Power system without rotating inertia — Is it possible?

7 min readFeb 6, 2024

How much variable renewable energy (VRE) can enter the grid system? Is there any technical limitation? I have been asking this question for years but have yet to find a satisfying answer.

Since VREs, such as solar PV and wind turbines, are intermittent power plants, it is understandable that people are worried about the reliability of the electricity supply during the night or windless winter. At the same time, there is currently a significant research interest in energy storage, and storage technology costs are decreasing. Seeing this trend, I believe that a grid system fully powered by VRE is not a technical feat but rather an economical one.

However, as I’m trying to dive deep into the details, it seems that VRE intermittency is not what the grid operator is concerned about the most. Instead, it is the risk of VREs making the grid system unstable because they lack inertia.

What is inertia? (This is a simplified explanation for non-engineers; you can skip this part)

Grid operators must always keep a stable condition called power system stability. One of the stability criteria is maintaining a steady 50 Hz frequency (it’s 60 Hz if you live in America). When there is a sudden jump in electricity demand, the system’s frequency drops and requires some power plants to increase their power output, balancing the supply and demand. Without a timely response from the generators, the frequency will keep dropping, and cities will suffer from blackouts. How quickly this frequency fluctuates strongly depends on the grid system’s inertia.

The larger the grid system (larger aggregate capacity, more interconnected), the larger its inertia. And the larger the inertia, the slower its frequency fluctuations. A grid with large inertia will better manage the frequency fluctuations (therefore, more stable) since there is more extended time available for grid operators to control the power plant’s output.

If you’re still confused, I hope the following analogy helps. You’re a driver and have the task to keep the vehicle’s speed at exactly 50 km/h. A motorcycle has a small mass (thus small inertia), making it vulnerable to disturbance. Compared to a car with much larger inertia, if there is a bump, the motorcycle will slow down faster than a car.

How is this all related to VRE?

In a grid system, the source of inertia is the rotating mass of turbines and generators. In the event of a disturbance, conventional powerplants will keep rotating and won’t suddenly stop. This intrinsic characteristic gives sufficient time for the grid operator time to respond appropriately.

The problem: Solar PV and wind turbines are not connected to a generator, which converts shaft rotation into AC power output in conventional power plants. Instead, VRE (including batteries) are connected to an inverter, which converts the original DC power output into AC power output for the grid system. An inverter is a power electronic that operates discretely, zero or one, thus contributing no inertia to the grid system.

The problem is actually twofold. First, during the disturbance, the grid operator cannot expect VRE to help stabilize the grid since its output depends on weather conditions. Second, with an increasing share of the VRE, the system inertia is degrading as there are fewer and fewer rotating generators. While the first problem can be solved with more energy storage, can we solve the second one? In a system dominated by VRE, will the grid operator be able to stabilize any disturbance?

Can we run the grid with very low inertia?

Unfortunately, I don’t have the answer as this topic is actively researched worldwide. However, I got a glimpse of hope after I found this image.

This NREL’s chart shows a general trend of VRE penetration in several grid systems. In a nutshell, high VRE penetration is common in small grids. Meanwhile, VRE penetration is still low in most large country-scale grid systems. There are challenges, but it proves that having a low inertia grid is not impossible to conceive.

In a microgrid, typically in a remote village or a small island up to 1 MW in size, full VRE+ battery systems exist and are operating today. Meanwhile, increasing VRE penetration up to 25% of annual energy in larger grids is relatively “easy”, as we see examples such as Ireland, Denmark, and Germany. However, jumping to higher than 50% VRE penetration seems to be a much more complex problem. Note that these figures are annual energy, meaning there are several hours of the year when almost all demand is covered with VRE, even with only 25% VRE penetration.

Rotating mass is slow; power electronics are fast. Do we need inertia anyway?

I sensed that the power system would evolve similarly to the music/film industry, which experienced analog-to-digital transformation in the past. At the early introduction of digital technology, there was a view that analog recording had superior quality over digital. Even though it is true, it turned out that consumers do not really need that level of quality. It may be analogous to the perceived need to maintain grid inertia.

Synchronous generators are relatively slower to respond than inverters. After detecting a disturbance, the control system takes about 5–10 seconds to open up the valve/inject more fuel. The rotation energy of the turbine and generator acts as a buffer to hold the frequency while performing the stabilizing control (technically called “droop control”).

Meanwhile, the inverter could respond much faster in 10–100 milliseconds, a two-order magnitude faster! There is an emerging view that inverter-based generators are fast enough to handle the contingencies and, therefore, inertia is no longer a critical factor of grid stability. In fact, engineers are trying to develop inverters with grid service capability today.

The new advanced inverter is getting smarter

The commonly used inverters are “grid-following” inverters. As the name suggests, this type of inverter tries to mimic the grid’s frequency (while assuming the grid is perfectly stable) and inject power. When there are frequency disturbances, it helps nothing but instead disconnects from the grid to protect the powerplant.

Recently, a more advanced version of grid following inverter has the capability of emulating “synthetic inertia”. The inverter’s new control algorithm enables it to increase or decrease real power output to stabilize frequency immediately. To do this, it must sacrifice around 10% of its maximum capacity so that there is enough headroom when it needs to inject more power.

Furthermore, there is ongoing research on “grid-forming” inverters. Grid-forming inverters can operate independently; they create and regulate grid voltage and frequency. Grid-forming inverters can also provide various ancillary services to the grid, such as inertia, system strength, voltage regulation, and frequency response. These services are essential for maintaining the power quality and security of the grid, especially when there is a high penetration of VRE.

The future?

Rather than making a precise prediction, I prefer to convey general trends of the power system based on what is already happening today.

On the generation side, there will be more inverter-based generators and less synchronous generators. Additionally, VREs are typically small and distributed. As more prosumers will be part of the electricity supply, power will increasingly flow bi-directionally (instead of just high to low voltage on consumer side).

A similar trend of “digitalization” is also happening on the demand side. More loads are based on power electronics (LEDs, VSD inverters, EVs, etc). Additionally, the load will not be passive but will become controllable, adjusting with the dynamics of the grid system (demand response). A new segment of load that can become a generator is emerging: electric vehicle charging and discharging to the grid (V2G).

Overall, the grid is evolving, and it is getting smarter.

Source: B. Kroposki et al., “Achieving a 100% Renewable Grid: Operating Electric Power Systems with Extremely High Levels of Variable Renewable Energy,” in IEEE Power and Energy Magazine, vol. 15, no. 2, pp. 61–73, March-April 2017, doi: 10.1109/MPE.2016.2637122.

These trends create new issues and demand for new solutions. It is not an insurmountable task that I believe our engineers can definitely solve. For readers who are electrical/power engineers, the well-known criteria of power system stability might need to be revisited to adapt to the evolving grid.

This article is not possible without the support and review from Gugun Bonar, an expert in Power System from PLN Indonesia who is currently completing his master degree in University of Queensland, Australia.