If you're like me, you look at statistics about the growth of generation resources like solar installations and microgrids and can't help but think about the need to push intelligence farther out on the grid.

This past April 22, Earth Day, folks at the Solar Energy Industry Association (SEIA) announced that they'd be celebrating every two and half minutes. That's how often a new solar installation is completed in America. Today, the U.S. has some 20 gigawatts of installed solar capacity, enough to power about 4 million homes. The solar industry now employs 175,000 workers, more than Google, Apple, Facebook and Twitter combined, an SEIA news release noted.

Meanwhile, Navigant Research says that microgrids have moved beyond research and development and the technology is moving rapidly toward full-scale commercialization. That industry is forecasted by Navigant to grow from $4.3 billion in 2013 to nearly $20 billion in 2020.

Through distributed intelligence, asset owners can improve their situational awareness needed to more effectively monitor and control remote distributed resources without having to loop back to the control center and out to the grid again. Those round-trip decision-making processes through a central office can take several seconds and even minutes. Systems may need to shut down unnecessarily. With interoperable, intelligent devices and systems at the local level, decision making is faster and better tuned for that specific location.

How do we create interoperable communication and decision-making capabilities at the grid edge? One intriguing approach is explored in Duke Energy's Distributed Intelligence Platform (DIP) Reference Architecture. This well-received technical vision provides a blueprint for a platform that pushes computing power closer to grid-edge distributed resources while leveraging non-proprietary standards-based solutions. Vetted and tested through Duke's Coalition of the Willing, a consortium of smart grid equipment vendors who unlocked their technology's interfaces so they could integrate with other vendors' technologies using a reference architecture framework, the Duke architecture relies on a key technology enabler: an open field message bus.

At the SGIP Winter meeting in 2014, Stuart Laval from Duke Energy's Emerging Technology Group approached the Smart Grid Interoperability Panel (SGIP) to expand on their vision and codify an Open Field Message Bus (OpenFMB) approach into a set of standards, recruit other utility partners and facilitate the creation of an Internet of Things (IoT) ecosystem for utilities. Similar to how the Green Button initiative provides a common application programming interface (API) on the customer side of the meter, SGIP's OpenFMB will provide a common interface for grid devices and target new applications that enable interoperability on the utility side of the meter. The resulting SGIP project, OpenFMB, was launched this past February at DistribuTECH, and a fast-tracked approach to getting standards in place is now underway.

Where the bus is headed
For the non-engineers who might be reading this article, a good definition of a message bus appears in the book titled Enterprise Integration Patterns by Gregor Hohpe and Bobby Woolf. According to this team, a message bus is a combination of a common data model, a common command set, and a messaging infrastructure to allow different systems to communicate through a shared set of interfaces.' According to DOE's Chris Irwin, "The OpenFMB framework will provide a common bus to allow utilities to communicate across the common utility languages like IEEE 1547 for inverters, IEC 61850 for substation automation, ANSI C12 for meters, and IEC 61968/70 Common Information Model (CIM) for Enterprise back office integration."

In addition to a space already crowded with existing standards that struggle to complement each other, the industry lacks a one-size-fits-all software or hardware technology that can adequately enable true interoperability between new Distributed Energy Resource (DER) systems and the existing grid automation infrastructure. Consequently, an abstraction layer between devices in the field is needed to simplify the integration of our existing systems that are traditionally very difficult and complex to maintain and scale in the utility back office.

The overall vision for distributed intelligence architecture and unlocking the edges of the grid requires that the message bus is in the field, out on the grid, not in centralized data centers. And, in order to gain traction and allow true interoperability across devices and across vendors, the bus must be open. It leverages the same kind of standards used by IoT and data model standards such as IEC 61850, a substation automation standard from the International Electrotechnical Commission, IEC's 61968 Common Information Model (CIM) that supports exchange of information about an electrical system, or MultiSpeak from the National Rural Electric Cooperative Association, which was chosen by the National Institute of Standards and Technology (NIST) as a key operations standard in NIST's Smart Grid Standards Framework and Roadmap.

With this history and Duke's real-world experience with its reference architecture, SGIP is now taking the lead in tackling three different initiatives for 2015. First, we are working on two standards that will outline an OpenFMB framework specification and the details of a reference implementation. The specification will include the OpenFMB framework structure and approach for implementing field message bus devices in an interoperable way. The reference implementation will include a working example using existing industrial and IoT protocols with existing utility data models.

The SGIP OpenFMB team also is designing a use case to be evaluated at three different test beds. And, finally, we're supporting collaboration among different test-bed efforts through a related catalog of test bed projects that will provide a centralized repository and characterization of North American Smart Grid test beds. Each of these activities will contribute to faster access to standards and frameworks that facilitate peer-to-peer communication on the grid.

Fast-tracking standards development
Typically, standards take three to seven years of committee work, debate and voting before they are ratified. Yet, SGIP aims to have two standards written in eight months of work and voted on during first quarter of 2016.

Here is how we're doing it: using Duke Energy's reference architecture as a starting point, the OpenFMB team has begun working with the North American Energy Standards Board (NAESB) to support NAESB's standards-development processes. The SGIP's multi-stakeholder team supports the standards process with crucial requirements, quick feedback on documents as they are created, and verification that the requirements we've identified are being addressed.

