MicrogridsEXPANDING APPLICATIONS,

IMPLEMENTATIONS, AND BUSINESS STRUCTURES

DECEMBER 2016

EPRI ID: 3002008205

2 SEPA & EPRI | WHITE PAPER

MICROGRIDS

TABLE OF CONTENTSINTRODUCTION ...........................................................................................................................................................................3

DRIVERS OF MICROGRID ADOPTION ....................................................................................................................................6

MICROGRID IMPLEMENTATION TYPES ................................................................................................................................6

MICROGRID BARRIERS AND CHALLENGES .........................................................................................................................8

MICROGRID SOLUTIONS FOR DIFFERENT CONTEXTS ................................................................................................. 12

CONCLUSIONS ......................................................................................................................................................................... 16

APPENDIX: RECOMMENDED READING ............................................................................................................................. 19

COPYRIGHT© Smart Electric Power Alliance and the Electric Power Research Institute, 2016. All rights reserved. This material may not be published, reproduced, broadcast, rewritten, or redistributed without permission.

AUTHORSNadav Enbar, Principal Project Manager, EPRI

Dean Weng, Senior Engineer, EPRI

Ryan Edge, Program Manager, SEPA

John Sterling, Senior Director, SEPA

ABOUT EPRIThe Electric Power Research Institute (EPRI) conducts research and development for the public benefit that is related to the generation, delivery, and use of electricity. An independent, nonprofit organization, the Institute brings together scientists, engineers, and experts from academia and industry to help address challenges in electricity.

ABOUT SEPAThe Smart Electric Power Alliance (SEPA) facilitates collaboration across the electric power industry to enable the smart deployment and integration of clean energy resources. Our focus centers on solar, storage, demand response, and other enabling technologies.

ACKNOWLEDGEMENTSSEPA and EPRI would like to thank the following people for their time and collaboration on this report: Jason Abiecunas, Black & Veatch; Peter Asmus, Navigant Research; Joe Herr and Chase Sun, Pacific Gas and Electric; Esrick McCartha, PJM Interconnection; and Howard Smith, Southern Company. SEPA would also like to thank Christine Stearn and Dami Soyoye, for their written contributions to the report, and K Kaufmann, Mike Kruger, Erika Myers, and Mike Taylor for their review. EPRI would also like to thank Tom Key and Aridam Maitra.

MICROGRIDS 3

IntroductionHistorically, microgrids have been employed as an additional layer of reliability for customers in remote locations with limited access to the grid, or for large institutions managing a campus-style energy system. However, efforts to modernize the electricity system to more effectively leverage rising penetrations of interconnected distributed energy resources (DERs), accommodate increased customer choice, provide critical or emergency services, and enable greater grid resiliency in response to more frequent extreme weather events, are stimulating interest in traditional and unconventional microgrid applications.

As shown in Figure 1, microgrids, working in concert with traditional utility infrastructure, can capture the benefits of the emerging “Grid 2.0”1 system, while mitigating associated challenges. For example, a parallel, bidirectional connection can improve reliability, lower costs, and diversify energy sources. At the same time, this connection can provide the option to separate, or “island,”

from the grid, enhancing resiliency via backup or emergency operation. In certain contexts, the combined value of resiliency, grid integration of intermittent resources (through aggregated control of multiple DERs, including demand side resources), and participation in electricity markets may justify microgrid development for electric utilities, producers, and end users alike. It may further contribute to a future grid that is more transactive in nature.

To realize the potential of microgrids, an increasing number of projects are exploring a range of structures and business models that could facilitate greater deployment and integration of these systems into the broader power grid across a variety of circumstances. Accrued experience and learning—along with policy and economic alignment—will likely raise the profile and value of microgrids as a way to safely meet future electricity needs, enable greater system flexibility, and help modernize the grid.

1 EPRI and SEPA, respectively, refer to the grid of the future as the Integrated Grid and the 51st State.

Source: EPRI, 2016

Note: The Utility PCC stands for the point of common coupling, a point in the electrical system where multiple customers or multiple electrical loads are connected.

FIGURE 1. MICROGRID AS PART OF A TRADITIONAL UTILITY SYSTEM

4 SEPA & EPRI | WHITE PAPER

MICROGRIDS

The evolving power system will require the coordination of an increasingly complex set of generation, delivery, and energy management assets. It will also require the development of novel processes, protocols, and business strategies to ensure that operational safety and grid reliability

objectives are met, and that any associated benefits are fairly distributed. Microgrids offer three main business models for integrating disruptive “grid edge” and grid automation technologies in ways that can help realize the underlying vision for the grid of the future.

WHAT IS A MICROGRID?The term “microgrid” is sometimes loosely used to describe a number of concepts involving distributed generation (DG). However, the industry-accepted definition, from U.S. Department of Energy (DOE), describes a microgrid as:

“. . . a group of interconnected loads and distributed energy resources (DER) within clearly defined electrical boundaries that act as a single controllable entity with respect to the grid, and that can connect and disconnect from the grid to enable it to operate in both grid-connected and ‘island’ mode.”2, 3, 4

The microgrid concept is not new; it is simply a reformulation of local power systems as they were originally designed. Like the traditional, centralized electric grid, microgrids generate, distribute, and regulate the supply of electricity to customers, but do so locally and on a much smaller scale. Microgrids provide “grid-interactive solutions”5 to utility grid flexibility and resiliency challenges associated with meeting the demand for continuous supply of electric power. Figure 2 illustrates examples of substation- and feeder-based microgrid architectures under this definition.6

Importantly, microgrids are not a replacement for utility distribution infrastructure. Instead, microgrids

form a self-contained organization of DG and load management that is capable of self-balancing, when necessary, within an isolatable portion of utility or non-utility infrastructure. Individual microgrids usually operate in a grid-tied mode, with power flowing both ways between the microgrid and the surrounding system. The option to separate from the grid provides a backup or emergency operation mode, as well as the opportunity for greater DER investments and coordination.

