In recent years, the integration of distributed photovoltaic (PV) generation into power grids has accelerated, driven by global carbon neutrality goals. However, the increasing penetration of nonlinear loads in AC microgrids, such as those comprising PV systems with energy storage, often leads to current waveform distortion and harmonic pollution at the point of common coupling (PCC). This deteriorates power quality, causing issues like reduced power factor and interference with grid operations. To address these challenges, we propose a multifunctional grid-tied inverter (MFGCI) that combines conventional grid-tied inversion with active power filtering capabilities. This system not only injects power from PV and storage sources into the grid but also compensates for harmonic and reactive currents, thereby enhancing overall power quality. Our approach leverages an LCL-filter-based inverter structure and a composite control strategy integrating PI and repetitive controllers, ensuring precise current tracking and robust performance under varying conditions. This article details the design, control, and validation of this multifunctional grid-tied inverter, emphasizing its dual functionality and effectiveness in diverse operational modes.
The core of our microgrid system is the multifunctional grid-tied inverter, which interfaces PV panels, an energy storage unit, and the AC grid. As illustrated below, the system includes a PV array, a bidirectional DC-DC converter for storage management, nonlinear loads (e.g., a three-phase uncontrolled rectifier), and the grid-tied inverter itself. The inverter is connected to the PCC via an LCL filter, which attenuates switching harmonics, while the nonlinear load draws distorted currents. The DC bus voltage is maintained at a stable level to facilitate power exchange between AC and DC sides. This configuration allows the grid-tied inverter to operate not only as a power converter but also as an active power filter (APF), compensating for load-induced harmonics and reactive power. The energy storage unit, typically a lithium-ion battery, plays a crucial role in balancing power fluctuations, ensuring system stability when PV output or load demand changes abruptly.
To achieve dual functionality, the control strategy for the multifunctional grid-tied inverter is divided into several key components: command current synthesis, DC bus voltage regulation, and current loop control. The command current is derived using the instantaneous reactive power theory (ip-iq method), which extracts harmonic and reactive components from the nonlinear load current. In the dq-reference frame, the load currents \(i_{La}\), \(i_{Lb}\), \(i_{Lc}\) are transformed as follows:
$$ \begin{bmatrix} i_{Ld} \\ i_{Lq} \end{bmatrix} = \frac{2}{3} \begin{bmatrix} \sin(\omega t) & \sin(\omega t – 120^\circ) & \sin(\omega t + 120^\circ) \\ \cos(\omega t) & \cos(\omega t – 120^\circ) & \cos(\omega t + 120^\circ) \end{bmatrix} \begin{bmatrix} i_{La} \\ i_{Lb} \\ i_{Lc} \end{bmatrix} $$
After low-pass filtering, the DC components \(i_{Lp}\) and \(i_{Lq}\) represent the fundamental active and reactive currents, respectively. The harmonic currents are obtained by subtracting the fundamental components from the original load currents. For reactive power compensation, \(i_{Lq}\) is set to zero. The final reference current \(i_c^*\) for the grid-tied inverter includes both harmonic compensation and an active current component \(\Delta i_p\) from DC voltage control, expressed as:
$$ i_c^* = i_L – i_{Lf} + \Delta i_p $$
where \(i_{Lf}\) is the fundamental load current. This ensures that the grid-tied inverter injects compensating currents while managing active power flow.
DC bus voltage stability is critical for proper operation of the multifunctional grid-tied inverter. A PI controller regulates the DC voltage \(V_{dc}\) by comparing it with a reference \(V_{dc}^*\) and generating the active current component \(\Delta i_p\). This component adjusts the power exchange to maintain voltage within a narrow band, typically ±5 V around 800 V in our system. The control law is:
$$ \Delta i_p = K_{p,V} (V_{dc}^* – V_{dc}) + K_{i,V} \int (V_{dc}^* – V_{dc}) \, dt $$
where \(K_{p,V}\) and \(K_{i,V}\) are proportional and integral gains. This approach enables the grid-tied inverter to stabilize the DC link, facilitating seamless power transfer from PV and storage sources to the grid.
