Q&A with Annabelle Pratt: Managing Energy at All Scales
Annabelle Pratt is a principal engineer for the Power Systems Engineering Center at NREL. Pratt specializes in microgrid and home energy management systems (HEMS), and her work has included the evaluation of microgrid controllers and advanced distribution management systems (ADMS). We sat down with Pratt to learn more about her work and the newest research in these areas. The following has been edited for length.
Tell me more about your background.
I specialized in power electronics in graduate school, and my first position after graduation was at a power supply company called Advanced Energy in Fort Collins, Colorado. While I was there, I participated in developing a prototype for a PV inverter. Then I went to Intel Corporation for about nine years, and I worked on power supplies for microprocessors, servers, and data center power delivery. My last four-and-a-half years there, I joined the Energy Systems Research Group within Intel Labs, which gave me an opportunity to get back to grid applications. I ended up working on microgrids and HEMS. The work I did there provided a bridge for me to bring that experience to NREL when I joined in 2014.
What has been your involvement with microgrid research?
I’ve worked on several microgrid projects, mainly focused on the role of microgrid controllers and the evaluation of microgrid controls in a laboratory environment. For one project, we partnered with the Electric Power Research Institute (EPRI) and brought a commercial microgrid controller with advanced capabilities into the Energy System Integration Laboratory. Through simulation of a medical campus and adjacent residential community, we evaluated a community EPRI was looking at modifying to run as a microgrid. We modeled that system using our real-time simulator capability, and then we interfaced that simulation with the actual controller hardware from the vendor. We also interfaced the simulation with a 540-kW battery inverter, with controller- and power-hardware-in-the-loop.
I was also involved with an exciting project around San Diego Gas & Electric Company’s (SDG&E) Borrego Springs community microgrid located in a valley just east of San Diego. We simulated the community using real-time simulators, and we tied it to a vendor microgrid controller along with two large inverters: a 500-kW PV inverter and a 540-kW battery inverter. The intent was to evaluate the performance of the controller under different conditions. The benefit of running the simulation at the lab is being able to run abnormal conditions, so you can simulate extreme events that you wouldn’t want to replicate in the field, and then evaluate the behavior of the controller.
I’m involved in a DOE-funded project that looks at networked microgrids. This project focuses on how interconnecting microgrids in a distribution system can improve resilience. The thought is instead of having only two or three islanded areas on a distribution system that can maintain power, why not have them share their resources to increase their resilience? For example, depending on how much generation capacity they have, they might be able to pick up other loads within the distribution system. This is really the new area in microgrid research—to look not only at single microgrids but also at how they can work together.
Evaluating and optimizing the value of advanced distribution management systems for utilities has played a key part in your work at NREL. What are you currently working on in this area?
When I introduce myself, I often say I work on energy management systems at different scales. The microgrid controllers are one scale, and the work I’ve done on HEMS is another. But over all of that is the work I’m currently involved in and really enjoying: the ADMS. These are software platforms that control the operation of an entire distribution system. ADMS has a lot of potential in terms of being able to manage a system with higher penetrations of renewables, increase resilience, interoperate with microgrids, improve operational flexibility, and bring more data security. Deploying an ADMS is a significant undertaking financially and in terms of time, and it really changes the way utilities operate their system. Currently, the ADMS adoption is low. The DOE Office of Electricity’s Advanced Grid Research program is hoping to address this problem and has worked with NREL to set up an ADMS test bed. That’s the project I’m leading at NREL. If there’s a utility that is considering deploying an ADMS and they’ve selected their ADMS vendor, NREL can then simulate that specific utility system, subject it to a wide range of load conditions and disruptive events, and determine the optimal ADMS settings for that utility’s specific needs and priorities. This lab testing allows the utility to understand how to best deploy their ADMS and exactly what the benefits are of that investment.
How has Xcel Energy been using the ADMS test bed in partnership with NREL?
Xcel Energy is interested in ADMS performance. Specifically, answering the question “how good does the model of the system need to be within the ADMS to get good performance?” To provide some context, the ADMS is a model-based platform that uses a representation of the distribution network and real-time sensor data to make control decisions. If the ADMS model doesn’t match the system that’s in the field, then you can expect some deterioration in performance. But the model doesn’t need to be perfect to achieve significant performance and financial benefits. The goal is to find the sweet spot and optimize the return on investment. Xcel Energy wants to use the output of the study we are doing to determine how much effort they should expend in updating their models so they can achieve a good performance. They’ve been an excellent partner, and this is a question that has real financial implications for them. This question has generated a lot of interest among our other utility and vendor partners as well. It’s wonderful to work on something that you feel is really impacting their operations and ultimately our power bills.
What’s another area you are working in that you are excited about?
Another set of projects I’ve done over the years here at NREL includes my work on HEMS. This work is really around the simulation of distribution systems to evaluate the impact of advanced controls within homes. We think about what happens when you have a high penetration of grid edge devices such as PV inverters, battery systems, smart thermostats, and smart water heaters, and how people manage those devices in response to a certain rate structure such as a time-of-use rates. It all comes back to the impact of technologies and different rate and tariff structures on the operation of the distribution system. We want to be able to continue to operate a reliable and cost-effective distribution system; it’s our lifeblood to have affordable and dependable electricity available to us. Many of my interests are around these new types of controls, technologies, and their impacts. With the simulation platform we’ve developed, we can demonstrate the effectiveness of particular controls, and after running an analysis, we can observe possible unintended consequences that will help refine those controls.
To learn more about NREL’s ADMS test bed, watch this 2-minute video or read our factsheet.
To learn more about NREL’s home energy management system testing capabilities, visit our webpage on NREL.gov.