The fifth anniversary of UK grid-scale storage
UKPN’s Leighton Buzzard 6MW/10MWh battery was the UK’s first grid-scale battery storage facility. It went live at the end of 2014 under the Smarter Network Storage innovation project. Grid-scale storage has now grown to 1GW since this first project five years ago. To-date the principal commercial service that these assets provide is that of frequency response; helping the grid to maintain the balance of supply and demand in real time.
Much has been written over the last 18 months on how the economic aspects of the Frequency Response (FR) markets have changed, however the operational utilisation of grid-scale storage within the Dynamic Firm Frequency Response (DFFR) market has also evolved since the Leighton Buzzard battery went live.
The main earner for grid-scale batteries; Dynamic Firm Frequency Response
Storage that provides DFFR must provide a power response as frequency deviates from the deadband (49.985Hz — 50.015Hz). Figure 1 shows the characteristic of this response, known as the droop curve.
Figure 1 — Frequency Response droop curve for a dynamic provider.
The more extreme or frequent the deviation of grid frequency over time, the harder the battery has to work at importing and exporting power, as determined by the droop curve. In general, the harder the battery works, the faster the available storage capacity degrades overtime. Battery manufacturers typically warrant that given a rate of energy throughput per year, the battery storage capacity will degrade at a known rate per year, and no more.
But there is a problem with the idea of providing a warranty based on a fixed annual energy throughput. Grid inertia has become increasingly time-varying due to the growth in renewable generation coupled with the retiring of large thermal generation. As non-synchronous generation (primarily wind, solar and interconnector imports) has increased over the past 10 years, the inertia of the system (its resistance to a change in frequency, arising mainly from the kinetic energy in heavy rotating turbines) has decreased. The monthly average percentage of synchronous generation on the GB system over the past 8 years is shown in Figure 2. As a direct consequence of this transition, the volatility of frequency has increased over time which leads to a higher Rate of Change of Frequency (RoCof). Figure 3 below shows the upward trend in variance of frequency overtime. The increase in frequency volatility leads to batteries (particularly shorter duration ones) reaching higher depths of discharge exacerbating degradation.
Figure 2 — Percentage of synchronous generation on the system from start of 2011 to end of 2019. Aurora Energy Research
Figure 3 — Daily Variance of UK grid frequency from start of 2015 to end of 2019
The decrease in inertia has meant that batteries are generally working harder at the end of 2019 than they were in 2015. Figure 4 shows the daily energy throughput for a 1C storage facility supplying a 24×7 DFFR service from 2015 to the end of 2019.
Figure 4 — Daily energy throughout of 1C battery providing 24 hour PSH DFFR
Due to the increased workload, the amount of time a DFFR battery is idle as the frequency is in the deadband (see Figure 1), has decreased over time. Figure 4 shows the monthly rolling mean of the time the battery is spent within the deadband:
Figure 5 — Time a 1C battery supplying 24 hour PSH DFFR spends in the deadband between 2015 and 2019.
The amount of time a battery spends in the deadband has fallen since 2015 and when outside the deadband frequency is further away, again resulting in higher depths of discharge. Additionally, since 2015 grid-scale storage started supplying FFR it has seen increasing volumes of import and export energy. Figure 6 shows how both daily net imports and exports have become more significant over time.
Figure 6–1C battery supplying 24 hour PSH DFFR — import v export.
This could be because as energy throughput has increased overtime, and because storage efficiency is less than 100% the amount of time the battery is importing has increased.
In summary, DFFR contracted grid-scale battery storage systems are now working nearly twice as hard, since the first commercial grid-scale storage system went into operation five years ago. This is likely to be a combination of the changing generation mix reducing inertia on the Grid. National Grid’s new frequency response products seek to address these issues through the requirement of faster response times and providing a more dynamic service.
How can Upside help with the operational effectiveness of battery storage?
Upside Energy is able to help asset owners and maintenance operators monitor the utilisation of their asset when running a DFFR contract or any other storage commercial services. The Upside Asset Management System (AMS) provides real-time diagnostic and monitoring tools for storage and distributed generation asset owners. These services are aimed at measuring asset storage performance in order to ensure the battery is performing in-line with manufacturer warranty conditions, and also to help quickly diagnose and alert operators to any unforeseen operational incidents. Real-time user defined alerts can be established to alert different user types.