The current loop employs a composite control scheme combining PI and repetitive controllers to achieve fast dynamic response and high tracking accuracy. The repetitive controller, based on the internal model principle, eliminates steady-state errors for periodic signals like harmonics, but it introduces a delay of \(N\) samples (where \(N = f_s / f_{grid}\), with \(f_s\) as sampling frequency). The PI controller compensates for this by providing rapid transient response. The overall control structure for the grid-tied inverter current is shown in the block diagram, with the plant model \(G(z)\) representing the LCL filter and inverter dynamics. In continuous time, the LCL transfer function is:
$$ G(s) = \frac{i_g(s)}{V_{dc}(s)} = \frac{1}{L_1 L_2 C s^3 + (L_1 + L_2)s} $$
For our design, parameters are \(L_1 = 1000 \, \mu\text{H}\), \(L_2 = 10 \, \mu\text{H}\), and \(C = 50 \, \mu\text{F}\). Discretizing with a sample time \(T_s\) yields \(G(z)\). The repetitive controller transfer function in the z-domain is:
$$ G_1(z) = \frac{z^{-N} K_r z^k S(z)}{1 – z^{-N} Q(z)} $$
where \(K_r = 0.8\) is the gain, \(k = 3\) provides phase lead, \(Q(z) = 0.5\) ensures stability, and \(S(z)\) is a second-order low-pass filter:
$$ S(s) = \frac{\omega_0^2}{s^2 + 2\xi \omega_0 s + \omega_0^2} $$
with \(\omega_0 = 2000 \, \text{rad/s}\) and \(\xi = 0.707\). The PI controller in discrete form is:
$$ G_2(z) = K_p + K_i \frac{T_s}{z – 1} $$
The composite control output for the grid-tied inverter is then:
$$ \frac{i_c(z)}{i_c^*(z)} = \frac{[G_1(z) + G_2(z)] G(z)}{1 + [G_1(z) + G_2(z)] G(z)} $$
This strategy enables the grid-tied inverter to accurately track reference currents, suppressing harmonics up to the 25th order with minimal total harmonic distortion (THD).
Energy storage coordination is essential for managing power imbalances due to PV variability and load changes. The storage unit, controlled via a bidirectional DC-DC converter, charges or discharges based on a power reference \(P_{\text{ref}}\). This reference is derived from the power balance equation:
$$ P_{\text{pv}} + P_B = P_L – P_g $$
where \(P_{\text{pv}}\) is PV output, \(P_B\) is storage power (positive for discharge), \(P_L\) is load power, and \(P_g\) is grid power (positive for grid supply). The grid current d-component \(i_{gd}\) relates to power as:
$$ P_g = i_{gd} u_{gd} $$
with \(u_{gd}\) as the grid voltage d-component. By regulating \(i_{gd}\) through the multifunctional grid-tied inverter, we achieve four operational modes, summarized in Table 1. These modes ensure that the grid-tied inverter maintains power quality while adapting to different power flow scenarios.
| Mode | Condition | Power Flow Description | Storage Action | Grid Power \(P_g\) |
|---|---|---|---|---|
| 1 | \(P_{\text{pv}} – P_L < 0\) | PV deficit: Storage and grid supply load | Discharge | > 0 (grid supplies) |
| 2 | \(P_{\text{pv}} – P_L > 0\) | PV surplus: Excess powers storage and grid | Charge | < 0 (grid receives) |
| 3 | \(P_{\text{pv}} – P_L < 0\), load surge | High demand: PV, storage, and grid supply load | Discharge | > 0 |
| 4 | \(P_{\text{pv}} – P_L > 0\), load drop | Low demand: Excess powers storage and grid | Charge | < 0 |
Simulation studies validate the performance of our multifunctional grid-tied inverter. Using MATLAB/Simulink, we modeled the system with parameters listed in Table 2. The grid-tied inverter was tested under the four modes, with nonlinear loads generating distorted currents (THD ~26% before compensation). After activating the APF function with PI+repetitive control, the grid current THD reduced significantly, as shown in Table 3. In all modes, the grid-tied inverter maintained THD below 5%, meeting IEEE standards, and improved the power factor to nearly unity. The DC bus voltage remained stable at \(800 \pm 1 \, \text{V}\), demonstrating effective regulation.
