Inverters on the Grid

Inverters catch up with me

The inverters catch up with me.

A few years ago, when I was writing Shorting the Grid, I realized that it would be a major task to understand how inverters affected the grid. The task was so big that I skipped it. I figured I could either write more about Inverter Based Resources or I could finish Shorting the Grid. I chose to finish the book.

Now the inverter issue has caught up with me. NERC (North American Electric Reliability Corporation) has issued a Level 2 Inverter Based Performance Issues Alert. Owners of grid-connected large solar arrays must respond to NERC by the end of this month.

As NERC wrote in their justification for the alert: These resources have exhibited systemic performance issues that could lead to potential widespread outages if they persist.

I guess I need to write about inverters.

What is an inverter?

The grid accepts AC (alternating current power) but many renewables make DC (direct current power). Inverters change DC to AC so that the grid can accept the power. Solar, wind, and batteries make DC power, and they all use inverters to put power on the grid. Solar, wind, and batteries are all Inverter Based Resources (IBRs).

(At the end of this blog post, I have a postscript on inverters.)

Task forces, orders, and standards, oh my!

Years after finishing the book, I have to return to discussing inverters.

But to some extent, I have the same problem as I had in the past. Anything about IBRs turns out to be a major rabbit hole. Luckily, it didn’t take a lot of digging to come up with FERC Docket No. RM22-12-000; Order No.901

This FERC order, issued on October 19, 2023, is titled, “Reliability Standards to Address Inverter-Based Resources.”

The FERC order is about 240 pages long. (The pagination is a bit wacky…I am using the page numbers in the table of contents.) Since the Inflation Reduction Act is approximately 1000 pages long, I guess 240 pages is reasonably short. If you begin reading government documents, your standards of brevity must change.

I looked at the FERC article and thought: “Okaayy….let’s dig in!” Doomberg often starts the main part of his articles with “Let’s dig in.” He writes great articles. But I could not dig in.

Not A Set of Standards

I finally realized. The FERC order is NOT a set of reliability standards. This order simply requires NERC to BUILD a set of reliability standards. NERC has until November 6, 2026, to do so, and to fill in any gaps in their rules.

The crucial part of the FERC order is the Performance requirements, starting on page 178. These consist of

1. Registered IBR Frequency and Voltage Ride Through Requirements ............. 178.

2. Bulk-Power System Planners and Operators Voltage Ride Through Mitigation Activities .......................................................................................................... 196.

1. Post-disturbance IBR Ramp Rate Interactions and Phase Lock Loop Synchronization ..................................................................................................... 200.

Let’s dig in, as deeply as we can. Which is not very deep. We need to remember that the whole order could be considered merely a “suggestion for further work.”

I will only address the first item in this blog post, because I hate when my blog posts turn into lengthy books.

Registered

1. Registered IBR Frequency and Voltage Ride Through Requirements .

The first crucial word here is “Registered.”

ISO-NE wrote a seven-page comment on the FERC order. This is what ISO-NE wrote about that first word, registered: (emphasis added)

“The NOPR (Notice Of Proposed Rulemaking) states that the new or modified Reliability Standards should ensure that registered IBRs provide frequency and voltage support during frequency and voltage excursions in a manner necessary to contribute toward the overall system needs for essential reliability services…..ISO-NE agrees that the new or modified Reliability Standards should require registered IBRs to provide frequency and voltage support. However, ISO-NE notes that placing these performance requirements only on registered IBRs means that, in New England, the requirements would not apply to the vast majority of IBR capacity. “

In other words, most of the inverter-based resources in New England are “behind the meter” solar. Rooftop solar and so forth. “Behind the meter” means (among other things) that these entities are not registered with the bulk power system. This means that these home systems do not have to contribute to overall system needs.

In fairness, FERC has asked NERC to submit a work plan on how it will identify IBRs that are connected to the bulk power system but don’t currently have to register. If they register, they will have to meet Reliability Standards performance requirements, which will be defined by NERC, supposedly soon.

But I wonder how many homeowners will be eager to register their solar arrays so that they can meet the performance requirements? How many politicians will want to require their constituents to meet those requirements?

Which brings up another problem. To meet performance requirements, you need to collect data. Will homeowner data be self-consistent and correct?

All right. Enough about “registered.”