Microgrids may contain dispatchable and non-dispatchable renewable generation, controllable loads, energy storage, electric vehicle-to-grid (V2G) discharging, and advanced grid modernization technologies, such as advanced metering infrastructure (AMI) and distribution automation.7 However, in most cases to date, thermal generation provides the primary power supply, using local waste fuels, diesel, or natural gas, and often configured for combined heat and power (CHP). Also, legally required backup and emergency systems, such as diesel generators in hospitals, may operate in parallel with the grid. In fact, some of the basic building blocks of microgrids may already reside in many networks.

2 “Summary Report: 2012 DOE Microgrid Workshop,” DOE EERE, Chicago, 2012.

3 T. Glenwright, “http://www.smartgrid-live.com/wp-content/uploads/2012/12/Introduction-to-Microgrids-by-Tristan-Glenwright.pdf.,” December 2012.

4 The CIGRÉ C6.22 Working Group definition is also useful: Berkeley Laboratory. Microgrid Definitions. Available at: http://building-microgrid.lbl.gov/microgrid-definitions.

5 NEMA MGRD 1-2016 Microgrid Primer. Jim Reilly, 2016.

6 Note that the term “nanogrids” falls under the definition of microgrids. Nanogrids are small microgrids that typically serve a single building or a single load.

7 In the context of a microgrid, grid modernization elements can include aggregated DER controls, such as a microgrid controller, load control, distribution asset monitoring and control (especially as it relates to protection), load/solar forecasting, economic optimization, and communication/interaction with the distribution system operator, transmission system operator (DSO/TSO).

MICROGRIDS 5

FIGURE 2. EXAMPLES OF MICROGRID ARCHITECTURE ON A RADIAL DISTRIBUTION SYSTEM

Source: EPRI, 2016

Note: Generation indicated in the graph includes distributed energy resources.

By definition, microgrids have a connection to the utility grid and the ability to island. A virtual power plant (VPP) or other shared metering arrangements may contain elements of a microgrid but are not considered to be one.8 VPPs typically provide the automation and control features required to remotely and automatically dispatch and optimize generation, load controls, and storage. However, VPPs do not operate within defined grid infrastructure boundaries. Similarly, an advanced energy community (AEC), or an electrically contiguous set of similar buildings that have implemented various DERs, also differs from a microgrid.9 Although an AEC aggregates services,

such as demand response, to the larger grid, it is not necessarily capable of islanding—a core microgrid feature.10

Separately, a utility may have the ability to reconfigure distribution feeders in response to contingencies, using looped or networked architecture. These feeders can actively reroute circuitry, but do not achieve islanded status. This is also not considered a microgrid due to the absence of local generation and the inability to balance generation and load within the area. Local generation that only operates within an interruptible load agreement, or on an emergency or standby basis, is also not a microgrid.

WHAT A MICROGRID IS NOT

8 A VPP can comprise distributed power stations such as wind farms, CHP units, photovoltaic (PV) systems, small hydropower plants and biogas units, as well as loads that can be switched off, in order to form an integrated network. Unlike microgrids, VPPs do not operate within defined boundaries.

9 An AEC would, for example, include zero net energy (ZNE) communities with integrated demand response or storage, in addition to PV, energy efficiency, and connected loads.

10 An aggregation platform with local controls, in many cases enabled through cloud-based data acquisition and controls operation, is typically used to supervise AECs, as opposed to a single central controller.

6 SEPA & EPRI | WHITE PAPER

MICROGRIDS

Drivers of Microgrid AdoptionEmerging lower-cost DER technologies, environmental concerns, and a growing reliance on uninterrupted electric service are increasing the potential value of microgrids and their available applications. Leveraging the broader power grid, a microgrid can help overcome power delivery situations with high-risk exposure and limited grid access. Meanwhile, advances in energy efficiency, DG, and storage technologies combined with energy management systems can help overcome the challenges of operating a small power system.

Table 1 lists the notable features and characteristics that support the business case for microgrids. The

value drivers listed in the table offer monetizable benefits. These drivers correspond to the assortment of applications that a microgrid can deliver. The technology and policy drivers contribute to the relative attractiveness of available microgrid applications. When combinations of these drivers are present, a well-designed and applied microgrid may increase reliability and reduce emissions at a relatively low incremental cost.11 Implemented correctly, a microgrid can enable a self-reliant customer (or group of customers) to operate during grid contingencies and to use the larger grid as a resource for energy balance, reliability, and access to energy markets.

Microgrid Implementation TypesMicrogrids are typically customized to their purpose and location, and thus the details of their design and construction vary significantly. For example, the proximity of critical loads, configuration of the surrounding electricity grid and thermal infrastructure, and availability of existing DERs inform microgrid design decisions as well as the merits of a microgrid’s intended applications.

The different types of microgrid implementations have expanded over the past decade (see Table 2). Today, various stakeholders are deploying microgrids to, among other things, enhance reliability for remote areas with limited grid access; serve military installations with energy security concerns; support data centers, hospitals, and other industries with critical up-time or power quality requirements; and help reduce the carbon footprint of university campuses via DER integration.

