Most people still picture microgrids as simple systems of solar plus storage. And for good reason: This classic formula has provided the most basic and reliable areas of functionality - a grid-connected system that provides owners the benefits of alternate and renewable energy generation while allowing for frequency regulation and voltage support for the macrogrid. They’re also seen as a way to improve resiliency, with the ability to disconnect from the grid and run independently, in what we call “island mode,” to maintain local power when the grid goes down.
But the times are changing, and this is great news. Microgrids are quickly adapting to perform as complex, meshed networks that manage many forms of energy generation, storage and consumption. The existing conception of microgrids being just photovoltaics wired up to lithium-ion batteries isn’t an accurate picture of the complexity of the modern microgrid. Today, microgrid operators are plugging in traditional power generation systems, like solar, diesel generators and cogen facilities, as well as more modern technologies for generation and storage, including methane, tidal energy, wind power, flywheels, high-pressure steam, and even hydrogen production, to name a few exciting technologies on the forefront. They need to operate dozens of variables, and often change on a moment’s notice to anticipate varying demand needs.
Evolution in microgrids requires advanced energy control systems
This modern evolution of microgrids, both in variety of Distributed Energy Resources (DERs) and types of storage, is creating new challenges for control systems, in that they now require intelligence to grapple with more complexity than their original designs dictated. Control systems for the modern microgrid have to interface with many kinds of DERs and effectively command actions by the second - whether that means increasing generation, reducing consumption, load shedding, or shutting down non-critical loads.
Here’s the problem: Today, most DERs by default are programmed to be the microgrids’ only “master,” but systems with multiple DERs need to be synchronized so that all the inverters are operating in phase with each other. This means that an adequate control system needs to be designed to serve as the “central brain” for a microgrid and assign DERs to be either master or slave on demand. Effective control systems not only govern all the various parts of a microgrid, they are able to bring intelligence into the system - predicting and acting on energy needs at just the right moment.
Stone Edge Farm – A healthy microgrid in practice
One case study of this evolving role of the microgrid is found at Stone Edge Farm, where an advanced microgrid involving multiple DERs is being controlled by DC Systems software and operating on a working organic winery in Sonoma County, California. The farm poses sophisticated energy management challenges: the control system must react to changing load conditions, such as agriculture irrigation pumps, monitor for high consumption, predict loads, and turn systems on and off appropriately to reduce loads on demand while simultaneously ensuring available power for future use.
The microgrid manages advanced storage and generation systems - hydrogen, for example, which is both used to produce power via fuel cells, and to fuel hydrogen vehicles onsite. The system must coordinate the required daily load for hydrogen production, along with the charging of other energy storage devices to use in meeting the evening and nighttime loads. Without the microgrid control system, hydrogen control and production would be manual and run at fixed times, without the predictive insight of determining future load requirements. With an appropriate energy management system in place, this layer of intelligence offers ready resources to be deployed as needed.
A sign of what’s to come
These kinds of decisions are only a sample of what future microgrid control systems will have to manage. In their ultimate application, there will be myriad control scenarios, where systems must disconnect from the grid and transfer to internal power without any interruption, as well as control generation, storage and consumption, and deal with DERs being added and removed dynamically from the microgrid, such as the case of electric vehicles being plugged or unplugged from the microgrid. The microgrid control system becomes a complex and ever-changing system, constantly solving a complex equation on a case-by-case basis.
Many advanced systems today are unique, proof-of-concept deployments that are challenging enough in isolation, but will soon need to scale into the hundreds, and even thousands, interoperating together. How these systems then communicate with each other, and manage energy deployment as a collective, becomes a further level of complexity and opportunity for future microgrid control systems.
As we transition to this new future of energy management, microgrids will be able to serve increasing demands to provide energy securely, resiliently, on-demand, and in alignment with the best possible pricing, environmental impact, and unique on-site needs. We find this both a challenge and an inspiration, with great opportunity to develop the software that will serve as the underlying intelligence behind microgrid operations.
By creating a new platform for on-boarding new clean energy alternatives, and managing them with a level of predictive insight and intelligence that machine learning can bring, we look forward to a new era of energy management that is secure, robust, and shared -- one where we can provide a better framework for the grid, and bring sustainable, reliable, and safe power to communities all around the world.