NREL’s Energy Systems Integration Facility is Helping the Electric Power Sector Solve Big Problems | SEPA Skip to content

NREL’s Energy Systems Integration Facility is Helping the Electric Power Sector Solve Big Problems

The Energy Systems Integration Facility (ESIF) at the U.S. Department of Energy’s National Renewable Energy Laboratory (NREL) in Golden, Colorado was launched in 2013 to serve as a research and development facility for the electric power sector. The facility’s goal, then as now, is to help solve the most complex and critical problems of integrating distributed energy resources (DERs) onto the grid, by bringing together top researchers from NREL with industry experts looking for practical solutions to key problems in the field.

K Kaufmann, communications manager for the Smart Electric Power Alliance (SEPA), recently sat down with Matthew Futch, global business development leader for NREL, to learn more about ESIF. Futch is responsible for maintaining and expanding NREL’s industry partnerships for ESIF and the lab in general.

The following interview has been condensed and edited for readability and clarity.

K Kaufmann: When people come to the ESIF facility, what’s the one place you show them, where their eyes get wide and they say,  “Whoa, this is so cool”? Or does it depend on what they’re involved in?

Matt Futch: Yeah, it depends on what kind of company, utility or vendor we’re hosting. We shape the tour according to what their interests and needs are. But I’d say there’s one consistent thing that we show them — what we call the Insight Center.

It is very, very cool. You basically are looking into a three-dimensional space in which we project industry data in a three-dimensional way.

Kaufmann: Wow!

Futch: Let me back up to why we have a space like this. For some time, we’ve had some difficulty — in the distribution and transmission system — to be able to visualize what consequences to the grid would occur, from a power flow or from an economic standpoint, if you were to change things. What happens if we don’t put in a substation but put in distributed energy resources? What happens if you put close to 10,000 PV (photovoltaic) systems in a distribution system similar to what’s happening in Pacific Gas and Electric’s service territory? What does it actually, physically look like over time, going forward on the system?

You can do that through operational experience, or you can actually research these questions and begin to design systems with the benefit of secure energy data. You can look at it in a three-dimensional space so that you can do things like put a large amount of DERs in one theater and then move them to another part of the distribution system that you’ve modeled, and physically look at that in time and space and see what the capital-cost or the power-flow changes would be.

Most of the industry is still looking at one particular interconnection for one particular type of system and then having to design for that, one thing at a time. We’re trying to speed up, and lower the cost, of designing power systems in a high renewable penetration scenario using high-performance computing and the data that we get from the utilities and the vendors. We’re combining hardware and software to actually construct and design new power systems and look at it in a three-dimensional space.

Kaufmann: ESIF is a relatively new facility. It has actually only been around about—this is coming on its fourth year?

Futch: That’s right. It is virtually new. You walk around, and just like Bryan Hannegan, our Associate Laboratory Director for Energy Systems Integration, says, it’s got that new lab smell and feel to it.

The building is composed of a high-performance computer, which takes up almost an entire floor. There are multiple bays where you can plug and play large devices or small devices together in either a microgrid or some other type of configuration. And it also has, of course, the offices for all the research scientists who are doing all the cool research.

Kaufmann: Let’s walk through ESIF’s collaborative model.

Futch: We always encourage our partners or our potential partners who want to collaborate with us to actually visit the facility — sometimes for two or three days. Specifically, we’ll start with bringing them in and giving them a program that allows them to see and touch and feel all resources in the facility, and meet with our researchers.

We try to scope out some technical work that is based on the fundamental problem that they’re trying to solve. We are interested in things that are difficult and challenging rather than focused on incremental improvements. We’ll take them to all these different areas in the facility that allow them to see the type of work we’re doing. We’ll show them examples from other utilities and other vendors that we’re working with, and then we’ll have a long, detailed discussion on the technical or market problem that they’re trying to solve, and then we’ll scope out work with them based on that visit. That’s the first step.

The second step is to begin to break that scope of work into what I call three different potential buckets, and those buckets really go from the simplest, smallest, and easiest way to engage with us, all the way to really long-term, fundamental industry relationships that could span three to five years and could evolve into many individual projects.

Kaufmann: Can you give us more detail on those three buckets of work?