Ride Through

Let’s move on to the rest of the short statement. “IBR Frequency and Voltage Ride Through Requirements”

First, we must discuss “Ride Through.” In this order (page 131) FERC defines the goal as follows:

“Registered IBR Frequency and Voltage Ride Through Requirements: In the NOPR, the Commission preliminarily found that the Reliability Standards should require registered IBRs to ride through system disturbances to support essential reliability services. Without the availability of essential reliability services, the Commission explained that the system would experience instability, voltage collapse, or uncontrolled separation.”

The next pages are notes from various grid groups (NERC, EPRI, others) about what ride-through means. They don’t all agree. In my understanding, the simplest version of what “ride-through” means is that if there is a fault on the system, the IBRs should not amplify the fault. The IBR generators and inverters should stay on through the fault. When they disconnect, it amplifies the fault.

Unfortunately, we have several examples of what “amplifying the fault would look like.” There are already examples of IBRs behaving badly. In the section “Commission Determination” beginning on page 138, FERC notes that

“This directive (for new or modified reliability standards) is supported by the comments, as well as the recommendations from multiple event reports, including the Blue Cut Fire Event Report, the Odessa 2021 Disturbance Report, and the 2021 Solar PV Disturbances Report.”

These are all incidents in which IBR systems on the grid helped a fault to spread, rather than damping it out.

Odessa, Texas

Let’s look at the “Odessa disturbance” because that is the one that many people have heard about. The report on this disturbance is also the source of the graphic at the head of this post.

The incident started when a surge arrester failed at a combined cycle power plant. This caused a circuit breaker to trip, and the 192 MW combined cycle power plant went offline. The disturbance this caused on the grid (mainly a frequency drop) cleared within a couple of electrical cycles. However, many generators that used IBRs tripped due to the disturbance. Over 1100 MW of solar and wind (mostly solar) went offline. As the NERC report said (page v):

“None of the affected inverter-based resources were tripped consequentially by the fault itself. Rather, all reductions (in output) were due to inverter-level or feeder-level tripping or control system behavior within the resources.”

In other words, the inverters could not ride through a fault. Instead, they tripped and amplified it.

Figure 1.4 from the NERC report (graphic above) shows the original fault (red) and the affected solar installations in blue. Affected wind installations are in orange.

Nothing is perfect, but you gotta try

When I was speaking in favor of nuclear energy, I used to be annoyed when an anti-nuclear person would come up to inform me that nuclear energy has corrosion issues. Yeah, baby, what problems do you think the members of NACE International are always trying to solve, for all sorts of industries? I was a member at the time. NACE is all about corrosion issues!

(NACE has mostly changed its name to the Association for Materials Protection and Performance. At the time it was the National Association of Corrosion Engineers. Though some of us were chemists, not engineers.)

Engineering solves problems. Not every problem is a showstopper. Many problems can be solved.

Showstoppers

The question for me is whether the renewable industry is trying to prevent problems. If we are going to have renewables on the grid, in high percentages, then these quotes from the NERC Odessa report are scarier (to me) than the map above. (page 9 of the report)

“In many cases, industry is not proactively identifying abnormal performance of inverter-based resources…the recommendations outlined in NERC reliability guidelines are not being adequately adopted….Plants stated that no mitigating actions are being done (or planned) to improve the performance of the resources involved in the event.”

Without action, problems can become showstoppers. Generation owners need to take care. They can’t neglect problems.

Our grid depends on it.

--------

P.S. for extra credit

What is an inverter?

Most traditional power plants have a spinning generator at their core. As the generator spins in a magnetic field, it produces alternating current (AC) power in a smooth sine wave. Most renewable plants (solar PV, wind) produce direct current (DC) power which must be changed to AC power to be put on the grid. Inverters change the DC power to AC power, and so most renewable power plants are called Inverter Based Resources. (IBR for short.)

Wind turbines make AC power, but they make it at a frequency that depends on the wind speed. However, the AC power that goes on the grid must be carefully matched to the grid frequency. The AC power (at whatever frequency) from the wind turbines is all changed to DC power (using a rectifier.) Then the DC power goes to an inverter to make AC power that matches the frequency on the grid. Wind turbines are an inverter-based resource.

Rotating machinery makes sine waves automatically, but inverters usually approximate sine waves through on-off switches. The normal output of a simple inverter is a square wave, but more sophisticated inverters approximate a sine wave more closely. Most inverters “follow” the existing sine wave on the grid to set their frequency.

The Wikipedia article on inverters can be a good start to more information.

The illustration below is from the Wikipedia article.

Acknowledgement:

I am grateful for the advice of Russell Schussler for his guidance during early stages of writing this post. You can often find Russ as “Planning Engineer” on various social media platforms.