In our previous article, we explored the three core values that make microgrids a key building block of modern power systems: ultra-reliable power supply, true energy independence, and an efficient path toward low-carbon transition.
After reading that, many global customers asked the same follow-up question: how does a microgrid actually work in real life? How does each kilowatt-hour move through the system from generation to storage, from storage to consumption, and through dispatch and control? What exactly are the devices doing in grid-tied mode and in off-grid mode?
In this article, we move beyond broad concepts and go deeper into device-level coordination, operating scenarios, and real-time control logic. We will break down how microgrids work at the system level, explain the technical principles behind both grid-tied and off-grid microgrids, and show how these systems adapt to different power environments around the world.
First, Understand the Smallest Operational Loop of a Microgrid
Every microgrid operating mode is built on the same four core functional modules. Understanding their roles is the foundation for understanding how a microgrid works.
1. Power Generation Units
The generation side is the energy supply source of the microgrid. It usually includes two categories: uncontrollable renewable generation and controllable backup generation.
Uncontrollable renewable sources mainly include solar PV arrays and wind turbines. These are the primary sources of clean electricity in most microgrids. Their output changes in real time based on sunlight, wind speed, and other environmental conditions, which means they cannot be precisely adjusted by command.
Controllable sources mainly include diesel generators, gas generators, and hydrogen fuel cells. These units are typically used as backup power in extreme conditions. Unlike renewables, they can be started, stopped, and ramped up as needed, with a typical startup response ranging from tens of seconds to a few minutes.
2. Energy Storage Units
The storage side is the stabilizing core of the microgrid. It usually consists of lithium battery banks and a bidirectional Power Conversion System (PCS).
The battery functions like an energy reservoir. It stores excess electricity when generation is higher than demand and discharges when there is a supply gap.
The bidirectional PCS is one of the most critical pieces of hardware in the entire system. It is the only unit that can manage two-way conversion between AC and DC, precisely control the direction of power flow, and actively regulate voltage and frequency when required. It is also the key device that enables smooth switching between grid-tied and off-grid operation. In fast-response applications, its control speed can reach the millisecond level.
3. Load Units
The load side represents the actual electricity users within the microgrid. It is also the anchor point for real-time power balancing.
In practice, loads are often divided into three priority levels:
- Tier 1 critical loads: loads that cannot tolerate outages, such as hospital ICUs, data centers, and precision manufacturing lines.
- Tier 2 flexible loads: loads such as pumps or HVAC systems that can be shifted or adjusted for short periods to support energy optimization.
- Tier 3 interruptible loads: non-critical loads such as ordinary office equipment that may be temporarily shed during extreme conditions to protect essential power supply.
Because load demand changes continuously with production and daily use, the microgrid must respond dynamically.
4. Control and Dispatch Units
The control layer is the brain of the microgrid. It usually includes three levels.
At the field level, local controllers are responsible for real-time data acquisition, command execution, and fast response to power fluctuations.
At the supervisory level, the Energy Management System (EMS) handles global optimization, dispatch strategy, peak-valley arbitrage logic, and operational planning.
The most critical physical interface is the Point of Common Coupling (PCC), which is the only connection point between the microgrid and the utility grid. The PCC is equipped with intelligent breakers and bidirectional metering devices, making it the physical gateway for switching between grid-connected and islanded operation.
In essence, the working logic of a microgrid is to maintain real-time balance between generation, storage, and loads under both grid-tied and off-grid conditions, while keeping system voltage and frequency within acceptable operating limits.
1. Grid-Tied Microgrid: How It Works in Full Operation
Grid-tied operation is the mainstream mode in areas with mature utility infrastructure. In this mode, the main grid is available and provides a stable voltage and frequency reference, such as 220V/50Hz or 110V/60Hz.
All microgrid devices synchronize to that reference. In this state, the PCS usually operates in PQ mode (constant power mode), meaning it controls active and reactive power output but does not establish system voltage or frequency.
Let us take a 1 MW commercial and industrial microgrid as an example, consisting of 800 kW PV, 500 kWh / 250 kW battery storage, and factory loads.
Scenario 1: Steady-State Operation
At 10:00 a.m., the PV system is generating 600 kW, the factory load is also 600 kW, the battery SOC is 50%, and the utility grid is stable.
