Microgrids

A microgrid is a controllable local power system. It usually includes electrical loads, distributed generation, energy storage, switchgear, protective devices, meters, communications, and a microgrid controller. The system may normally operate in parallel with the utility grid, but it is designed so it can separate from the grid and continue serving selected loads when needed.

Although wording varies by organization, most engineering definitions describe a microgrid as a group of interconnected loads and distributed energy resources within a defined electrical boundary that can act as one controllable entity. The most important idea is not simply “local power.” A facility may have rooftop solar, a standby generator, or a battery system without being a complete microgrid.

A microgrid works by continuously balancing local generation, storage, and load demand within a defined electrical area. When the utility grid is available, the microgrid may import power, export power, charge batteries, reduce demand peaks, or support local loads with renewable generation. When the grid becomes unavailable or the system is intentionally islanded, the microgrid disconnects and uses local resources to keep priority loads energized.

Grid-connected mode

In grid-connected mode, the utility grid often acts as the voltage and frequency reference. The microgrid can serve local loads, charge batteries, offset utility purchases, or export power if the interconnection agreement allows it. Protection and metering at the point of common coupling are especially important because the microgrid is operating in parallel with the larger electric system.

Microgrid islanding

Islanding is the process of separating the microgrid from the utility grid and operating it as a local electrical island. This can happen during an outage, a planned resilience test, or an intentional operating event. During islanded operation, the microgrid must maintain voltage, frequency, load balance, and protection without relying on the utility system.

Reconnection and synchronization

Reconnection is not just closing a breaker. The microgrid must be synchronized with the utility grid so voltage magnitude, frequency, phase angle, and protection conditions are acceptable. Poor reconnection logic can create damaging transients or unsafe operating conditions.

Black start capability

Black start capability means the microgrid can restart selected loads without first relying on the utility grid. This may require a generator, grid-forming inverter, battery system, or staged start sequence that energizes controls, auxiliaries, switchgear, and priority loads in a deliberate order.

This simplified power balance is the operating logic behind a microgrid. During islanded operation, \(P_{\text{grid import}}\) becomes zero, so local generation, battery discharge, and load shedding must keep the system balanced in real time. The expression shows real power balance; actual microgrids also require reactive power support, voltage regulation, harmonic control, ramp-rate management, and stable transient response.

Microgrids are built from familiar power system equipment, but the equipment must be coordinated as a local operating system. The controller, protection scheme, communications, and switching logic are what allow the microgrid to behave as more than a collection of independent devices.

Many of these components are also used in conventional distribution systems. The difference is that a microgrid design must coordinate them around islanding, critical-load service, energy management, and reconnection instead of treating each device as a standalone asset.

The microgrid controller is the operational brain of the system. It monitors loads, DER output, battery state of charge, utility conditions, switchgear status, meters, and protective devices. It then issues commands for dispatch, load shedding, voltage control, frequency control, islanding, and reconnection.

Grid-forming vs grid-following resources

Modern microgrids often depend on inverter-based resources, so the difference between grid-forming and grid-following behavior matters. A grid-following inverter typically needs an existing voltage and frequency reference. A grid-forming inverter can help establish that reference during islanded operation. If a microgrid is expected to run on solar and batteries after separation from the utility, the design must identify what source forms the islanded grid.

Protection in grid-connected vs islanded mode

Protection can become more difficult after islanding because the available fault current may drop, especially when inverter-based resources dominate the microgrid. A relay setting that works while connected to the utility may not detect or coordinate properly during islanded operation unless the protection study considers both modes.

Utility coordination

Utility coordination is not a paperwork detail. The microgrid’s point-of-common-coupling protection, export behavior, reclosing logic, anti-islanding requirements, and reconnection sequence must align with the serving utility’s interconnection requirements.

Microgrids can be grouped by ownership, operating mode, location, and electrical architecture. The label matters because a remote community microgrid, hospital microgrid, campus microgrid, and industrial microgrid may all use similar components but have very different design priorities.

Most utility-connected microgrids are AC systems because buildings, switchgear, utility feeders, and many loads are AC. DC and hybrid architectures are increasingly important where solar PV, batteries, electric vehicle charging, data centers, or power electronics dominate the load and resource mix.

Microgrids are often discussed as resilience tools, but their value depends on what the owner needs the system to do. A microgrid may be justified by outage resilience, renewable integration, energy cost management, power quality, fuel security, or operational control. Each benefit comes with a design tradeoff.

What affects microgrid cost?

Microgrid cost is controlled by the resource mix, storage duration, critical-load size, switchgear, protection upgrades, controls, communications, interconnection studies, civil and electrical construction, commissioning tests, cybersecurity requirements, and long-term operations and maintenance. Cost estimates should be based on the required operating mode and resilience target, not only on total kilowatts installed.

Microgrids are often confused with related power system concepts. The distinction usually comes down to whether the system has a defined electrical boundary, coordinated controls, islanding capability, and a plan for serving specific loads as a local grid.

Microgrid sizing starts with the loads that must remain energized, then works backward to the resources needed to serve them for the required duration. Oversizing every source is rarely the best answer; a strong design separates critical loads, uses storage strategically, and defines what the system is allowed to shed during islanded operation.

Consider a hospital campus with utility service, rooftop solar, a battery energy storage system, emergency generators, critical medical loads, life-safety systems, HVAC loads, and administrative loads. The microgrid value comes from how those pieces are sequenced during an outage.

Normal operation

During normal operation, the hospital imports utility power while solar offsets daytime load and the battery maintains a reserve state of charge. The controller monitors the utility meter, battery state of charge, generator status, and critical-load demand.

Outage and islanded operation

When the utility outage is detected, the point-of-common-coupling breaker opens and the microgrid isolates. The battery or another grid-forming source stabilizes the transition, critical loads remain energized, emergency generators start, and noncritical loads are shed or delayed if the islanded resource margin is limited.

Restoration

When utility service returns, the controller verifies acceptable voltage, frequency, and phase conditions before reconnection. Loads can then be restored in stages, the battery reserve can be rebuilt, and the system returns to normal grid-connected operation.

Microgrid diagrams often make the system look simple: utility grid, switch, solar, battery, generator, and loads. Real projects are harder because controls, protection, utility requirements, equipment ratings, maintenance access, communications, cybersecurity, and operating procedures all affect whether the system performs during an outage.

Engineers also need to consider how the system will be maintained. Batteries age, generators require fuel and exercising, protective settings change when equipment is modified, and operators need clear procedures. A microgrid is an operating asset, not just a construction package.

The simplified microgrid concept breaks down when the system cannot maintain power balance, cannot protect itself under islanded conditions, or cannot transition between operating modes without unacceptable voltage, frequency, or safety issues.

• Insufficient local capacity: The DER and storage mix cannot support the selected islanded loads for the required duration.
• Weak protection assumptions: Fault current is lower or different in island mode, causing protective devices to miscoordinate or fail to trip as expected.
• Unclear load priority: Too many noncritical loads remain connected, draining storage or overloading generation during islanded operation.
• Poor transition control: The system experiences voltage or frequency excursions during islanding, black start, or reconnection.
• Untested operating logic: The controller sequence looks correct in design documents but has not been validated under realistic conditions.
• Unplanned O&M burden: Batteries, generators, relays, communications, sensors, and control systems require periodic testing and maintenance.

Many microgrid mistakes come from treating the concept as an equipment purchase rather than a coordinated operating system. A battery container, solar array, or standby generator can be valuable, but the microgrid performance depends on how those assets are controlled and protected together.