The rapid integration of renewable energy sources, primarily through power electronic converters, is fundamentally reshaping modern power grids. Among these, solar inverters play a pivotal role in converting the direct current (DC) output from photovoltaic (PV) panels into grid-compliant alternating current (AC). However, the displacement of traditional synchronous generators by these inverter-based resources (IBRs) has led to a critical reduction in system inertia, posing significant challenges to grid stability, particularly frequency stability. In response, Grid-Forming (GFM) control technology has emerged as a cornerstone for the future grid. GFM solar inverters employ advanced control algorithms to emulate the inertia and damping characteristics of synchronous machines, actively regulating voltage and frequency to support the grid. Unlike their Grid-Following (GFL) counterparts that rely on a stable grid voltage signal for synchronization, GFM inverters can establish and maintain grid voltage autonomously, behaving as voltage sources.
A particularly effective GFM strategy for power-limited sources like PV is the Matching Control. This ingenious method establishes a direct correspondence between the DC-link voltage and the virtual rotor speed of a synchronous machine. It leverages the energy stored in the DC-link capacitor to simulate rotational inertia, thereby providing frequency support without necessitating constant power reserves from the PV array. This allows the PV system to operate near its Maximum Power Point (MPP) more often, balancing grid support with generation efficiency. The advent of such sophisticated control in solar inverters necessitates equally advanced testing methodologies to validate their performance and reliability under realistic and challenging grid conditions.
Traditional testing setups often fall short. A standard grid simulator typically focuses solely on replicating AC-side grid characteristics—voltage sags, swells, frequency deviations, and harmonic distortions. While essential, this approach ignores the critical influence of the DC-side power source dynamics. For GFM solar inverters employing Matching Control, where the control logic is intrinsically linked to both AC grid frequency and DC-link voltage, testing only the AC response is insufficient. It fails to capture the complex, coupled dynamics of the inverter’s AC-DC coordinated control strategy. A true assessment requires simultaneous simulation of both the AC grid’s electrical environment and the dynamic output behavior of the PV generator itself. This paper, therefore, articulates an innovative dual-port operating condition simulation method, proposing a comprehensive test platform designed to meet the rigorous demands of next-generation GFM solar inverters.
Architecture of the AC-DC Dual-Port Simulation Platform
The core innovation of the proposed methodology is the holistic simulation of the inverter’s complete operational environment. The test platform is architected to interface with both the AC and DC ports of the Device Under Test (DUT), which is the GFM solar inverter. The platform consists of two main programmable power electronic converters sharing a common DC bus, supplied by a relatively low-power external DC source.
1. The AC Grid Simulator: This is a bidirectional AC-DC converter (typically a two-level or multi-level inverter) connected to the AC terminals of the DUT. Its primary function is to emulate a realistic grid connection point, including:
- Grid Voltage Source: Generating a fundamental AC voltage with programmable magnitude, frequency, and phase.
- Grid Impedance: Emulating the Thevenin equivalent impedance of the grid, typically represented as a series RL circuit, which defines the grid strength (Short-Circuit Ratio – SCR).
2. The DC Source Simulator: This is a bidirectional DC-DC converter connected to the DC terminals of the DUT. Its role is to emulate the dynamic I-V and P-V characteristics of a PV array or a generic DC source. A Dual Active Bridge (DAB) topology is highly advantageous for this role due to its inherent galvanic isolation, which blocks zero-sequence current paths in the test loop, enhancing safety and signal fidelity.
3. Power Flow and Efficiency: The three converters—the AC Grid Simulator, the DC Source Simulator, and the DUT—form a closed power loop. The active power generated by the DC simulator is processed by the DUT and injected into the AC simulator, which then feeds it back to the DC bus. The external DC source only needs to supply the total power losses of the three converter stages, drastically reducing its required power rating and the overall test system’s energy consumption.
The synergy of these components creates a flexible, efficient, and powerful testbed capable of subjecting GFM solar inverters to a vast array of combined AC and DC stress scenarios.
Control Strategies for Realistic Condition Emulation
The fidelity of the simulation hinges on the control algorithms implemented in the grid and source simulators. These controls must accurately replicate physical phenomena in real-time.