Any microgrid implementation could exhibit the characteristics of more than one use case. Likewise, individual technologies are not exclusive to any application. For example, military and campus-based microgrids share common attributes (e.g., both enhance reliability for a single customer in a confined geographic area). Military applications, however, require more stringent external hardening. In either case, a microgrid could also aggregate DERs while grid connected and provide backup power when islanded.

11 “Microgrids Evolution Roadmap - Working Group C6.22,” CIGRE Working Group C6.22, 2012.

MICROGRIDS 7

TABLE 1. DRIVERS OF MICROGRID DEVELOPMENT

VALU

E D

RIVE

RS

RESILIENCY/RELIABILITY

Microgrids provide contingent electric service capacity when the grid is unavailable. They can also serve as an alternative to ensuring power quality through traditional measures, such as the use of a large uninterruptible power supply in a data center.

ENERGY/CAPACITY/GRID SERVICES (ENERGY COST SAVINGS)

By aggregating local resources, microgrids can provide reserve and regulation services (i.e., real and reactive power) to the electrical grid as a means of capturing economic value. Examples include demand response capacity (which has increased sharply since 2010), local generation sources, energy efficiency, and load management control. Microgrids can use these resources to exploit time-of-use pricing to lower consumer costs through energy arbitrage, or provide needed energy/capacity to the grid.

DER INTEGRATION/AGGREGATION

Microgrids can aggregate and manage a number of installed DERs and operate them as a single entity, from the utility or grid operator perspective, to provide ancillary services, for example. This aggregation capability can inform grid planning and operations.

EMISSIONS REDUCTION

Carbon emissions reductions can be attained through the alternative use of a microgrid’s renewable (or higher efficiency) DERs compared to conventional generation assets—particularly within specific contexts.

ENERGY

Microgrids offer the capability to island for long periods of time and provide an uninterruptable power supply to entities of high national interest or to remote communities. For the former, the increased reliability can serve heightened cybersecurity needs. More generally, renewable generators mitigate risk from petroleum supply volatility or disruption.

INVESTMENT ALTERNATIVE/DEFERRAL

Utilities can use local resources within a microgrid to reduce grid energy use during peak hours, and in turn, defer distribution asset upgrades. These local resources can also provide an alternative investment strategy to bolster resiliency and reliability improvements in place of conventional measures (e.g., supplanting the need for a redundant transmission line).

TECH

NO

LOG

Y/PO

LICY

DRI

VERS

TECHNOLOGY AND COST

In the past decade, renewable energy (e.g., wind turbines, photovoltaic (PV) solar) costs have decreased significantly, while performance has improved. The cost of traditional forms of DG (e.g., internal combustion engines, small combustion turbines) and energy storage has likewise declined due to technical advances and economies of scale. Energy conversion to electricity close to end users is also creating greater opportunities for waste heat capture in CHP systems. Microgrids can capture all of these values.

ADVANCED CONTROLS AND POWER ELECTRONICS

New power electronics technologies, such as power conditioning and control, are improving inverters (needed for direct current sources, such as PV and battery storage, that are becoming prevalent in microgrid installations). By providing grid support functions, these technologies are enabling better DER integration and control.

PUBLIC POLICY

Current public policies and incentives—which include renewable energy tax credits, renewable portfolio standards, grid modernization grants, emissions restrictions, and net metering—favor DG that offers improved efficiency, lower emissions, and enhanced power system security.

ENERGY USER AWARENESS

Enhanced energy user awareness of alternative power resources and their associated economics are increasing willingness to consider onsite generation options.

Note: * Because microgrid capacity is currently small, microgrids are today unlikely to be an attractive solution for reducing emissions at the state and federal levels. However, at the local level, a number of university campuses are embracing emission reduction targets, which can drive microgrid deployment.

Source: SEPA, EPRI, 2016

8 SEPA & EPRI | WHITE PAPER

MICROGRIDS

TABLE 2. EXAMPLE MICROGRID IMPLEMENTATION TYPES

GENERAL CHARACTERISTICS EXAMPLES

COMMERCIAL/INDUSTRIAL

§ Primary goal during normal operations: reduce demand- and consumption-related costs.

§ During outages: the operation of critical functions is paramount, especially for customers that operate continuous processes such as data storage and production lines.

Manufacturing, chemical processing, shipping and processing facilities, and data centers

COMMUNITY/CITY/UTILITY

§ Improve the reliability of critical infrastructure, defer asset investment, meet emission and energy policy targets, and promote community participation.

Municipal buildings, utility infrastructure, supermarkets, gas stations

CAMPUS-STYLE

§ Meet the high reliability needs of research labs, businesses, and campus housing

§ Enable reduced cost for large heating and cooling demands

§ Decrease emissions

§ Leverage existing assets (e.g., partially or completely owned distribution infrastructure and backup resources)

§ Serve as a power source for emergency shelters for surrounding communities during extreme events.

College and university campuses, business parks

PUBLIC/INSTITUTIONAL

§ Improve reliability and enable lower energy consumption at public health and safety facilities. The additional requirement of uninterrupted electrical and thermal service increases the attractiveness of CHP-based district energy solutions.

Hospitals, police and fire stations, sewage treatment plants, schools, correctional facilities, airports

MILITARY

§ Achieve high reliability for mission-critical loads

§ Meet pressing needs for cyber, physical, and fuel security at operating bases

§ Satisfy greenhouse gas (GHG) emission reduction goals

Training facilities, forward operating bases, proving grounds, airbases, naval bases

RURAL/REMOTE COMMUNITIES

§ Microgrids in remote communities are typically connected to rural distribution systems where constructing new transmission service for backup is cost-prohibitive due to distance or other physical barriers. Many locations currently use diesel generators for backup generation. Microgrids provide prime options for incorporating renewable energy, thereby improving system reliability targets, deferring investments, and reducing supply chain risk.