The local controller rapidly collects operating data and confirms that PV generation matches the load. The EMS then issues a self-consumption priority command. The PV inverter outputs 600 kW, and all generated electricity is consumed directly by the factory.
In this situation, the PCS remains on standby. No power is imported from or exported to the grid through the PCC. The bidirectional meter continues recording electricity data for energy accounting, carbon tracking, and utility settlement.
Scenario 2: Power Surplus
At noon, PV output rises to 800 kW, while the factory load drops to 300 kW during a break. This creates a 500 kW surplus.
The controller detects the surplus and reports it to the EMS. The EMS follows a “store first, export second” strategy. The PCS immediately begins charging the battery at its rated 250 kW. The remaining 250 kW is exported to the utility grid through the PCC.
The bidirectional meter records the exported electricity, allowing the customer to earn revenue where local policy supports energy sales. Once the battery SOC reaches its upper protection threshold, the EMS stops charging and all remaining surplus power is exported.
Scenario 3: Power Deficit
At 6:00 p.m., PV output drops to zero, while the factory evening load rises to 700 kW. A 700 kW power gap appears, and local electricity prices are at peak levels.
The controller detects the shortfall and reports it immediately. Based on cost optimization logic, the EMS instructs the battery system to discharge first during peak-tariff hours. The PCS switches to discharge mode and delivers 250 kW. The remaining 450 kW is imported from the utility grid.
This allows the microgrid to reduce expensive peak-time electricity purchases while maintaining uninterrupted power supply and stable voltage.
Scenario 4: Grid Fault and Seamless Islanding
Now consider an abnormal grid event such as voltage sag, frequency deviation, or total blackout.
Once the protection devices at the PCC detect that grid conditions are outside the acceptable range, they send an islanding signal. The breaker at the PCC opens, physically separating the microgrid from the utility grid.
At the same time, the PCS shifts from PQ mode to VF mode (constant voltage and frequency mode), taking over as the reference source for the local system. The EMS then executes load-priority dispatch, ensuring that critical loads remain energized while lower-priority interruptible loads may be curtailed if necessary.
This is how a well-designed hybrid microgrid achieves seamless transition and keeps essential loads running without noticeable interruption.
2. Off-Grid Microgrid: The Core Logic of a Fully Autonomous Power System
Off-grid operation is what truly distinguishes a microgrid from conventional distributed generation or backup generators.
In off-grid mode, the microgrid is completely disconnected from the utility grid. There is no external voltage or frequency reference. The system must create and maintain its own stable electrical standard while balancing generation, storage, and load in real time.
This is the defining technical difference between grid-tied and off-grid microgrids.
Master-Slave Control Architecture
The most mature off-grid microgrid designs use a master-slave control structure.
Because there is no utility grid to define voltage and frequency, one device must take the lead. In most practical systems, the bidirectional storage PCS acts as the master unit and operates continuously in VF mode, functioning like a “virtual utility grid” inside the microgrid.
Other devices, such as PV inverters, wind converters, and diesel generators, act as slave units. They operate in PQ mode, follow the voltage and frequency established by the PCS, and adjust output according to dispatch instructions.
Let us take a 1.2 MW off-grid microgrid as an example, consisting of 1 MW PV, 1 MWh / 500 kW storage, and an 800 kW diesel generator, serving a remote African village and a small processing plant where no utility grid is available.
Scenario 1: Stable All-Renewable Operation
At 10:00 a.m., the PV system is generating 500 kW, total village and factory load is 400 kW, and battery SOC is 60%.
The storage PCS, acting as the master device, continuously provides a stable 230V/50Hz reference. All other equipment follows that standard.
The EMS prioritizes clean energy, so the PV inverter delivers 400 kW directly to the loads. The remaining 100 kW is used to charge the battery through the PCS.
In this state, the microgrid achieves balanced operation across generation, storage, and load with no instability.
Scenario 2: Sudden PV Drop
Suppose PV output was originally 800 kW while the load was 500 kW, leaving 300 kW available for battery charging. Then a fast-moving cloud causes PV generation to fall sharply to 200 kW within a very short time.
This creates a 300 kW shortfall.