DC-Side PV Array Emulation
The DC Source Simulator (DAB) is controlled to mimic a PV array’s behavior. For many stability and grid-support function tests, the primary characteristic to emulate is the power delivery dynamic. The PV source can be modeled as a controlled power source, where the power command can be set to a constant (for MPPT operation) or varied dynamically (to simulate cloud passing events). The DAB utilizes a power control loop. It measures its output voltage and current to calculate the actual output power, compares it to the desired power reference (P_ref), and uses a PI controller to generate the phase-shift modulation index (d) for the DAB bridges, thereby regulating the power flow.
$$P_{out} = V_{dc\_dut} \cdot I_{dc\_dut}$$
$$\delta = PI(P_{ref} – P_{out})$$
Where $P_{ref}$ is the emulated PV power setpoint, which can follow profiles like the one shown in Table 1.
| Time Segment (s) | Power Reference (kW) | Emulated Scenario |
|---|---|---|
| 0.0 – 1.0 | 20.0 | Steady-state, full sun |
| 1.0 – 1.1 | 20.0 → 15.0 | Fast cloud-induced ramp-down |
| 1.1 – 2.5 | 15.0 | Partial shading |
| 2.5 – 2.6 | 15.0 → 20.0 | Cloud passage, ramp-up |
AC-Side Grid Emulation with Virtual Impedance
The AC Grid Simulator must replicate both an ideal voltage source ($v_s$) and the series grid impedance ($Z_{line}(s)$). This is achieved through a combined Virtual Impedance and Voltage/Current Control strategy. The control objective is to make the Point of Common Coupling (PCC) voltage behave as if it were sourced through a physical RL line.
The desired PCC voltage reference ($v_{g\_ref}$) is calculated by subtracting the voltage drop across the virtual impedance from the ideal source voltage:
$$v_{g\_ref} = v_s – i_g \cdot Z_{line}(s) = v_s – i_g \cdot (sL_{line} + R_{line})$$
where $i_g$ is the current injected by the DUT.
The direct implementation of the derivative term ($sL_{line}$) is problematic as it amplifies noise. A practical and stable implementation uses a low-pass filter to approximate the virtual impedance transfer function $G_{VI}(s)$:
$$v_{g\_ref} = v_s – i_g \cdot G_{VI}(s) = v_s – i_g \cdot \frac{sL_{line} + R_{line}}{1 + T s}$$
Here, $T$ is a small time constant chosen to provide adequate noise filtering without introducing significant phase delay at fundamental and low harmonic frequencies. This $v_{g\_ref}$ is then tracked by cascaded voltage and current controllers, typically employing Proportional-Resonant (PR) regulators for zero steady-state error at the fundamental frequency.
Grid-Forming Control in Solar Inverters: The Foundation for Testing
To appreciate the test requirements, one must understand the control paradigms of the DUT—the GFM solar inverter. Two primary GFM control schemes are relevant.
Virtual Synchronous Generator (VSG) Control
VSG control directly mimics the swing equation of a synchronous generator. It calculates the virtual rotor speed ($\omega$) and angle ($\theta$) based on active power imbalance, providing inherent inertia (J) and damping (D_p).
$$\omega = \omega_0 – \frac{1}{J s + D_p}(P – P_0)$$
$$\theta = \int \omega \, dt$$
The voltage magnitude (E) is regulated based on reactive power (Q) deviation:
$$E = E_0 – D_q(Q – Q_0)$$
While effective, a VSG controlling the grid-side inverter of a PV system requires the DC source (or the PV-side converter) to provide rapid power adjustments, often forcing the PV off its MPP.
Matching Control for PV Systems
This control is particularly suited for solar inverters as it elegantly ties DC-link dynamics to grid support. It establishes a “matching” relationship between the DC-link capacitor energy and the virtual rotor kinetic energy. The key equation replaces the active power feedback with a DC voltage squared term:
$$\omega = \omega_0 – \frac{1}{D_p}(u_{dc}^2 – u_{dc\_ref}^2)$$
The reactive power-voltage loop remains similar to VSG. This method allows the DC-link voltage to deviate, absorbing or releasing energy from the capacitor to provide inertial response, while the primary power source (the PV array) can maintain a near-constant power output. The test platform must therefore accurately simulate the DC source’s constant-power behavior and the AC grid’s frequency excursions to properly stress this control loop.
| Feature | Virtual Synchronous Generator (VSG) | Matching Control |
|---|---|---|
| Primary Input for Frequency Regulation | Active Power (P) | DC-Link Voltage Squared ($u_{dc}^2$) |
| Energy Buffer for Inertia | Requires external energy storage (e.g., battery) or de-rated PV power. | DC-link capacitor. |
| Impact on PV Operation | Typically requires power de-rating for headroom. | Allows operation closer to MPP. |
| Key Test Focus | AC grid frequency/power disturbance response. | Coupled AC frequency and DC voltage dynamic response. |
Simulation-Based Verification of the Methodology
The effectiveness of the proposed dual-port simulation concept is validated through detailed simulation models. A system is constructed in a tool like PLECS, comprising the DUT (with Matching Control), the AC Grid Simulator with virtual impedance, and the DC Source Simulator (DAB).