Our OpenFMB team represents all kinds of players within the smart grid industry. We have asset owners, manufacturers of hardware, software and firmware, governments, testing and certification organizations, as well as regulators to represent consumers.

We also have a number of leading utilities participating, including the largest investor-owned utility in the U.S. - Charlotte, N.C.-based Duke Energy - as well as the nation's largest municipal utility - CPS Energy in San Antonio, Texas - and the nation's largest cooperative - Pedernales Electric - which also is in Texas. Other notable participating utilities include Ameren Services, American Electric Power, BC Hydro, DTE Energy, Southern California Edison and Southern Company Services.

Along with these utility giants, we have some major industry giants: ABB, General Electric, Itron, National Instruments and Omnetric Group. In addition, the team includes thinkers from some of the nation's top laboratories and research organizations, such as the Pacific Northwest National Lab (PNNL), the National Renewable Energy Lab (NREL), FREEDM Systems Center and the Electric Power Research Institute (EPRI). A complete list of players appears in the sidebar (on page 14).

Representatives from all of these organizations are coming together to contribute ideas and expertise, which SGIP is coordinating into comprehensive input for the standards writers. The team meets weekly by teleconference as well as through in-person meetings, with the most recent one held at the National Renewable Energy Lab in Golden, CO. SGIP is acting as the driving force, making sure the standards-creation process includes utilities, vendors and test beds to fully leverage prior work and create a truly interoperable open field message bus framework. The standards informed by this multi-stakeholder group will be comprehensive enough to serve the broad array of industry participants.

Putting it to the test
The SGIP team also is creating microgrid use cases to test at three different microgrid test bed sites - NREL, CPS Energy, and Duke Energy. We are working with organizations such as UCA International Users Group (UCAIug), Electric Power Research Institute (EPRI), and the Industrial Internet Consortium (IIC) to help support and coordinate the testing activities. Already, other test beds have approached SGIP expressing an interest in the use cases. The idea is that no test bed will test the exact same configuration and equipment but will use the OpenFMB framework to show true interoperability across equipment, vendors and even technologies.

The use cases have already been defined and UML models are under development. The first deals with the unintentional islanding from grid-connected mode due to one of two scenarios: a larger grid outage or a security threat.

In the first scenario, a grid outage is detected by an island recloser at the point of common coupling (PCC) and initiates the unintentional islanding transition. In the second scenario, a security event detected by the security platform notifies the island recloser at the PCC to start the unintentional islanding transition. Upon opening of the recloser at the PCC, the battery inverter receives the recloser open status and switches from current-source mode to voltage-source mode, which initiates the microgrid controller to begin optimizing the local microgrid in island mode. The microgrid controller calculates how long it can remain in islanded mode and provides regular status updates to the headend systems.

The second microgrid use case deals with normal daily operations of a microgrid, both grid connected and islanded. The use case models activity for day-ahead and intra-day scheduling, and it also examines internal microgrid optimization as well as optimization with power export.

The use cases will be finalized in May, setting the stage for an exciting summer as SGIP OpenFMB team members work with utility engineers at the three different microgrid test beds. We will coordinate the effort, working with utilities, test beds and vendors to support the use case implementation. The busy summer will be followed by a busier autumn when demonstrations are conducted at EPRI in September and at the SGIP annual conference the first week of November in New Orleans.

Who's doing what with test beds?
In conjunction with the OpenFMB test bed coordination activities, the SGIP team is developing a Catalog of Test Beds for North America. We have a list of 52 different smart grid test beds in North America and are conducting surveys of test bed leads to gather information on the type of testing the group is doing, the equipment in place, how others can collaborate on the project, whether there are costs for participation, the level of testing underway, simulation capabilities and other details. The information gathered will help identify opportunities for collaboration across test beds and identify gaps in testing that need to be filled. The SGIP plans to assist its utility and vendor members to locate opportunities for hands-on interoperability field testing on their own systems and equipment.

SGIP's efforts stem from the belief that microgrids and distributed resources are capable of providing even greater potential with a mixture of centralized command and control and distributed intelligence. But, SGIP members and staff also recognize the interoperability gaps that currently exist for utilities. As power delivery becomes increasingly localized, distributed intelligence will move from a nice-to-have notion to a must-have requirement for optimized grid operations.

Ultimately, all SGIP activities - the Catalog of Test Beds, the microgrid use cases, and the standards-writing support - will be completed by SGI, by year-end 2015. This is an aggressive schedule, but the team is confident and truly engaged as a collaborative machine with field device interoperability and distributed intelligence as the goal. SGIP is proud to be part of the efforts that will codify these requirements into industry standards.

If you're interested in learning more about the OpenFMB project and joining this effort, please contact info@sgip.org.

About the Author

A prominent contributor and technologist in the utility industry, Stuart McCafferty has lead large multi-stakeholder collaborative programs. He is the VP of Operations at SGIP where he oversees all operational aspects of programs to advance and accelerate grid modernization. In 2013, he received the international Distinguished Project Award from the Project Management Institute (PMI) for his work on SGIP programs.