Renewables combined with storage play a key role in grid stability. (Image: iStock)
A few weeks ago, the national grid of Spain disconnected from the rest of Europe, resulting in a major blackout. In March 2022, Taiwan’s national grid was also separated into two, and all conventional power plants in the southern grid failed, requiring a black start.
How can these types of events occur? And how can renewables and batteries provide more comprehensive grid services to avoid similar events in the coming years? This is what this article is going to discuss.
Frequency oscillations on the power grid
Under normal operational conditions, the grid frequency is kept the same throughout the entire continental Europe, indicating a balance between aggregated supply and demand. However, when the connection becomes too weak or the power imbalance becomes too large across nations, the frequency of different national grids can begin to deviate from each other. One common phenomenon (which also occurred prior to the blackout in Spain) is frequency oscillation.
Frequency oscillation is a behavior one can derive from the swing equation, the mathematical formulation of the law of energy conservation in power system engineering. For better understanding, we can use a mechanical analogy to explain why an oscillation would occur:
- First, as shown in (a), consider a ball connected to the ceiling. In our analogy, the ball is a national grid with a weak connection to the rest of the continental grid, while the ceiling is the rest of the continental grid. The ball is at equilibrium, which is analogous to the national grid operating under normal conditions.
- Then, as in (b), we apply a small tangential displacement to the ball. In our analogy, this displacement is a perturbation of the phase angle of the weakly connected national grid.
- As a result, the ball in the mechanical system begins to swing back and forth, as in (c). In our analogy, this would be the cause for frequency oscillations.
The mechanical analogy of frequency oscillations. (a) The equilibrium state without perturbations. (b) A small perturbation is applied. (c) The ball swings back and forth.
From the analogy above, we can see that factors affecting the stability of the power system can include: inertia (the capability to resist sudden changes in frequency) among national grids, the size of the perturbation, and the frequency control after the perturbation. They will determine whether the power system can eventually return to normal or the amplitude of the oscillation becomes too large such that various national grids can no longer remain synchronized with each other.
Batteries: the conductor of the power system orchestra
Well then, how can renewable energy and batteries deal with various grid stability issues such as the aforementioned frequency oscillations?
Most renewable energy power plants apply "grid-following" control when connected to the grid: these power plants measure the magnitude and phase angle of voltage on the grid and set their reference current output accordingly to maximize renewable energy fed into the grid. Currently, some renewable energy power plants and batteries can perform "grid-supporting" control, which dynamically adjust their reference power output according to the measured voltage magnitude and grid frequency. This allows them to provide synthetic inertia, fast frequency response, or other frequency and voltage services.
As more and more renewables and batteries are installed, requiring some of the batteries to perform "grid-forming" control will be inevitable in a power system dominated by renewable energy[1]. While other grid-following power plants only measure the magnitude and phase angle of voltage at the point of connection, power plants adopting grid-forming control will keep the magnitude and phase angle of their voltage output near a reference point, thereby becoming a voltage source in the power grid. During a contingency, these grid-forming power plants keep the magnitude and phase angle of grid voltage stable, allowing other grid-following power plants to continue to operate.
We can use a musical analogy to elaborate further: if the power system is an orchestra, then the batteries adopting grid-forming control are equivalent to the conductor of the orchestra, whose job is to provide stable and consistent signals to the members in the orchestra (grid-following renewable energy power plants and batteries) to follow. The conductor themselves might not perform music (just as batteries cannot generate electricity on their own), but without them, the orchestra cannot function as a whole.
It should be stressed here that inverter-based renewable energy and batteries can provide faster and more accurate flexibility[2] compared with conventional power plants. Therefore, either performing grid-supporting or grid-forming control, they can make the power grid more stable. This has been proven in grids with very high share of renewables, for example, South Australia and Hawaii. In these examples, after deploying batteries capable of providing synthetic inertia, fast frequency response, or grid-forming control, the responses of the power system during contingencies with similar levels of frequency perturbations become faster and more effective.
Renewable energy and batteries are usually scattered in the distribution grid, with less physical inertia but also faster control, similar to a fleet of cyclists. Conventional power plants on the other hand are more centralized, with more physical inertia but also slower control, similar to heavy goods vehicles.
Another point that should be stressed is that for the power system, more inertia and voltage sources are not always better. Returning to the mechanical analogy in the previous section, a power system with a large inertia but ineffective frequency control is similar to a wrecking ball swinging back and forth with little air drag; this obviously is detrimental to the stability of the system (see the figure below). In the meantime, too many voltage sources within the same area can also cause local current instability issues[3]. Therefore, a more comprehensive strategy is to harmonize the inertia and control parameters in the system and to balance the amount of grid-forming and grid-following inverters, such that the power system is always controllable under any plausible circumstances.