Remote and rural locations

Sources: SEPA, EPRI, 2016

MICROGRIDS 9

Microgrid Barriers and ChallengesAlthough the technology and equipment necessary for creating microgrids is available today, off-the-shelf commercial solutions are rare. A number of

technical, economic, and regulatory issues must be addressed to unlock the technology’s full potential.

TECHNICAL CHALLENGES n Establishing a Generation/Load Portfolio—Investing in a balanced energy portfolio can help support the economic value proposition for microgrids. Investment exclusively in a single technology will not allow for a permanent, grid-parallel DER installation. However, while a microgrid is islanded, incorporating a diversity of resources can support a wide range of priorities, including risk management, fuel cost hedging, and GHG reduction. Technical challenges surround efforts to combine inverter-based systems (e.g., PV, energy storage) with dispatchable, synchronous systems (e.g., combustion engines or turbines with or without CHP). Asset response times and ramping characteristics, for example, need to be coordinated. Moreover, local load characteristics (e.g. variability, baseload) typically dictate generation and storage technology selection and sizing.

n Connection and Protection—Once the energy sources are in place, the interconnection and protection technology must likewise be evaluated. Transfer switches and reclosers must be installed such that the system may be safely islanded at the point of common coupling (PCC) with the utility. Breakers and reclosers designed to operate with fault currents commensurate with the larger system connection may need to be redesigned for compatibility with a microgrid’s islanded mode. In this case, fault currents might be sufficiently lower, and coordination between devices might be needed.

n Grid Transitioning—Considerable technical challenges exist when toggling a microgrid between grid-connected and islanded modes. During transition to island mode, phase and frequency drift is highly likely, causing loads and distributed energy resources to trip. Without a finely calibrated synchronization process, grid reconnection could damage generators and loads within the microgrid and in surrounding systems.12, 13

n System Stability and Control—Balancing generation and load in the microgrid system is among the most difficult challenges for microgrids. The grid operations at large are based on the vast majority of generation occurring through rotating machines. Conversely, microgrids—especially those built around PV and battery storage—have orders of magnitude less inertia. Consequently, controllable loads and the algorithms in the microgrid controller may need to operate much more quickly in order to preserve energy balance and system stability. Alternatively, energy storage located on the system can provide additional support, but often at a higher cost.

n Power Quality—Because microgrids may result in higher short-circuit impedance compared to a larger system, total voltage harmonic distortion in microgrids could be significant.14 Similarly, capacitor-switching and other transients must be managed in order to avoid equipment damage.

12 B. Feero, D. Dawson and J. Stevens, “Protection Issues of The MicroGrid Concept,” Department of Energy (EERE), 2002.13 T. Peterson, “Distributed Renewable Energy Generation Impacts on Microgrid Operation and Reliability,” EPRI, Palo Alto, 2002.14 Higher impedance causes voltage harmonic distortion. Harmonics have a wide range of impacts such as communications interference

and degradation to conductors, motors, and transformers.

10 SEPA & EPRI | WHITE PAPER

MICROGRIDS

The development of a capable microgrid controller is expected to help solve many of these technical challenges. As the “brain” governing a microgrid’s entire operation, the controller is key to enabling core functions of grid transition, real-time local energy balancing, protection coordination, as well as ensuring power quality. Research and

testing efforts now underway are developing control architectures. To date, however, no commercially available controller can reliably perform the full suite of core functions. Consequently, most systems identified as microgrids prefer to maintain a grid connection whenever available.

ECONOMIC AND REGULATORY CHALLENGESAssigning value to microgrids is difficult due to an intersecting and fluid set of economic and regulatory issues. For example, regulatory and market uncertainties affect the upfront costs and life cycle economics of microgrids and associated DER technologies. A second challenge is that difficult-to-monetize or even non-monetizable microgrid costs and avoided costs (benefits) can complicate value stream calculations.The changing landscape of subsidies for renewables, such as the federal Investment Tax Credit and the Production Tax Credit, as well as market incentives, such as renewable energy certificates, influences the first costs of the resources that comprise a microgrid. Moreover, federal and state mandates (e.g., renewable portfolio standards), emissions reduction targets, trading systems, and incentives (e.g., the Regional Greenhouse Gas Initiative, Zero Net Energy frameworks, Clean Power Plan) can significantly impact DER costs. Unpredictable modifications to these and other constructs can considerably alter the economic outlook of DERs and, in turn, microgrids. Uncertainty about future electricity and fuel prices can also influence the long-term economic attractiveness of individual DERs within microgrids. For example, fluctuating fuel and electricity prices over the life of a natural gas-based CHP unit will change the frequency of its dispatch. A favorable spark spread will encourage use of a high-efficiency CHP unit, while the reverse might conversely motivate power purchase from the grid. Shifting price signals and their attendant structures—informed by incentive policies, market supply and demand, and regulatory directives—further shape the overall economic picture for DER technology types and sizes. The following can also inform DER dispatch:

n Utility tariff structures (e.g., flat, volumetric, time-of-use, or real-time rates; and demand charges)

n Market constructs (e.g., day-ahead, real-time, capacity, and ancillary service markets)

n Utility marginal cost considerations (in regions where markets do not exist)

Though fraught with challenges, developing supportive price signals, tariff structures, and regulations for DERs and microgrids at large is a necessary ingredient for the technology’s uptake.Further consensus and consistency will also be required around methodologies for evaluating the contextual financial benefits of reliability and resiliency. A data center or manufacturing facility may calculate revenue loss from a given power outage event to justify the cost of a microgrid. But evaluating avoided costs for maintaining critical services (e.g., hospital care, national security, and public safety) is a more challenging undertaking. Avoided cost calculations, regardless of their context, require a range of assumptions about the expected frequency and magnitude of grid disturbances over the life of a microgrid’s operation. Ownership of a microgrid’s generation equipment and wires can itself pose concerns. Ownership is relatively straightforward if the generating entity is providing power only to its own buildings and facilities (e.g., in the case of a university or an industrial complex). However, if multiple entities cooperatively create a microgrid, with some entities generating and delivering electricity to other entities, the economic framework for sharing costs and ownership rights as well as accruing benefits across multiple users can become complex.

