Energy Literacy: Let's talk about inertia

I don’t remember the first time I heard an electric power industry executive talk about inertia, but I know I was immediately confused.

I mean, inertia was about things being stuck, not moving, right? But this person was saying it had something to do with turbines and spinning reserves – whatever they  were – which were somehow essential for keeping the lights on.

“Inertia” is the kind of industry jargon that traditionally has allowed executives and other experts to keep their meanings and motivations purposefully obscure. More to the point, between the current uproar over demand growth from data centers and President Trump’s drive for U.S. energy dominance, it has become increasingly politicized. 

In the interests of full disclosure, I will readily admit I spent several years working around the word, nodding along whenever it was discussed. 

Based on what I heard, I had the impression that inertia was some kind of backup power that could come online very quickly and possibly run for extended periods of time — hence the insistence on those spinning reserves. Coal- and natural gas-fired power plants could do it; solar and wind couldn’t.

If you had too much solar and wind on the grid, you wouldn’t have enough inertia in the event of a sudden loss of power — an argument almost immediately but erroneously raised after the April 28 power outage in Spain and Portugal.

U.S. grid operators, along with various policymakers and regulators, are also invoking the need for inertia as part of their rationale for fast-tracking approvals of new natural gas and nuclear plants, again, to ensure reliability while meeting growing power demand from data centers.

Renewables are being demonized as “unreliable,” when they could be a major part of the solution if those industry leaders would look at grid reliability in new, dynamic ways.

All of which is why I decided to write this month’s Energy Literacy column on inertia. I was determined to tackle the jargon head on, to find out exactly what inertia is, what it isn’t and how much we actually need.

Inertia 101

To begin, yes, according to Merriam-Webster, inertia does mean an “indisposition to motion, exertion or change,” which is how most people understand it. But it’s got this other, more complicated definition based on Sir Isaac Newton’s (1643-1727) first law of motion: bodies in motion tend to stay in motion; bodies at rest tend to stay at rest.

A 2020 paper from the National Renewable Energy Laboratory — “Inertia and the Power Grid: A Guide Without the Spin” — provides a pretty straightforward explanation of how such Newtonian physics apply to the grid. 

“Inertia in power systems refers to the [kinetic] energy stored in large rotating generators and some industrial motors, which gives them the tendency to remain rotating,” the report says. “This stored energy can be particularly valuable when a large power plant fails, as it can temporarily make up for the power lost from the failed generator. This temporary response—which is typically available for a few seconds—allows the mechanical systems that control most power plants time to detect and respond to the failure.”

Going into more detail, the report describes how “inertia is derived from hundreds or thousands of generators that are synchronized, meaning they are all rotating in lock step at the same frequency. … All the individual generators that are online and spinning can work together to contribute to grid inertia.” 

Now, the one wonky bit here is frequency, which is a measure of how fast electrons move, or oscillate, on the grid per second, measured in something called Hertz (Hz). To maintain grid reliability, frequency has to stay within a very narrow range around 60 Hz. For most U.S. grid operators, a drop to 59.5 Hz can trigger further outages. (If you really want to go into the weeds, a more technically detailed explanation of frequency can be found here.)

The chart below, from the NREL report, shows how traditional inertia works to maintain frequency and reliability. When a power plant goes offline (the contingency), kinetic energy –- inertia –- from other power plants is converted into electricity and used to slow down the loss of frequency on the grid, all within two seconds. Additional power can then be brought online (primary frequency response, or PFR), along with other support services, to move the frequency back up toward 60 Hz and prevent any further outages — the under frequency load shedding (UFLS) on the chart .

The topline take-away here is that while inertia is one of a range of tools that can be used to maintain grid reliability, it is not the only one and, as the grid evolves, may not be the best or most effective. 

Further, the need for inertia occurs in super-short time frames, and the amount needed can vary. For example, the report notes that having more renewables on the grid may reduce the amount of inertia that is available, but could also reduce the amount needed. 

Wind, solar and storage may not have turbines that spin, the report says, but they “can respond much faster than conventional resources, reducing the amount of inertia actually needed … This represents a paradigm shift in how we think about providing frequency response.” 

In fact, the NREL report says, the grid’s “reliance on inertia to date results largely from the legacy use of synchronous generators” – those fossil fuel-fired power plants with turbines spinning in lock step. 

Doing inertia differently

So, how might we think differently about inertia, and how much we actually need, within the current landscape of increasing demand growth and increasing renewables on the grid? 