Case 1: AC Grid Frequency Disturbance
Scenario: The grid frequency undergoes a ramp increase from 50 Hz to 51 Hz over 2.5 seconds, simulating a sudden loss of generation.
Expected DUT Response (Matching Control): According to the control law $\omega = \omega_0 – K(u_{dc}^2 – u_{dc\_ref}^2)$, an increase in grid frequency ($\omega$) should cause a decrease in $u_{dc}^2$, meaning the DC-link voltage must drop to release energy and oppose the frequency rise. Concurrently, the active power output should decrease.
Simulation Result Metrics:
- DC-link voltage deviates from 750V to a lower steady-state value.
- Active power output reduces proportionally.
- The rate of change of frequency (RoCoF) is mitigated by the inverter’s response.
This test validates the inertial response of the GFM solar inverter and the platform’s ability to impose precise AC frequency transients.
Case 2: DC-Side Power Transient
Scenario: The emulated PV power from the DC Source Simulator steps down from 20 kW to 15 kW, simulating a cloud cover event.
Expected DUT Response: The instantaneous power imbalance causes the DC-link voltage to fall as energy is drained from the capacitor. The Matching Control interprets this voltage drop as a need to reduce the output frequency/power to restore equilibrium. The inverter’s power output should smoothly track the new available power.
Simulation Result Metrics:
- DC-link voltage dips momentarily but is stabilized by the control.
- AC output power settles at 15 kW.
- Grid frequency and voltage remain stable during the transition.
This test demonstrates the platform’s capability to simulate source-side dynamics and verifies the inverter’s stable power tracking under variable generation.
Case 3: Advantage of DAB Topology in Zero-Sequence Current Suppression
A comparative simulation between a non-isolated DC-DC converter (like Buck-Boost) and the DAB topology for the DC simulator reveals a critical advantage. In a back-to-back test setup, parasitic capacitances can create zero-sequence current loops. The simulation shows:
- Non-isolated Topology: Significant high-frequency zero-sequence current ($i_0$) circulation is observed, which can interfere with measurements, increase losses, and cause EMI issues.
$$i_{0\_buckboost} = \text{Significant (e.g., several Amperes)}$$ - DAB Topology: The high-frequency transformer provides galvanic isolation, effectively blocking the zero-sequence path.
$$i_{0\_DAB} \approx 0 \text{ A}$$
This validates the choice of an isolated DC-DC converter as a best practice for constructing a clean and reliable test platform for solar inverters.
Conclusion and Perspective
The transition towards grids dominated by inverter-based resources like solar inverters necessitates a paradigm shift in testing philosophies. The proposed AC-DC dual-port full-condition simulation method addresses a critical gap in the validation of advanced Grid-Forming solar inverters. By simultaneously and synchronously emulating the dynamic characteristics of both the AC grid and the DC power source, this platform enables a holistic evaluation of the inverter’s coupled control strategies, such as the increasingly important Matching Control.
The key benefits of this methodology are multifold:
- Comprehensive Validation: It tests the integrated AC-DC behavior, revealing interactions that single-port tests would miss.
- High Flexibility: Grid strength (SCR), fault types, voltage/frequency profiles, and PV generation curves can all be programmed via software, enabling a vast test matrix without hardware changes.
- Energy Efficiency: The closed-loop power flow minimizes the utility power consumption of the test bench, reducing operational costs.
- Improved Signal Integrity: The use of an isolated DAB topology for DC simulation mitigates problematic zero-sequence currents, leading to cleaner and more accurate measurements.
This methodology establishes a robust foundation for the type approval, performance certification, and resilience analysis of next-generation solar inverters. As grid codes evolve to mandate specific GFM functionalities, such test platforms will become indispensable tools for manufacturers, utilities, and certification bodies to ensure that the future fleet of solar inverters will reliably and securely support the transition to a sustainable energy system.