Greater inertia does not always help stabilize the grid. When other conditions are the same, a system with smaller inertia (real line) can neutralize perturbations faster than a system with larger inertia (dashed line).
In recent years, as more variable renewables are deployed onto the grid and more conventional power plants are being phased out, people often focus on metrics where grid-following renewable energy power plants are inferior to conventional power plants. While discussions on retrofitting conventional power plants into synchronous compensators and adding grid-forming inverters can directly address these concerns, we should also not neglect metrics where renewable energy and batteries are superior to conventional power plants. For example, renewable energy power plants and batteries, when performing grid-following control, need not worry about phase angle instability issues. This means that we should not request all batteries to provide synthetic inertia or grid-forming control; instead, we should strategically set up stricter connection rules and deploy flexibility resources at weakly-connected or crucial locations.
The future of electricity supply is renewables, and the future of grid stability is batteries
The continual growth of renewables has long made them the main protagonist in energy transition. At their side, batteries act as the crucial supporting role that will stabilize the power system.
As the global energy transition unfolds, blackouts will inevitably occur in parts of the world with high renewable energy shares. Neglecting the more common examples where renewables and batteries are controlled adequately such that blackouts are avoided or mitigated, it could be easy for energy transition skeptics to conclude falsely that “renewables can only threaten the grid”.
For example, before the blackout in Spain, there were no known issues with electricity supply from different sources. When almost all the conventional power plants disconnected from the national grid during the first phase of the blackout, it was some of the renewable power plants which remained on-line, so a complete blackout could be avoided. Finally, after the blackout, the power output of wind and solar power plants returned to normal in no more than two days, when nuclear power plants were just beginning to ramp up their power output from scratch, a process that took more than a week to complete.
When the blackout stroke Spain, almost all conventional power plants were disconnected (nuclear power output dropped from 3.3 GW to zero, while gas power output dropped from 2.2 GW to 0.42 GW). Some renewable power plants remained on the grid, so a complete blackout was avoided.
Despite the empirical evidence that renewables are more stable and more resilient than conventional energy sources, energy transition skeptics still blamed solar energy as the culprit of the blackout as soon as it occurred. Interestingly enough, the same logic (that the technology with the highest share in the mix should be blamed) was not applied during the 2021 blackout of Texas, and renewables were still blamed even with a smaller share in the electricity mix back then.
The real message behind these narratives is that “no matter the share, renewables will threaten the grid, so we should stop deploying them”, which cannot be further from the truth. The technologies and regulations around renewables and batteries are advancing rapidly around the world; any event on the grid should not be an excuse to stop the development of renewables and batteries but rather the opportunity to speed it up. After all, new flexibility resources such as batteries, electric vehicles, and heat pumps are the best solutions to deal with these types of grid contingencies.
After the summer blackouts of California in 2020, they deployed almost 7 GW of batteries in 4 years; Texans also built renewables and batteries at a similar speed after 2021. Energy transition can and should be a common vision that transcends partisan politics.
In Taiwan, 5.5 GW of batteries are to be deployed by 2030, and we should start encouraging some of them to provide synthetic inertia and grid-forming control. As batteries provide more grid services, the need for physical inertia can be reduced, with the ultimate goal of allowing the grid to operate with 100% renewables.
[1] Renewable energy power plants can also provide grid-forming control, but this usually means additional power output curtailment, so existing batteries should be prioritized to provide such services.
[2] The speed and accuracy of control using inverters can be empirically proven from field tests such as those conducted by NREL, see Demonstration of Essential Reliability Services by a 300-MW Solar Photovoltaic Power Plant or Fast Grid Frequency Support from Distributed Energy Resources. In the meantime, batteries can provide more inertia to the power system compared with steam turbine power plants of the same nominal capacity, while their real and reactive power output control are not constrained by physical limits due to mechanical rotation.
[3] Discussion in Revisiting Grid-Forming and Grid-Following Inverters: A Duality Theory can be helpful to understand this concept. The premise that grid-forming inverters (voltage sources) are beneficial to the system lies in the assumption that most inverters are grid-following (current sources) at the location. However, if grid connection regulations require grid-forming inverters to be deployed everywhere, the benefits of increasing voltage stability might not worth the costs of reducing current stability.
Have insights on energy or carbon issues? Share your perspective with us! Send your submission to [email protected] for a chance to be featured. Submissions may be edited for clarity and style.
.jpg)