One of the great challenges facing the growing microgrid market is the wide range of applications, climates, cultures, resources, and costs associated with a highly dispersed market. The HOMER® microgrid optimization software is a useful tool for determining which particular technology combination and configuration may be economical and effective for microgrid projects of all types and sizes. We suggest there are currently four microgrid categories into which most systems can fit and which form the structure of our training curricula around microgrid design and certification using the HOMER software. This microgrid classification system also defines submarkets and therefore suggests how companies may position themselves within a diverse market. Each of these submarkets will have a distinct set of quality standards.
HOMER Energy has been involved with optimizing microgrids for several decades. Its principals first created HOMER at NREL over twenty years ago. In 2009, they formed HOMER Energy and became the exclusive licensee to support and enhance it. HOMER Energy has continued to develop HOMER and has provided training, analytical, and consulting services to microgrid developers, utilities, municipalities, system integrators, governmental and non-governmental organizations, and component distributors and manufacturers. This complete focus on microgrids for many varied applications has led us to understand that there is a need to define distinct categories within the broad spectrum of microgrid systems.
Peter Lilienthal first suggested a microgrid classification system in April 2013 in his post How to Classify Microgrids: Setting the Stage for a Distributed Generation Energy Future. He suggested four basic microgrid types, which can be displayed as a 2×2 matrix:
While the grid-connected versus remote division is fundamental, the ways in which these two groups further divide is only roughly correlated with small/large, and we now suggest that microgrids can be more accurately subdivided by the following tree: 
The scale issue is still very important because there are many qualified power project developers prepared to develop and finance larger projects, but the greatest potential market for microgrids are projects that are too small to attract those companies. Innovative development and financing models are required for microgrids to achieve their full potential. These innovative approaches will require more involvement by local stakeholders and more standardized design choices and financing processes.
The two grid-connected types are distinguished by the reliability of the existing grid service. Many locations have grid service that is sufficiently reliable for most applications, but not adequate for critical emergency and public safety services. By contrast, most developing countries have growing middle classes and commercial enterprises that operate backup generators on a regular basis. Improving the reliability of their electric service is widely recognized as a key enabler of their continued economic development.
Remote microgrids have no grid connection. The most important distinction for remote microgrids is whether there is an existing utility company or not. Each of the four microgrid types has its own challenges. There are technical challenges, such as integration issues and system design and equipment choices. There are also development, policy, and financing challenges related to transaction costs, support infrastructure, capital requirements, and tariffs. The different types of questions these challenges raise lead to both different quality standards and different modeling approaches within HOMER.
Grid-Connected Microgrids
Grid-connected microgrids provide backup power to a national or regional grid. There are currently two primary markets in grid-connected systems:
Type 1 – Microgrids connected to reasonably reliable utility grids: These systems either need extremely high reliability, have a consistent thermal load for combined heat and power (CHP), or particular value propositions that make microgrids attractive. Critical infrastructure includes telecommunications facilities, data centers, military bases, medical centers, police stations, and firehouses but also many less obvious facilities, such as gas (petrol) stations. The series of severe storms that have hit the northeastern United States in recent years has focused on the need and potential for these microgrids to be able to provide reliable power for extended (weeks at a time) periods. Some of these facilities cannot tolerate even very momentary disruptions in power quality. The value of reliability is so high in these applications that the value of energy savings may not be a deciding factor whether to deploy a microgrid.
Applications with a consistent thermal load have the opportunity to deploy a CHP system. CHP systems substantially improve the financial return of a microgrid, making them potentially attractive anywhere with access to low cost fuel, such as pipeline gas. Campuses are particularly attractive because they can be relatively large loads with a single owner. Unlike many commercial facilities, university campuses often have long financial horizons for capital projects.
This is a relatively new category of microgrid. Utilities, regulators, and other policy makers are still working out their relationship to the central grid. They have the potential to be a major supplier of ancillary services and demand response to the grid if the appropriate regulatory, financial, communication, and control structures can be agreed upon.
Although development interest in this category is mostly limited to larger facilities, storage cost reductions resulting from the advent of packaged systems and electric vehicles may make this a residential option at some point in the near future. These systems can overcome the constraints that utilities are starting to consider for additional distributed renewables.
Type 2 – Microgrids connected to unreliable grids: The electric service in most developing countries is so unreliable that commerce is all but impossible without backup power, and every middle and upper-class family that can afford it already owns their own backup power systems. These grids may be unreliable because of random outages, rolling blackouts, or regularly scheduled part-time operation. Unlike Type 1 Microgrids, these backup microgrids are attractive because they are used often enough that simple diesel backup generators have unsustainable fuel costs and maintenance, noise, and emission issues for their owners. This contrasts them with the Type 1 Microgrids that have different value propositions. These microgrids also do not need the same level of sophistication for their switchgear, controls, and communications. For some of the smaller microgrids in this category, solar plus storage may provide sufficient backup reliability without a diesel generator.
Isolated Microgrids
These microgrids are not economical to connect to a larger grid either because of the cost of submarine cables or because they are too small and far from a central grid. There are two main groups that are distinguished by the extent of the existing infrastructure.
Type 3 – Island Utilities: These larger isolated systems are managed by a real utility company with billing and engineering departments and distribution systems that they maintain. All but the smallest of them also have switchgear and synchronizing gear that allows them to run multiple generators in parallel. They do not need to be actual geographic islands, as there are about 200 isolated villages in Alaska, 300 in northern Canada, and 3000 in Russia that cannot be connected to a central grid. There is no obvious upper size limit, but we limit our focus to systems small enough that their primary conventional generators are liquid-fueled. This can include systems with peak loads in excess of 100 MWs. Island utilities are currently the locations with the greatest integration issues, as they have strong economic incentives for aggressively adding renewables. The island nation of Tokelau, which has a single integrated generation system on each of its relatively small islands, is nearly 100% renewables plus storage, the first-ever island nation to achieve this goal.
Type 4 – Village Power: These are the smallest systems, which serve a small village or group of houses. Similar systems can also serve small ecotourism facilities and remote research sites. These systems have no more than one backup diesel generator, combined with renewable generation and storage. Because the grid is often unreliable in many of these areas, village power microgrids may be a better choice even when grid extension might appear to be economical. Without an existing utility, the primary challenge for these systems are developmental, because they are not at a scale to attract professional power project developers, but they still need financial and support services, including metering, billing, and maintenance.
Remote telecomm systems and cell towers are a special case. They have the locations and typical size of Type 4, Village Power systems, but the high reliability requirement of Type 1 systems. Part-time diesels are another special case. Many remote communities have a diesel generator that is only scheduled to run for a few hours per day. These systems straddle the two remote categories that we describe above because they already have some infrastructure, but are very different from typical utility plants that have multiple generators and on-site engineers. Part-time diesels can range from utility-owned systems with a conventional distribution network to very informal arrangements with extension cords strung between trees. The conversion to a hybrid renewable microgrid can dramatically improve power quality, but there are important issues around tariff structures and energy efficiency that need to be addressed in order to develop sustainable systems.
As the microgrid market further develops, the expertise and processes around these various microgrid types will likely create a natural division of markets and operators. For now, we suggest that these four major groupings form the basis for capacity building and quality standards. HOMER Energy is developing a software certification program for 2015, based on the HOMER Pro software, that will use these four types as areas of expertise.