MICROGRIDS 11

THE CURRENT LANDSCAPECurrent technical standards offer guidance to microgrid development, but do not address more nuanced issues germane to microgrid design. For example, further definition is required for protocols governing advanced protection coordination, multi-layer/device communications and controls, microgrid-to-grid interactions, and grid resynchronization. Today, DERs within microgrids must comply with interconnection standards such as IEEE 1547-2003 (“Standard for Interconnecting Distributed Resources with Electric Power Systems”). Such standards provide guidance on utility voltage and voltage-regulating equipment, overcurrent protection, effective grounding, islanding prevention, harmonics, voltage flicker, telemetry and metering equipment, among other issues which have direct relevance to microgrids operating in grid-tied mode.15

Specific standards (e.g., IEEE 1547.4) also provide guidance on microgrid design, implementation, and operations for purely islanded systems. Other standards, such as IEEE 2030.2, address interoperability issues specific to energy storage and the power grid, and are likely to impact future microgrid controls and communications, and inform microgrid development.16

FUTURE DIRECTION OF MICROGRID STANDARDSMicrogrid standards still require further definition of communication and controls, and their

associated testing procedures. With advances in monitoring and control systems, microgrids can now be established with controlled islanding—a key benefit for resiliency.17 However, to accomplish this feat, microgrid control systems must be able to satisfy local utility requirements, such as voltage and frequency limits and ride-through capabilities.

The pending IEEE P2030.7 (“Standard for the Specifications of Microgrid Controllers”) covers the control functions that define the microgrid as a system that can manage itself, operate autonomously or grid-connected, and seamlessly connect to and disconnect from the main distribution grid for the exchange of power and the supply of ancillary services.18 P2030.8 (“Standards for theTesting of Microgrid Controllers”) is being developed to establish standardized microgrid controller testing procedures that allow for component and control algorithm flexibility and customization, while ensuring that minimum requirements are met.19 These standards are intended to (1) address the technical issues and challenges of microgrid controller operation that are common to all microgrids, and (2) present the control approaches required from the distribution system operator and microgrid operator.20

Other notable efforts underway include IEC TS 62898-1—“Guidelines for the General Planning and Design of the Microgrid,” and 62898-2 Ed. 1.0—“Technical Requirements for Operation and Control of Microgrids.”

DEVELOPING AND APPLYING MICROGRID STANDARDS

15 More than 40 states have adopted IEEE 1547-2003 to guide the interconnection process.

16 An outgrowth of the Smart Grid Interoperability Panel (SGIP) framework, this standard allows alignment to use case analysis. Among the issues it addresses are power system interoperability, communication system interoperability, information system interoperability, security and privacy, interoperability and cyber security—all topics germane to microgrid development.

17 Due to operational and safety concerns, local intentional islanding has historically only been conducted after effective isolation from the utility grid.

18 P2030.7 was approved in mid-2014, but is not yet officially published.

19 P2030.7 was approved in June 2015, but is not yet officially published.

20 The scope of P2030.7 does not cover interconnection requirements for individual DERs and single controllable entity microgrids; it pertains only to multi-entity microgrids and microgrid controllers.

12 SEPA & EPRI | WHITE PAPER

MICROGRIDS

Microgrid Solutions for Different Contexts Microgrids are contributing to the electric grid’s steady evolution toward DERs, flexible loads, and increased resilience and security. The potential roles and structural arrangements of microgrids are changing, however, as part of the larger utility system reformation. Meanwhile, a diverse set of utility and other business models are emerging that support microgrid development and reflect the shifting landscape of generation and delivery services.

The specific context of each microgrid installation ultimately informs its appropriate business model. Generally, the structural arrangement of a microgrid varies depending on a number of dimensions, some of which are illustrated in Figure 3. Core considerations include:

n Who owns and controls the microgrid and its component resources?

n Does the microgrid interface with a wholesale market or is it tied to a vertically integrated utility?

n Is the microgrid connected to a distribution-only utility? Is the entity a full- or partial-requirements customer of another utility?

n Does the microgrid serve a single customer or many customers?

n What technologies comprise the microgrid?

These questions bring utility rate issues to the forefront with respect to different microgrid business models. For installations owned by and serving a single customer, rate impacts may be easily confined. For microgrids that serve multiple customers, cost allocation becomes murky, especially for community resiliency applications designed for societal benefit beyond the encompassed customers. How should these costs be allocated to a utility’s rate base? Should customers outside the installation have to pay to reinforce hardware that does not ultimately connect to their meters? What if no widespread outage occurs and the microgrid never islands?

Another rate issue arises when a microgrid encompasses multiple customers, and the utility does not own the onsite generators. When these microgrids island, a third party sells power to end users, a practice that may violate the utility compact in some jurisdictions. These examples demonstrate how DER technologies are outpacing utility regulatory oversight, perhaps at the expense of customers, when proceedings restrict DER deployment.

