Microgrid Controller

The controller executes intentional islanding by detecting upstream grid disturbances via IEEE 1547 voltage and frequency ride-through curves. It opens the static transfer switch at the point of common coupling within a single cycle to isolate the microgrid. During seamless reconnection, the controller synchronizes voltage magnitude, frequency, and phase angle to within specified limits before issuing a close command, preventing transformer inrush and power quality events.

The controller implements a hierarchical control architecture. Primary control uses fast droop control curves on grid-forming inverters to share real and reactive power proportionally without communication. Secondary control is a centralized PI loop that restores frequency to nominal and corrects voltage deviations caused by droop action. Tertiary control optimizes economic dispatch across assets based on marginal cost curves and market signals.

The controller performs real-time state of charge management for battery energy storage systems. It enforces depth-of-discharge limits to prevent accelerated degradation and manages charge/discharge cycling to maximize cycle life. During islanded operation, it reserves a calculated energy buffer to sustain critical loads through the forecasted outage duration. The algorithm balances grid-forming energy reserves against energy arbitrage revenue opportunities.

When generation capacity is insufficient to meet demand, the controller executes load shedding with millisecond latency. It uses a prioritized tripping table where loads are categorized by criticality. Adaptive protection logic recalculates the available generation headroom in real-time and dynamically adjusts the shedding threshold. Under-frequency load shedding relays are coordinated with the controller's centralized scheme to arrest a frequency nadir before it reaches inverter trip settings.

The controller possesses black start capability to re-energize a completely de-energized microgrid. It follows a pre-configured energization sequence: first establishing a voltage reference via a grid-forming inverter or a diesel generator with black start capability, then incrementally adding feeder segments and soft-starting loads to avoid inrush current collapse. The controller manages the cold load pickup challenge by staggering load restoration based on diversity factors.

For economic optimization, the controller employs model predictive control that uses forecasts of renewable generation and load to solve a constrained optimization over a receding horizon. The algorithm computes optimal setpoints for dispatchable assets while respecting battery state-of-charge constraints, generator ramp rates, and network thermal limits. This minimizes operational cost and maximizes renewable self-consumption across the prediction window.

The controller must distinguish between grid-forming inverters, which establish voltage and frequency references independently, and grid-following inverters, which act as current sources synchronized to an existing waveform. During a transition to islanded mode, the controller must instantaneously switch the operational mode of designated inverters to prevent a collapse of the voltage reference.

• Grid-Forming: Creates the microgrid's electrical backbone without external reference.
• Grid-Following: Requires a stable voltage source to inject power.
• Transition Logic: The controller manages the seamless handoff between these modes during islanding.

A decentralized load-sharing algorithm that acts as the primary layer of hierarchical control. The controller implements P-f droop (frequency vs. active power) and Q-V droop (voltage vs. reactive power) to allow multiple parallel generators to share load proportionally without requiring high-speed communication links.

• P-f Droop: A 5% frequency drop corresponds to 100% active power output.
• Q-V Droop: A 4% voltage drop corresponds to 100% reactive power output.
• Virtual Impedance: The controller can emulate output impedance to improve power sharing accuracy.

The controller relies on Generic Object Oriented Substation Events (GOOSE) for high-speed, peer-to-peer communication with intelligent electronic devices. This protocol bypasses the TCP/IP stack to publish binary status changes and analog measurements directly onto the Ethernet layer, achieving transfer times under 3 milliseconds for critical protection and control signals.

• Publisher-Subscriber Model: One device publishes a message; multiple devices receive it simultaneously.
• Retransmission Logic: Messages repeat at increasing intervals to ensure delivery.
• VLAN Tagging: Prioritizes GOOSE traffic over less critical network data.

At the tertiary control level, the microgrid controller uses Model Predictive Control to optimize dispatch over a receding time horizon. The controller solves a constrained optimization problem using a dynamic model of the battery state of charge, renewable forecasts, and load predictions to minimize fuel consumption or electricity costs while respecting voltage and thermal limits.

• Prediction Horizon: Typically 15 minutes to 24 hours.
• Control Horizon: The first step of the optimized sequence is executed.
• Constraints: Battery depth of discharge, generator ramp rates, and line ampacity.

The automated process of resynchronizing an islanded microgrid with the main utility grid. The controller monitors the voltage magnitude, frequency, and phase angle on both sides of the point of common coupling (PCC). It issues corrective setpoints to the grid-forming assets to minimize the vector difference before issuing a close command to the static transfer switch.

• Sync Check Relay: Verifies the voltage difference is within 5%, frequency within 0.1 Hz, and phase angle within 5 degrees.
• Phase-Locked Loop (PLL): Tracks the utility voltage angle for synchronization.
• Bumpless Transfer: Ensures no power surge or transient during reconnection.

The ability of the microgrid controller to orchestrate a full system restoration from a completely de-energized state without external grid support. The controller must sequence the startup of a grid-forming asset (typically a battery or diesel generator), establish a stable voltage island, and then incrementally add grid-following assets and load blocks to avoid frequency excursions.

• Soft Start: Loads are added in small increments to prevent voltage collapse.
• Inrush Management: Transformer energization is staggered to avoid nuisance tripping.
• Dead Bus Detection: The controller must first verify the bus is de-energized before initiating the sequence.