Futch: Bucket one is what we call a user agreement, where technology companies or utilities come in and use a particular piece of equipment that they may not have in their own facility or they may not have the funding for. They essentially rent out space or researchers’ time for a very limited amount of time to test this specific set of issues that they’re trying to resolve.

The second bucket is what I call a technical services agreement, which is simply a larger version of the previous example, except for more challenging problems. It involves more researchers and maybe some counterparts on the company side. The company is paying for the time and the research to get their problem solved. It usually lasts anywhere from six months to a year, and they’re just solving that one particular problem.

The third bucket is what I call long-term research relationships, which allow companies to work with us for a longer period of time to solve more than one problem.

For example, Google came to us with something called the “little box challenge,” which was aimed at getting the same level of inverter power — ancillary power, reactive power, power electronics — that allows you to do ancillary (grid-support) services. They wanted to get the same level of functionality that is in the current size of the standard inverter, which is a relatively big box near your house or near the feeder, and get that into something the size an Apple iPad. A very difficult problem.

We brought in a bunch of vendors in the inverter world, and we offered a reward to come up with a design for this, and maybe a partnership with Google or utilities that might be interested in this solution. We had them test out different electrical configurations to see if they could actually deliver the same level of functionality in something that was that smaller size. And one of them was able to do it.

Kaufmann: So what’s the next step?

Futch: The long-term is that Google as a corporation can now look at other things they might be interested in, such as efficiency in their data centers or some of the artificial intelligence algorithms that they use in their software. The arrangement we have with them allows us to work on multiple projects over a longer period of time.

Consul General of the Netherlands Gerbert Kunst tours NREL’s ESIF’s Insight Center. (Source: NREL)

Kaufmann: Where are you seeing significant advances in technology? What’s coming in the door?

Futch: Almost every utility that I’ve interacted with in the last three months, and I’d say a high proportion of the vendors I’ve spoken with, are talking around the edges of the same question.

If you use inverters to replace a lot of the existing spinning reserves that provide the necessary inertial power or reactive power to keep the system stable, can those inverters collectively provide the same level of inertial response as a large-baseload gas or coal power plant?

This issue is beginning to capture a lot of attention on the hard power system engineering side, and we are seeing advances. More and more inverter manufacturers are trying to push their power electronics to enable that type of inertial response.

Kaufmann: What are the areas we need to be looking at right now if we’re going to keep integrating DERs? What areas are really ripe for research, where you’re saying, “OK, companies, come on in. We think there are really good opportunities for partnerships here.”

Futch: There’s a big question from the industry right now about whether things like cloud computing can actually lower the cost and improve the speed at which interconnected devices at the edge of a distribution network can be made more reliable.

For example, both Pacific Gas and Electric (PG&E) and Tokyo Electric Power have deployed  — or will be deploying — very large numbers of smart meters. PG&E has something like eight or nine million. Tokyo Electric Power is going to have about 30 million deployed over the next two years. Those smart meters produce vast amounts of energy consumption data very rapidly, but that data is not really always being used in an operational way. It’s mostly being used to provide financial records or billing. And the promise of the industry for so long is that things like smart meters or smart inverters could throw off the right amount of data where you could begin to operate the power system in a way that’s very different from the way that it’s being operated now.

Instead of having a reactive system where you’re providing power a day ahead in the wholesale market or at the network operator’s utility, you could actually use that data to be very precise about when you need power, where you need the power, and whether you could produce a lot more renewable energy at the right time and place.

It’s going to be very hard to use that data just at the very edge of the system or with only one utility’s software. It really needs to be aggregated,  in a very power, robust and fast way to provide real, true business value. I think that’s an area that if you combine the high-performance computing that we have at ESIF with the actual power electronics where you can put pieces of things together, and you get a clearer look of where that data can provide real business value for utilities, their customers and the grid.

For more information about the Energy Systems Integration Facility, please visit the NREL website. ESIF also hosts industry days with tours of the facilities and demonstration projects. Representatives from ESIF will be onsite at the SEPA Utility Conference, April 24-26 in Tucson. Additionally, NREL researchers will be presenting at the conference about grid modernization and interconnection queues.