MICROGRID BUSINESS MODELSUntil recently, a large portion of microgrids have been third-party installations serving a single customer. However, utility-owned microgrids are also being developed, primarily due to state-level policies and directives. In tandem, technology maturity and expanding microgrid applications are also facilitating their development. Figure 4 lists high-level features of the three main microgrid business models that have emerged in the current market, with their relative degree of utility control.

The Third-Party Model places control of the microgrid in the hands of end users and/or their solution providers. The microgrid can be owned directly by the end users or indirectly through an ownership vehicle or entity; meanwhile, the end users or utility can own the encompassing distribution system. Similarly, the physical operations of the microgrid controls (including DERs) may be operated directly by the end users or contracted to a third-party operator.

MICROGRIDS 13

FIGURE 3. SAMPLE DECISION TREE FOR DETERMINING THE STRUCTURE ARRANGEMENT OF A MICROGRID

Premium Reliability/ Resilience District Regulated Rates

3rd Partyor Utility?

Provides Continuity of Essential Services

for Local Area. One Customer, Multiple Facilities,

Regulated Rate @ PCC

Mini-Utility withInternal Customers,

Behind Meter @PCC Internally Unregulated

Premium Reliability/ Resilience District, Regulated Rates

PUBLICUTILITY

One Customeror Mini-Utility?

NON-UTILITY3RD PARTY

MINI-UTILITY

ONECUSTOMER

Source: EPRI, 2016

Notes: PCC = point of common coupling; Public Utility = public service utility; Mini-utility = third party microgrid that serves multiple customers.

Sources: SEPA and EPRI, 2016

FIGURE 4. MICROGRID BUSINESS STRUCTURES ARRANGED BY LEVEL OF UTILITY CONTROL

THIRD-PARTY MODEL n End user(s) or 3rd party own and finance microgrid

n End user(s) or 3rd party determine economic dispatch (potentially with utility guidance)

n Utility, end user(s) or 3rd party agree on appropriate islanding conditions

n End user(s) see net change in bills

UNBUNDLED MODEL n Utility or 3rd party owns and finances microgrid on behalf of end user(s)

n Utility or 3rd party dispatches DER assets on behalf of customer(s)

n Utility and end user(s) agree on appropriate islanding conditions

n End user(s) pays utility for grid assets, pay implementer (utility/3rd party) for microgrid assets, receives credit from DER

INTEGRATED UTILITY MODEL

n Utility owns and finances microgrid

n Utility dispatches DER assets based on system economics

n Utility and end user(s) agree on appropriate islanding conditions

n End user(s) pays utility for resiliency/premium power service

CUSTOMER CONTROL

UTILITY CONTROL

14 SEPA & EPRI | WHITE PAPER

MICROGRIDS

When grid-tied, microgrid assets can be dispatched primarily for the economic benefit of the end users (e.g., to perform demand charge reduction), but may respond periodically to favorable price signals from the utility or power markets, such as a demand response event. During an emergency outage, the primary objective of the microgrid is to isolate from the power grid and maintain end-user loads; this supersedes all other objectives of the utility or grid operator.21 In recent years, some third-party engineering, procurement, and construction management companies and operators have begun to offer “microgrid-as-a-service” products. In this arrangement, the end user is completely divorced from the microgrid’s financing, construction, and operations, and pays a fee to reap the resulting microgrid benefits.

The Unbundled Model encompasses the most wide-ranging set of possible ownership and operating arrangements for microgrids, and is less established than the other two models depicted in Figure 4.

As its name suggests, the model’s structure is unbundled with respect to ownership, financing, operations, and transactional configuration. One entity—the utility, end user, or third party—owns or finances the microgrid, while another entity operates the microgrid. Meanwhile, both grid-tied and islanding operations can have single or multiple objectives. For example, a third party operating the

microgrid might be free to dispatch DERs for the benefit of end users, subject to certain distribution or bulk system constraints that the utility or grid operator sets. Financial gains can flow to a single entity or multiple parties. For example, a traditional customer-utility payment structure could be established, as could a split-payment structure.

The Integrated Utility Model places ownership and control of all of the microgrid infrastructure as well as generation and storage assets with the utility. When grid-tied, microgrid operations prioritize the objectives of the bulk system (e.g., exporting energy back to the power grid), distribution system (e.g., peak shaving to prevent violation of thermal limits), and/or the end user (e.g., providing heat from a CHP unit or reducing local GHG emissions).

The utility is likely to have a contract with the end user, stipulating conditions under which the microgrid will island. As opposed to the third-party model, the utility can opt to provide support via demand- or supply-side demand response, for instance, rather than island the end user if doing so can prevent a larger grid outage. The utility can also use the microgrid for blackstart support for sections of the power grid after an islanding event. From the end user’s perspective, the customer-utility relationship fundamentally does not change: the end user pays according to a tariff that now includes a premium for the microgrid service.

REPRESENTATIVE MICROGRID PROJECTSAccording to GTM Research, roughly 1.6 gigawatts (GW) of operational microgrid capacity has been cumulatively installed to date, and the market is forecasted to reach 4.3 GW by 2020.22 Fossil fuels power the majority of today’s installations and continue to serve military and university campus applications under the third-party business model. However, the increasing diversity of project arrangements indicates a maturing understanding

of the contexts within which microgrids provide value. In fact, the known pipeline of planned microgrid generation capacity appears to favor a more diverse resource portfolio, with increased inclusion of renewables, in addition to more varied use cases and applications.

