In short. A grid-connected microgrid runs with the utility present most of the time, using local generators, batteries and solar to reduce cost, shave peaks and raise self-consumption. The utility provides the voltage and frequency reference, fault current and system strength. The point of common coupling must manage import, export and protection, and the design must state exactly what happens when the utility disappears, because the grid was doing far more for the site than supplying kWh.
Normal operation with the utility present
In grid-connected mode the DISCOM network is the reference. It sets voltage and frequency, absorbs mismatches between local generation and load, and supplies large fault currents that let protection operate quickly. Local sources work as grid-following resources: the battery charges and discharges to setpoints, PV injects what it can, and generators, if permitted to parallel, run at fixed output. The microgrid controller’s job in this mode is mainly economic and contractual.
Typical strategies
- Peak shaving to keep import below contract demand, described in peak-shaving systems.
- Self-consumption, using solar on site and storing surplus where export is limited or poorly paid.
- Time-of-day shifting, charging the battery at off-peak rates and discharging during peak tariff hours.
- Resilience readiness, keeping the battery above its reserve level and generators ready so the site can island at any moment.
The point of common coupling
The PCC is the electrical and contractual boundary. It needs a breaker capable of separating the site under fault, metering that measures import and export, and protection that meets the DISCOM’s requirements, typically under and over voltage, under and over frequency, rate of change of frequency or vector shift for islanding detection, reverse power where export is not allowed, and directional overcurrent. The controller must know the PCC breaker status, power direction and utility condition at all times.
| PCC function | Purpose | Typical device |
|---|---|---|
| Isolation | Separate the site on fault or utility failure | ACB or VCB with motorised operation |
| Islanding detection | Stop local sources feeding a dead utility network | Multifunction relay with ROCOF and vector shift |
| Export control | Hold export at zero or the approved limit | Export limiter feeding the controller; reverse power relay |
| Synchronising check | Permit reconnection only when matched | Sync-check relay |
| Metering | Billing and controller feedback | DISCOM meter and a separate controller meter |
Paralleling generators with the utility
Many sites want diesel or gas generators to run in parallel with the grid, for peak shaving, closed-transition testing or no-break return. That needs synchronising switchgear, protection approved by the utility and a clear operating permission. Indian DISCOMs vary in their requirements and some restrict DG paralleling altogether. The switchgear side is covered in utility paralleling and the control side in utility and DG synchronisation.
What the grid was doing for you
The moment the utility is lost, the site changes from a load connected to a strong network into an autonomous power system. Everything the grid silently provided must now come from inside the boundary: voltage reference, frequency reference, reactive power, inertia, fault current, and the absorption of every load step. If no local source is ready to form the grid, the island collapses and the site goes dark until a generator starts through AMF logic. A grid-connected design must therefore state, for each source, what it does at the instant of utility loss and who takes over the reference. Read islandable microgrids for that transition.
Pgap = Pload − Plocal
Pgap is the utility import at the moment of loss, which local grid-forming sources must pick up at once. If a site imports 400 kW with 300 kW of PV running, losing the grid leaves a 400 kW gap (illustrative), and the PV itself may trip on the disturbance, raising the gap to 700 kW.
Planning for Indian supply conditions
Grid-connected microgrids in Telangana and Andhra Pradesh face supply conditions that a European design template may not anticipate: frequent short interruptions, voltage sags from upstream faults, scheduled maintenance outages announced at short notice, and wide voltage variation at the end of long rural feeders. Each of these is a possible islanding trigger. The controller settings must distinguish a genuine loss of supply from a brief sag that the site should ride through, otherwise the microgrid islands several times a day and wears out its switchgear. Recording a few weeks of supply quality at the PCC with a power quality analyser before setting thresholds is time well spent. The same data show whether voltage regulation by the battery inverter, through reactive power, could reduce the number of sags that trip sensitive production equipment.
Returning to the grid
When utility power returns and has been stable for a set time, the site either reconnects by synchronising its island to the grid in a closed transition, or by briefly dropping the load in an open transition. Closed transition needs synchronising equipment at the PCC and utility permission. Seamless transfer in either direction is not an inherent feature of every microgrid; it has to be designed, tested and approved.
Common mistakes
- Designing only the grid-connected economics. The site discovers on the first outage that nothing can form an island.
- Assuming solar inverters will keep running when the grid fails. They trip under anti-islanding rules and the gap doubles.
- Paralleling generators with the grid without approved protection. The DISCOM disconnects the site or imposes penalties.
- Spending the battery reserve on tariff arbitrage. The transition fails because the battery is empty.
What to send us
Send the single-line diagram from the DISCOM meter downward, the sanctioned load and contract demand, twelve months of interval data, details of PV, battery and generator assets, and any correspondence with the DISCOM on paralleling or export. Submit through the application engineering enquiry and we will outline the PCC scheme and operating modes.
Questions engineers ask
Can my microgrid export power to the DISCOM?
Only under the utility’s net-metering, gross-metering or other approved arrangement, with the required metering and protection. Many industrial connections are set for zero export, which the controller must enforce.
Will the site stay powered when the grid fails?
Only if a local source is ready to form the grid and the PCC opens fast enough. Without a grid-forming battery or a running generator, the site sees a break until the standby generator starts.
What is anti-islanding protection?
It disconnects local sources when the utility network is lost so that they do not energise lines the DISCOM believes are dead. The PCC can be designed to open first, after which the site can island deliberately.
Is a grid-connected microgrid worth it if the grid is reliable?
The economic functions such as peak shaving and solar self-consumption can still justify it. The resilience functions add cost and are worth it mainly where outages are frequent or costly.
Technical content reviewed by Srinivas Reddy, Managing Partner, Reddy Generators. Last updated 19 October 2026.