Microgrids, DERs, and Data Centers: What’s The Deal?

Data center leaders are constantly on the search for cost and energy savings. One solution garnering increasing curiosity is the idea of microgrids, the ability to temporarily operate independently from a primary electric source. If facilities like hospitals and military bases use them, can they work for data centers?

The discussion centers around distributed energy resources (DERs), which can come in many forms: battery energy storage systems, conventional generators, or renewable energy devices like solar panels or wind turbines. Emerging trends in data center sustainability are putting extra emphasis on exploring DERs, as they can be the key to decarbonization. Optimizing with DERs, though, can be complicated, as load profiles present a challenge for data center operators looking to take advantage of the technology.

Rehlko’s Benjamin Rapp recently sat down to discuss what sets what sets data centers apart from most other DER use cases, and how backup generators can play an important part in a data center’s DER plan:

For people who don't know, at the simplest level, for a beginner, how would you describe DERs?

Benjamin Rapp (Rehlko): I think of distributed energy resources as small, modular energy generation and storage technologies. You might think of these as batteries, solar plants, wind farms perhaps, and even backup generators. When you look at all those options, there are a couple of different types, which you can think of as dispatchable and non-dispatchable. You can't really control when the sun is shining or the wind is blowing, so those are your non-dispatchable loads. Dispatchable resources are those that you can choose to turn on and off as you'd like. Diesel generators are, in fact, a dispatchable load, as well as battery storage and fuel cells.

You've got a background in engineering. From that perspective, what are some of the essential issues around DERs that people in the mission critical sector are talking about right now?

Rapp: Resiliency is top of mind, especially with storms and natural disasters that can occur and shut down the grid. Facilities, especially data centers with critical applications, are extremely focused on making sure they have a reliable connection to electricity. When you think about distributed energy resources, what emerges as top of mind is having various different pieces of distributed energy resources — those different types I mentioned — all together in one regional location, also known as a microgrid.

Essentially, that microgrid and these assets can function connected to the grid, but in the event of an outage, they can disconnect and then operate your facility as if you didn't have a connection to the grid in the first place, maintaining all the same performance you'd expect. When it comes to data centers, a key focus is really understanding what this could look like for your facility, understanding your load profile, getting high-quality data to understand where your energy demands are, and then based on that determining if there's a possibility to optimize with distributed energy resources.

What are some specific use cases for data center DERs right now?

Rapp: For data centers, reliability has always existed as a priority and will continue to exist. But now there's big talk about decarbonization. Microgrids can be that avenue. It's not only looking at emissions from grid-connected power, but if we have assets on-site like a backup diesel generator, is there a way to decarbonize? Rehlko has its own solutions, which we'll probably talk about in a bit, but there have also been some use cases in the industry where data centers have implemented distributed energy resources such as battery systems. I think Google is testing that in Europe, for example, as well as some renewable natural gas-type turbine systems implemented by Microsoft, I believe in Texas. There are some use cases where data centers are typically exploring these, but probably on a smaller scale than some other industries. A lot of that comes down to the balance between dispatchable and non-dispatchable assets and understanding how that fits together with a very consistent load profile. Data centers don't see a lot of modulation in their load profile, so it's really a matter of understanding what kind of distributed energy resources are best used and how those can optimize our energy requirements.

Rehlko, with the backup generator, has been one of the original players in the DER game. How do you see the company continuing this trend and expanding in it?

Rapp: I touched on decarbonization, and it's been a focus of our operation plan for our products: how do we better optimize our products themselves to then have our customers be optimized in their operations? This question is really advancing our engine technology, advancing our efficiency within the system, and we've adopted technologies to allow for the use of hydrotreated vegetable oil. Altogether, it comes down to helping our customers be more sustainable in their operations, but still maintaining that same level of on-demand resiliency you'd expect from a backup generator.

When we think of generators in the distributed energy space, they typically have just been assets that sit idle. They exercise monthly and then start up in the event of an issue. One of our current goals is about transitioning that from being an insurance policy to being more actively used. That's where you start having a conversation around Tier 4 generators, so they can run more often, maybe on a demand response program or shaving peaks on your load profile. Those can be used because of the aftertreatment system, and Rehlko has one of the simplest aftertreatment systems for our Tier 4 generators in those applications.

Taking it a step further, we've acquired a company called Helia Technologies. This brings in a controls aspect. When we think about distributed energy resources and this microgrid concept, this Helia system is able to apply controls at each asset. Typically and historically, we've heard of centralized controllers controlling all these assets, but Rehlko's system is now able to control each asset, and they're able to bid against each other, effectively removing that single point of failure that a centralized controller could provide, adding to that sense of resiliency for the total system. What's nice about that is, I talked about understanding data. Often, when microgrid developers are modeling these systems for a data center customer, there's a difference between what actually happens in the field and what was modeled. With Rehlko's system and those controls with Helia Technologies, they're able to use the same algorithms that are in those controls for the modeling. So what you see in the model is what you're going to get at the end. I think that's a huge point when it comes to payback for a data center customer.

Is there anything else that you think is really important for somebody just getting into this and learning about DERs to know and understand?

Rapp: I think, as I've mentioned, it's really about understanding what your goals are as a customer and finding the right partner to meet those goals. Throughout the portfolio of businesses within Rehlko, we have a place to play in that conversation, and we're happy to help with any data center needs in this space.

To hear more from Rapp, check out his interview at Yotta here. Or, to continue learning about how Rehlko offers a cleaner power advantage for data centers, read more here.