Figure 5 presents four high-level examples of microgrid implementations representing the range of projects operating today. Although enhanced

21 The microgrid would likely still be required to inform the utility and/or grid operator when disconnecting or reconnecting to the main grid.

22 U.S. Microgrids 2016: Market Drivers, Analysis and Forecast. GTM Research, Boston, MA: August 2016.

MICROGRIDS 15

FIGURE 5. REPRESENTATIVE MICROGRID IMPLEMENTATIONS

Source: SEPA, EPRI, 2016

Note: CHP includes gas-fired engines, turbines, microturbines, and fuel cells; DG = diesel generators; ES = battery-based energy storage; PV = solar photovoltaics.

16 SEPA & EPRI | WHITE PAPER

MICROGRIDS

reliability is a common objective across the project examples, and applications and featured technologies overlap, each project embodies unique design and operational characteristics that are the result of their business models.

For example, the 29 Palms and Illinois Institute of Technology projects operate under the third-party model and have been designed to broadly meet end-use reliability objectives. While the former is focused on grid hardening and resiliency as a primary means for securing energy supply, the latter is more directly intended to reduce energy costs and peak load.23

The Borrego Springs Substation project is organized under the integrated utility model. It is operated under San Diego Gas & Electric’s (SDG&E) directive to enhance grid resiliency for a remote desert community of 500 residential, commercial,

and industrial customers in a way that is more economical than creating grid line extensions. The utility is assessing the feasibility of “graceful degradation,” a practice originally developed in telecommunications, in which functions can be diverted to customers with critical needs. SDG&E is also testing price-driven response systems using in-home storage, plug-in electric vehicles, smart appliances, smart meters, and home area networks to reduce peak demand by 15 percent.

The Rutland community microgrid represents a project organized under the unbundled model, in which Vermont’s largest power company, Green Mountain Power, has partnered with NRG Energy. The installation is intended to incorporate DERs and distribution equipment to supplant conventional investments in aging infrastructure. The microgrid aims to enable peak shaving and provide backup power to critical facilities in grid-islanded mode.

Conclusions Though individually unique, microgrids tend to consist of similar technologies that provide different benefits to the customers they serve or the grid at large. Recent and anticipated advances in microgrid controllers and DER management systems are further broadening the values that can be realized from these installations. To support the utility grid’s evolution, initiatives such as SEPA’s 51st State and EPRI’s Integrated Grid are exploring the effects of potential market, technology, and regulatory changes that may arise in the foreseeable future.

Given that microgrids can encompass virtually any DER technology, research findings and associated lessons learned for ongoing research and development are anticipated to have broad application to many other DER solutions. EPRI’s portfolio of microgrid projects conducted under the banner of its Integrated Grid initiative is, for example, advancing research in the areas of power system modeling,24 controls and communications,25 device testing,26 grid integration, and cost-benefit analysis. EPRI microgrid projects in various stages of

23 Prior to its microgrid project, IIT incurred an average of three outages per year and approximately $500,000/year in total damages, time lost, and restoration costs. For more information, see http://www.iitmicrogrid.net/microgrid.aspx.

24 Research encompasses steady state, dynamic, fault, and transient analyses, for both the microgrid itself (during islanded/grid-connected) as well as in the boarder context of central grid planning and operations.

25 Research encompasses control schemes (e.g., DER control modes and settings, protection coordination, transition to island, resynchronization, black start, and economic dispatch), microgrid controller architecture, distribution system architecture, device communication interfaces, etc.

26 Hardware, Hardware-in-the-Loop (HIL)

MICROGRIDS 17

implementation currently span diverse geographical locations (nine states); residential, commercial, industrial, public, and military sectors; integrated, nonintegrated, and self-regulated utility and regulatory structures; and various use cases and business models.

The broader context of existing and future regulatory models, and market frameworks unique to each utility footprint will significantly influence how microgrid use cases and business models evolve over the long term. As seen in the wide range of submissions received as part of the SEPA’s 51st State, the possibilities for how DERs

might be deployed and participate on the grid of the future are limitless.27 However, a finite set of decision points will drive those discussions. In time, these decision points will affect how customer bases, market participants, and local utilities in each jurisdiction approach the deployment of microgrids.28

As outlined in Figure 6, microgrids can be justified across a wide variety of use cases based on a specific set of major drivers. Behind these use cases, the ownership and control of the component technologies range along a continuum, resulting in a variety of markets potential business models. Just as

OVERVIEW OF SEPA’S 51ST STATE INITIATIVE AND EPRI’S INTEGRATED GRID PROJECT

Source: SEPA, EPRI, 2016

27 See, e.g., http://www.sepa51.org/phaseII/51stState_PhaseII_SummitReport.pdf.

28 For more details on potential market transformation outcomes and options, see Sterling, J., Stearn, C., Kaufmann, K., & van Zalk, J. (2016). Blueprints for Electricity Market Reform: Building a Structure for Collaborative Stakeholder Discussions. Smart Electric Power Alliance. http://51st.report.

SEPA launched the 51st State Initiative (sepa51.org) in 2014 to provide a collaborative dialogue across the power sector about the future of the electric industry. This future

will be shaped by technologies that allow for new ways to deliver electricity, as well as changing customer priorities like low-carbon energy and resiliency. The traditional utility business model based on rate recovery of large capital investments is being challenged. The 51st State is intended to serve as a conceptual alternative to the contentious debates surrounding market and rate reforms occurring in many jurisdictions. It has created an ongoing, safe platform for experts and industry leaders to present, sound out, and provide feedback on direction and innovation to support utility sector evolution.