The storage PCS detects the disturbance almost instantly and, without waiting for EMS-level scheduling, switches to discharge mode and injects 300 kW to cover the gap. At the same time, the local controller reports the event to the EMS.
Throughout this process, the PCS continues to hold voltage and frequency stable, so the loads do not experience flicker or interruption.
If the low-PV condition continues and battery SOC falls to a low threshold, the EMS automatically starts the diesel generator. Once online, the generator follows the PCS reference and supplies both the load and battery charging demand.
Scenario 3: Sudden Load Increase
Now assume PV output is 400 kW, the stable load is 300 kW, and the remaining 100 kW is charging the battery. Suddenly, a 200 kW industrial motor starts and total load rises to 500 kW.
This creates a 100 kW power gap.
The PCS detects the frequency dip and responds within milliseconds by discharging 100 kW to stabilize the system. The EMS then evaluates whether the new load level is temporary or sustained. If it continues, the system may increase renewable output where available and prepare the diesel generator as backup.
Once the load returns to normal, the microgrid resumes its economically optimized operating state.
Scenario 4: Black Start and Extreme Backup Protection
Consider an extreme case: three consecutive rainy days reduce PV generation to nearly zero, battery SOC falls to the deep protection threshold, and the diesel generator has also stopped because of a fuel issue. The entire microgrid is shut down.
In this case, the system must perform a black start.
With black-start capability enabled, the storage system uses its reserved battery energy to start the PCS without any external power source. The PCS first establishes the voltage and frequency backbone of the microgrid.
The EMS then energizes only the highest-priority critical loads, such as clinics or communication stations. Once the local power backbone is restored, the system can supply starting power to the diesel generator, bring it online, and then recharge the battery while supplying the remaining loads.
When solar resources recover, the EMS automatically shifts the system back to the normal “PV first, diesel backup” operating strategy, minimizing fuel consumption and operating cost.
3. Hybrid Microgrid: Seamless Switching Between Grid-Tied and Off-Grid Modes
Today, the vast majority of commercial microgrid projects are built as hybrid microgrids that can switch seamlessly between grid-tied and off-grid operation.
This is where the true competitiveness of microgrids lies.
A hybrid microgrid is not simply a combination of two separate modes. Its key technology is synchronization and seamless transition control.
Under normal conditions, the system runs in grid-tied mode to maximize solar self-consumption, reduce electricity costs, and perform peak-valley arbitrage where applicable.
If the utility grid fails, the microgrid disconnects within milliseconds and immediately enters islanded mode to keep critical loads energized.
Once the utility grid recovers, the microgrid does not reconnect abruptly. Instead, it first adjusts its own voltage, frequency, and phase so they match the utility grid precisely. Only then does the PCC breaker close, allowing smooth reconnection without current shock or supply interruption.
This “grid-connected for economics, islanded for resilience” architecture gives users the best of both worlds: lower operating costs and higher power reliability.
That is why hybrid microgrids are increasingly used across a wide range of applications worldwide, including hospitals, data centers, precision manufacturing plants, remote communities, island regions, and disaster-prone coastal areas.
Conclusion
As we can see, the difference between grid-tied and off-grid microgrids is not simply whether the system is connected to the utility grid.
It is a complete technical framework that combines hardware control, software dispatch, and real-time response under different operating conditions.
Grid-tied mode uses the utility grid as a stable reference to maximize renewable consumption, optimize electricity cost, and improve financial returns.
Off-grid mode relies on the voltage and frequency support capability of energy storage to achieve full energy autonomy, making reliable electrification possible in remote or weak-grid areas.
Hybrid mode combines both economic performance and resilience, which is why it has become the mainstream architecture for modern microgrid deployment.
This is the core advantage of microgrids compared with conventional distributed solar or standalone backup generators: a microgrid is not just a power source. It is a complete, intelligent, and flexible power system that can adapt to different scenarios and deliver reliable, autonomous, and low-carbon energy.
For global customers, however, one key question always remains: how much money can a microgrid really save? What does its full life-cycle cost structure look like? And how long is the payback period?
In our next article, we will break down the economics of microgrids from three dimensions: initial investment, operating cost, and value-added returns. We will also show how microgrids can turn electricity from a cost center into a value-generating asset.