The NREL report points first to the Electric Reliability Council of Texas — the state’s grid operator, commonly referred to as ERCOT — as ahead of other U.S. grid operators on stretching the connection between inertia and reliability.

Texas has a relatively small grid, which means it has less inertia to respond to a plant going offline to begin with, the report says. It also has increasing amounts of wind and solar online. One of the main things ERCOT has done is to lower the level at which a drop in frequency triggers other outages, providing more time and flexibility to restore balance on the system.

Another strategy is demand management, which, the report says, “consists of equipping certain large industrial loads [plants] with sensors that measure frequency and are programmed to disconnect automatically when the frequency drops below a certain level.”

The disconnection lasts for about half a second. Taking part in the demand management program is voluntary, and the companies involved are paid for participating. 

A more recent report from Duke University grabbed headlines in the industry media with its proposal to extend ERCOT’s use of demand management to data centers themselves. 

Data centers — or other large industrial plants — could “temporarily reduce their electricity consumption from the grid during periods of system stress by using on-site generators, shifting workload to other facilities, or reducing operations,” the Duke report says.

Reducing this demand by as little as .25% could allow grid operators to accommodate 76 GW of new demand on their systems. A 1% reduction could make room for 126 GW of new demand, which some studies have estimated is about the same amount of new power the U.S. could need for data centers by 2030.

In other words, we wouldn’t have to build new natural gas plants if we could enlist hyperscalers like Google, Amazon and Microsoft to help us use the grid we have more flexibly.

Texas is, once again, leading the pack. Republican Gov. Greg Abbott on June 20 signed a new law that will require data centers or other large industrial plants that use a lot of electricity to curtail their demand – that is, briefly reduce their power consumption – if the grid is under stress. The law also calls for new voluntary demand management programs that will allow utilities to ask these large megawatt-guzzlers to reduce their demand — in some cases, by firing up their onsite backup power — to maintain grid reliability.

Outdated views, slow-moving institutions

Now, we get to the really fun stuff, using renewable energy – solar, wind and storage – to keep grid frequencies at reliable levels, traditional inertia not required. 

As previously noted, renewables may not spin, but they have inverters — the devices that convert the direct current (DC) they produce to the alternating current (AC) used on the grid — which can respond to a drop in frequency much faster than turbine-dependent inertia. 

The result has been dubbed “fast frequency response,” and federal regulations approved in 2018 require that all new generation connecting to the grid — with or without spin — must be able respond to a drop in frequency. According to the NREL report, both Quebec and ERCOT have required wind turbines to have “frequency-responsive reserves” for more than a decade. 

“After the time required to sense frequency and initiate a response, wind can increase output by as much as 25% per second, while [solar] can increase output over its full range [that is, 100%] in less than a second,” the report says. 

Small, zero-inertia microgrids combining renewables and storage “have been in operation for decades, which demonstrates that inertia is not needed to operate an AC power system.”

So, if solar, wind and storage can respond to potential energy emergencies faster and better, why do we still have all this brouhaha about inertia?

The Trumpian drive to fossil fuels is at least part of the motivation behind current efforts to convince the public that inertia is hardwired to reliability and that how we produce our electricity has no connection to climate change.

But at a more basic level, many grid operators and regulators simply don’t trust renewables. They raise questions like, what will happen if there’s a drop in frequency on the grid, and your energy storage is out of power? 

Stoking fears about potential blackouts caused by worst-case, but unlikely scenarios is their stock in trade, and it remains very effective. Or, as a recent report from the nonprofit think tank Energy Innovation Policy and Technology put it: “Outdated views on grid reliability are colliding with slow-moving institutions.”

In 2020, the NREL report acknowledged that more research on the impact of renewables on the grid and their ability to provide fast frequency response would be needed. Federal regulators say standards for frequency response from inverter-based renewables must be set, which could take years.

The Energy Innovation report argues for a more holistic approach to grid reliability, one in which inertia plays a very small and potentially unnecessary role. With demand and renewables both increasing, we must “[move] beyond a reliability construct that depends primarily on baseload power to one that depends on a broad portfolio of resources and load flexibility,” the report says.

 Right now, power plants that have spinning turbines – coal, natural gas, hydropower and nuclear – produce about 85% of the country’s electric power. The new power waiting to get online – an estimated 2,000 GW – is overwhelmingly solar, wind and storage. 

Don’t believe the hype. We’ve got plenty of inertia to tide us over till we don’t need it any more and plenty of ways to get there.