EPRI’s Integrated Grid initiative, launched in 2012, is exploring advances in technology, grid planning and operations processes,

and DER cost-benefit methodologies to help facilitate the smooth transition to a future grid paradigm. The underlying vision of the Integrated Grid seeks to fully leverage the value of high penetrations of DER and customer optionality while maintaining established standards of electric quality and reliability. As part of the multi-year research activity, EPRI is pursuing a range of R&D projects across DER technology categories, including one for microgrids, that are intended to inform the state-of-the-art.

TheIntegrated

Grid

18 SEPA & EPRI | WHITE PAPER

MICROGRIDS

no two markets or utilities are alike, microgrids will continue to proliferate based on the unique served loads, targeted drivers, and deployed technologies. The ability of multiple ownership models to both exist and persist will be important.

The state-level and federal rules that govern electric power generation and distribution across the United States are complex, and the specific terms of these and future regulations will be key to determining the role and business case for microgrids. Better understanding, perhaps enabled through experimentation, will be needed to further define microgrid value, based on overlapping technology performance, regulatory, and market variables. The degree to which upfront DER costs and price signals can be reasonably communicated over the long term will be central to improving the economic viability of microgrids. Ultimately, interoperability between more traditional utility assets and DERs—whether they are owned and controlled by utilities, customers, third parties, or even independent system operators—will shape how all parties value and compensate (or are compensated for) reliability.

Microgrids can serve as a robust tool to enhance the reliability and resiliency of defined areas within the electrical distribution system by effectively aggregating DERs and other resources to provide energy, capacity, and ancillary services. The extent of their increased deployment and use will depend not only on technology and research advancements, as outlined above, but also market and regulatory reforms that seek to adequately capture their value proposition. The latter include—but are not limited to—reforms pursued as part of New York’s Reforming the Energy Vision (REV) proceedings and California’s distribution resource plan (DRP) activities. These deliberations have the potential to empower microgrids as a tool for evolving the grid ecosystem toward greater decarbonization and sustainability.

FIGURE 6. MICROGRID MATRIX: BUSINESS MODELS, IMPLEMENTATION TYPES, AND APPLICATIONS

Source: SEPA, EPRI, 2016

3RD PARTY MODEL

UNBUNDLED MODEL

INTEGRATED UTILITY MODEL

RESILIENCY/ RELIABILITY

ENERGY/ CAPACITY/

A.S.

DER INTEGRATION/ AGGREGATION

EMISSIONS REDUCTION

ENERGY IN-DEPENDENCE

INVESTMENT ALTERNATIVE/

DEFERRAL

MILITARY n n n nCOMMERCIAL/INDUSTRIAL n n n

CAMPUS-STYLE n n n nPUBLIC/INSTITUTIONAL n n nCOMMUNITY/UTILITY n nREMOTE/ RURAL n n n n

MICROGRIDS 19

Appendix: Recommended ReadingPowering Microgrids for the 21st-Century Electrical System. National Electrical Manufacturers Association, NEMA MGRD 1-2016, Roslyn, VA: 2016.

51st State: Blueprints for Electricity Market Reform. SEPA, Washington, D.C.: 2016.

Business Models for Distributed Power Generation: The Case of Microgrids. University of California San Diego, San Diego, CA: 2016.

Program on Technology Innovation: Microgrid Implementations: Literature Review. EPRI, Palo Alto, CA: 2016. 3002007384.

Beyond the Meter: Distributed Energy Resources Capabilities Guide. SEPA, Washington, D.C.: 2016.

Beyond the Meter: Addressing the Locational Valuation Challenge for Distributed Energy Resources. SEPA and Nexant, Washington, D.C.: 2016.

Beyond the Meter: The Potential for a New Customer-Grid Dynamic. SEPA, Washington, D.C.: 2016.

November Member Brief: Virtual Power Plants: Buzzword or Breakthrough?. SEPA, Washington, D.C.: 2016.

The Integrated Grid: A Benefit-Cost Framework. EPRI, Palo Alto, CA: 2015. 3002004878.

Microgrids for Critical Facility Resiliency in New York State, New York State Energy Research and Development Authority (NYSERDA), Report Number 14-36, Albany, NY: December 2014.

The Advanced Microgrid Integration and Interoperability. Sandia National Laboratories, SAND2014-1535, Albuquerque, NM: 2014.

Microgrids: A Regulatory Perspective. California Public Utilities Commission, Policy & Planning Division, Sacramento, CA: April 2014.

Microgrids—Benefits, Models, Barriers and Suggested Policy Initiatives for the Commonwealth of Massachusetts, Massachusetts Clean Energy Center (prepared by KEMA), Burlington, MA: February 2014.

The Integrated Grid: Realizing the Full Value of Central and Distributed Energy Resources. EPRI, Palo Alto, CA: 2014. 3002002733.

Distributed Renewable Energy Generation Impacts on Microgrid Operation and Reliability. EPRI, Palo Alto, CA: 2002. 1004045

White Paper on Protection Issues of The MicroGrid Concept, Consortium for Electric Reliability Technology Solutions (CERTS), Prepared for the Department of Energy, March 2002.

EPRI Smart Grid Resource Center, Use Case Repository (NEDO, NIST, and ORNL microgrid use cases): http://smartgrid.epri.com/Repository/Repository.aspx.

3420 HILLVIEW AVENUE PALO ALTO, CA 94304

650-855-2000© Electric Power Research Institute, Inc.

2001-2016 All Rights Reserved.

1220 19TH STREET NW, SUITE 800, WASHINGTON, DC 20036-2405

202-857-0898©2016 Smart Electric Power Alliance.

All Rights Reserved.