Part I: Opportunities at the intersection of energy storage, demand response and smart buildings

Written By: Stratton Report
October 4, 2016


An interview with Peter Kelly-Detwiler of NorthBridge Energy Partners

Recently, Advanced Microgrid Solutions announced that Macquarie Capital will provide $200 million in capital for the development and construction of a fleet of AMS’s energy storage projects, located at commercial, industrial and government host sites, creating what AMS terms “hybrid buildings.” Looking to understand the dynamics of this combination of real estate and energy storage, we spoke to consultant Peter Kelly-Detwiler of NorthBridge Energy Partners, who both advises AMS and is a recognized expert in demand response. Peter previously served as Senior Vice President of Energy Technology Services at Constellation NewEnergy,. where he established the Demand Response group and oversaw its growth to 1700 MW of dispatchable load.

Stratton Report: At the moment, how well does energy storage work with demand response?

Peter Kelly-Detwiler: There’s different types of demand response. If we are talking about frequency regulation or ancillary services, where what is needed is a response with a very short lead time, and where the power is only required for a short duration, that type of demand response works well with energy storage today. Where energy storage doesn’t work so well in terms of demand response is when capacity must be available to the system for up to six hours or longer.

SR: What do you see happening in the future?

PKD: Energy storage prices are expected to come down over the next five years by about 50 percent, and then you will have a different ballgame. Five years out we’re going to see a lot more opportunity for longer-term demand response being provided by batteries.

SR: What will drive these price reductions?

PKD: Partially it will be improvements in battery technology. Similarly to solar, as the central technology gets cheaper, the whole supply chain is forced to get more efficient, and prices get squeezed on all the system inputs. But you will also see significant reductions in cost unrelated to technological improvement, occurring simply as a result of manufacturing at greater scale. This is true of any manufacturing process. I believe this will be very visible in the large battery production volume that will be coming out of Asia. An additional source of lower costs will be the use of “pre-owned” EV batteries. Although no longer useable in electric cars, these batteries typically will have about 80 percent of their residual function still left and will still be suitable for grid-connected storage. Current estimates are that those batteries could come in at a third to a quarter of the price of new batteries. In fact, Underwriter’s Laboratory is now starting to work with some of the various vendors to create the standards around the use of “pre-owned” EV batteries. My analogy would be that for energy storage, new batteries will function as the name-brand drugs and used EV batteries will be the generics. They might not be quite as good but they’re certainly going to be a heck of a lot cheaper.

SR: Who will be able to take advantage of opportunities at the intersection of storage and DR?

PKD: I see existing demand response players as best-positioned to offer energy storage as a demand response application. The first requirement of making money out of demand response is the ability to participate in capacity markets. PJM and New England have “three-year-out” capacity auctions, and if you want to participate in demand response you have to take a position, a short position, and say that you are going to provide this much capacity three years from now and of course you bid in a price. Current demand response players are already active in such markets. If you didn’t participate in those auctions in past years, the only way to play in a formal DR market will be to find someone who has been participating so that you can market your energy storage capacity through them. Of course, there are other areas of the country where they don’t have formal demand response markets, for example in California. There, companies like STEM, Advanced Microgrid Solutions, Sharp, Green Charge Networks are already positioning themselves to monetize their demand response capabilities in other ways.

SR: What do you mean by “other ways” to monetize the demand response capabilities of energy storage?

PKD: I see monetizing approaches for demand response as likely to evolve in two different ways. One is the use of formal, organized DR markets like we see in PJM, in New England, in New York and probably in Texas as well. The other way is where utilities pay a certain amount for an asset that they can control when they want to control it. This is the idea of what is termed a “virtual power plant.” For example, ConEd Solutions is working with SunPower and SunVerge to create a virtual power plant where they are installing solar and storage capacity in the same system. The utility will then reaches out and activates that solar-charged energy storage capacity when they want to reduce load on the grid. An interesting sidelight to that arrangements is that there may be other times when the customer can use that energy storage capacity to create additional value, especially if the energy storage capacity is deliberately overbuilt relative to the what the utility needs.

SR: Are energy end-users aware of the potential for using energy storage to earn money via demand-reduction?

PKD: I recently spoke with representatives of NRG and Constellation, both of whom are installing energy storage on their customers’ sites and I asked them what are the barriers to growing this business? And they said one barrier is that their customers simply don’t know yet that they can do this. Of course, end-users would be more aware of potential revenue from their energy storage if regulators had made it clearer that they were aware that batteries can perform demand response functions and that they intend to compensate battery owners for that performance. in California both the regulators and the end-users are very aware of it. Places like Hawaii are becoming much more aware of it because they need it to integrate more solar into the system

SR: Your client, Advanced Microgrid Solutions, has coined a term, “hybrid buildings” to indicate real estate properties with installed energy storage capacities. How does that work?

PKD: AMS won a contract from a California utility, Southern California Edison. The utility paid for 50 MWX amount of the storage that’s being installed by AMS onsite in various commercial buildings, with the understanding that the utilities will be able to call on that stored energy at certain times. When the utility’s not using the battery, the contract allows AMS and the buildings to utilize it for other activities. One example is that the buildings can use the storage capacity to reduce their peak demand charges. They also get back-up power during outages. That kind of approach is increasingly common. The challenge for energy storage providers like AMS is to move beyond places like California where there is a strong regulatory and financial support mechanism, and into other markets like the northeast and perhaps Texas. That’s a question of essentially waiting for the cost to fall and for regional markets to get smarter about what they need.

SR: Are there opportunities at the intersection of energy storage and smart buildings?

PKD: Energy storage is a discrete system; it has its own software and it knows “what to do” to perform its function. The challenge of smart buildings is arranging communications and protocols for all the energy assets—lighting, HVAC, etc.—in the building so that they all can talk together and take direction as a team. And ideally you want to have that language be consistent across a whole fleet of buildings so that you’re not facing a Tower of Babel scenario. And that’s been a real challenge. Today there are only a limited number of smart buildings that interact with the grid because building energy systems are not like the Internet, we can’t all search the same body of data with the same query. The promise of smart buildings in terms of demand response is that it is very cheap to shed load by changing the thermometer set points and dimming the lights and that sort of things in an existing building, cheaper than by installing batteries. On the other hand, particularly as energy storage costs come down, it’ll be interesting to see whether people will go to the effort to deal with the challenge of smart buildings. They may just decide that because batteries are easier to control right out of the box, and no behavioral changes are required, they’ll just skip the smart building and go with some energy storage.

SR: Are these same problems as difficult to solve in new buildings?

PKD: No, it’s much easier with a new building to control its function in a unified energy management system. If you put in a new LED lighting system you can put the whole mesh network system in all at once. You can think it through beforehand, and if you want, you can make it market-aware right from the get go. So it’s much easier to think about new construction. Unfortunately new construction adds less than one percent to the total building fleet each year so if we really want to make changes in society’s energy use patterns we really have to focus on the retrofit issue