activated. The trick is to strategically combine all individual markets and revenue streams so that they collectively achieve the commercial optimum.

 

The importance of power market interconnection

Power markets in Europe arguably need stronger physical interconnection across geographical regions. If the UK, a country with relatively little sunshine, does not have enough power, being able to procure it seamlessly from Spain, a country with relatively heavy sunshine, would solve the issue. Conversely, on the off chance that Spain does not produce enough solar, they can import wind power from the UK. While a structure for intercountry energy trading is in place, the physical lines facilitating this exchange have much room for improvement.

Southeast Europe experienced severe heatwaves in the summer of 2024, impacting the energy production of conventional power plants. Household power consumption then increased because people were using air conditioning and fans. With conventional power plants unable to operate, affected countries had to call on the interconnections to fill the gaps in supply. These, in turn, were struggling to provide the required energy. If a situation like this occurs in multiple markets at the same time, the risk of blackouts increases significantly, especially if the infrastructure for cross-border electricity transfers is poor. And this is again where long-duration storage comes in.

 

What duration should a battery use case have to be profitable?

A large-scale rollout of long-duration storage depends on how quickly the transition to renewables progresses. The cannibalistic effect of renewables causes negative prices when there are generation surpluses. Batteries help mitigate this. However, the market today has not yet passed that decisive threshold where the majority of power production comes from renewable energy sources, which is why the 2-hour system remains the most reliable battery business case. 4-hour systems already demonstrate high compatibility for co-location and will find broader application as storage needs increase.

While 4-hour systems bridge the supply gap with their ability to provide short-duration services and use their MWhs for longer periods, they will be of even higher relevance in the future, in which wholesale dominance is expected universally among forecast providers. As system prices decline and 4-hour systems become more cost-competitive, much like the 1-hour vs. 2-hour conversation in 2022 and 2023, the focus now is on 2-hour vs. 4-hour. Building the longer duration system today may cost slightly more; however, as an investor, you may be better placed to capture market opportunity as it evolves away from ancillary service dominance – a factor most markets will experience.

 

The next question to logically arise is: why not 6- or 8-hour systems?

Whilst these systems will have a role to play in the future of the electricity system, they are likely to play a different role. The theoretical maximum of a 2-cycle/day asset is approximately 6 hours – 24 hours divided by 4 active events (2 charges + 2 discharges) = 6 hours. With this, there will be wasted hours when a net-zero result of a charge and discharge occurs due to identical wholesale prices. This makes the 4-hour system a more effective solution in the longer-duration model, as it can capture more value without having to utilize cycles to avoid energy stagnation.

Therefore, battery systems with a duration of 6 hours and above will have unique use cases due to their suitability for providing energy backup, reducing peak load during office hours, and time-shifting energy use to avoid peak charges. However, the exact function and monetization of a long-duration BESS asset depend on the owner’s bespoke use case.

 

Are you uncertain about how to dimension your battery project?

Let's find a solution together.