Although new microgrid configurations, technologies, and business models are still evolving in the U.S., some lessons have been learned in the past few years. Aside from the fact that financing nontraditional/non-campus microgrids is hard, if there’s one overarching lesson, it’s that a microgrid designed to provide only one benefit or rely on only one generation source is unlikely to succeed.
The following lessons have been distilled from comments made at the March 2016 Infocast Microgrid Markets Summit East as well as other POWER research.
In contrast to pilot projects or “science projects” to test the operation of a particular component, microgrids work best when all components, benefits, and stakeholders are considered from the start. Just wanting to add renewable generation to a building project, for example, isn’t sufficient.
A related lesson is to wring as much value from a project as possible from the start—even though some future benefits may accrue that cannot be anticipated (see “Stacking Microgrid Benefits” below). For example, use economies of scale and diversity, as far as possible, to aggregate multiple loads with complementary demand profiles.
Model Before You Build
For years, microgrid developers have recognized that, although the parts of a microgrid are all familiar, designing an effective installation poses new sorts of system integration challenges. Especially in an arena where you have new players and familiar players in new roles, there are going to be unknown interconnection behaviors that can potentially damage expensive equipment. The more you model, the less you’ll have to figure out while operating a microgrid.
One new tool to assist with this task is the Microgrid Controller Hardware-in-the-Loop Demonstration Platform (see the photo below) developed at the Massachusetts Institute of Technology’s (MIT’s) Lincoln Laboratory (Lincoln Lab).
Several researchers involved in the Lincoln Lab project were on hand at the Microgrid Summit to display and discuss their “physical model” of a microgrid—essentially a test integration model. Researchers hope it will be used as a development, deployment, and standards-testing platform to help accelerate microgrid deployments, MIT’s Kendall Nowocin explained. The platform uses the configuration of an anonymized real-world power system (code name “The Sheriff”) operating as a microgrid.
There are three levels of control on a grid of any size, Nowocin told POWER: primary—for performing equipment protection at microsecond to millisecond speeds; secondary—for controlling forcing functions (frequency, voltage, power, power factor) at millisecond to second speeds; and tertiary—the supervisory level that doesn’t directly control equipment and operates at seconds to minutes speeds. The Lincoln Lab model can import a one-line diagram (schematic) of any existing or planned microgrid in the 1-MW to 50-MW range to test how microgrid controllers (the tertiary level) from various vendors would behave—thereby eliminating many of the current unknowns that can spell trouble for one-off projects.
Anyone, including vendors, can use the “Sheriff” model (contact G73-HIL-PLATFORM@ll.mit.edu), and the lab plans to release the model’s source code publicly in April. Phillip Barton, director of Schneider Electric’s North American Microgrid Competency Center, said he recommends working with Lincoln Lab and that, as his company moved into hardware-in-the-loop testing, they learned it is less risky than onsite testing.
Consider Different Configurations for Different Purposes
Although there are benefits to standardization and replicability, some circumstances may call for atypical solutions. Direct-current (DC) microgrids are one example.
Andrew Yip, a business development manager with Robert Bosch (Bosch), told POWER his company is developing high-voltage building-level DC microgrids. It has two systems in place in North Carolina (one at Fort Bragg) and two more going in this year, one at a Bosch facility in Michigan and another supported by a California Energy Commission grant for a project in Los Angeles.
By coupling solar and storage with AC loads that are converted to DC loads—especially LED lighting, industrial fans, and forklift chargers—Bosch says it can save 3% to 4% on conversion losses. That can add up to substantial energy savings for businesses.
Reliability Is Always Important
Whether you are designing a microgrid primarily for backup power or to participate in energy markets to reduce power purchase costs, reliability as a means of securing resiliency for a building, process, or community should always be considered—even if assigning this benefit a dollar value is difficult. If you are not providing some level of resiliency, you’re leaving value on the table.
Third-Party Financing Remains Important
As in the early days of solar power, external financing sources can help get a microgrid project off the ground. California Public Utilities Commission advisor Jamie Ormond noted that different funding options can be available depending on the microgrid’s goal. In her state, for example, there are options for projects related to energy and water management.
In Search of the Best Microgrid Pancake Recipe
Microgrid developers and financers often talk about the importance of “pancaking” value streams to make a project financially viable. Just providing backup power, for example, is unlikely to make a project pencil out. More recently, “resiliency” for a customer or community—supported by the electricity reliability enabled by a microgrid—has become a primary motive for considering microgrid development. The problem is that, while few would argue against resiliency, there’s currently no way to make it a bankable benefit for financing purposes.
That’s why everyone with any experience in this energy supply area agrees on the importance of finding multiple value streams. For example, looking at the graphic below, that might mean bundling rooftop and carport photovoltaic generation with battery storage, electric vehicle (EV) charging/energy storage, and a small fossil-fueled genset—all of which might enable (where allowed) the microgrid to earn revenue from selling ancillary services to the grid, lowering overall electricity costs, providing tenant/customer EV convenience as well as power supply security.
Note that each wholesale energy market and state has different rules for what sorts of generation can be owned by customers behind the meter, by what sorts of entities, and how power can be moved over what geographic lines. States also vary in interconnection requirements, net metering rules, and what distributed energy resources a utility may own behind the meter. All those factors combine to create an enormous amount of variability in terms of determining what’s legally—let alone financially—viable for a new microgrid.
As a result, especially when you look beyond building-level microgrids, project design is likely to require customization for the foreseeable future.
This article is a preview of a longer article on microgrids that will appear in the May issue of POWER, which will also be available at powermag.com.
—Gail Reitenbach, PhD, editor (@GailReit, @POWERmagazine)