| Component | Parameter | Value |
|---|---|---|
| PV Array | Maximum Power | 5-25 kW (variable) |
| Energy Storage | Capacity | 20 kWh (LiFePO4) |
| Grid | Voltage & Frequency | 380 V (line-line), 50 Hz |
| LCL Filter | \(L_1\), \(L_2\), \(C\) | 1000 μH, 10 μH, 50 μF |
| DC Bus | Reference Voltage \(V_{dc}^*\) | 800 V |
| Switching Frequency | \(f_{\text{sw}}\) | 10 kHz |
| Control Gains | \(K_p\), \(K_i\) (current) | 0.5, 100 |
| Repetitive Control | \(K_r\), \(Q(z)\), \(N\) | 0.8, 0.5, 200 |
The dynamic response of the grid-tied inverter was evaluated during load transitions. For instance, in Mode 3, when load power increased suddenly from 7 kW to 17 kW at t=1 s, the grid-tied inverter quickly adjusted its output, maintaining sinusoidal grid current with minimal overshoot. Similarly, in Mode 4, a load reduction at t=1.5 s resulted in smooth power feedback to the grid. The composite control strategy proved superior to PI-only control, reducing THD from 3.73% to 2.45% in comparative tests. Waveforms showed that grid voltage and current remained in phase (or 180° out of phase during reverse power flow), confirming reactive power compensation. The multifunctional grid-tied inverter thus successfully operated as both a power converter and an active filter across all modes.
| Mode | THD Before Compensation (%) | THD After Compensation (%) | Power Factor | DC Voltage Error (%) |
|---|---|---|---|---|
| 1 | 26.07 | 2.63 | 0.998 | 0.125 |
| 2 | 26.07 | 1.89 | 0.999 | 0.125 |
| 3 | 26.07 | 2.20 | 0.997 | 0.125 |
| 4 | 26.07 | 1.90 | 0.999 | 0.125 |
Further analysis involves the stability of the grid-tied inverter control system. The frequency response of the composite controller was examined using Bode plots. The repetitive controller provides high gain at harmonic frequencies (e.g., 100 Hz, 150 Hz, etc.), effectively rejecting periodic disturbances, while the PI controller ensures adequate phase margin for robustness. The overall system stability margin exceeded 45°, indicating reliable operation even under grid impedance variations. Additionally, the impact of parameter uncertainties on the grid-tied inverter was studied via sensitivity analysis. Variations in LCL filter components up to ±10% caused less than 1% change in THD, demonstrating the controller’s resilience.
Energy management in the microgrid relies heavily on the coordinated control of the grid-tied inverter and storage. The power reference \(P_{\text{ref}}\) for storage is generated based on the SOC limits and power balance requirements. For example, when SOC is within safe bounds (e.g., 20%-80%), the storage unit balances PV-load mismatches, allowing the grid-tied inverter to focus on power quality. If SOC reaches limits, the grid takes over balance duties. This logic ensures long-term sustainability of the storage system while optimizing the grid-tied inverter’s dual functions. The control algorithm can be summarized as:
$$ P_B = \begin{cases}
P_L – P_{\text{pv}} & \text{if } SOC_{\text{min}} < SOC < SOC_{\text{max}} \\
0 & \text{otherwise}
\end{cases} $$
In practice, hysteresis bands are added to prevent frequent mode switching. The grid-tied inverter continuously monitors \(P_{\text{pv}}\), \(P_L\), and SOC to adjust \(i_c^*\) accordingly, ensuring seamless transitions between modes.
Comparative studies with conventional grid-tied inverters highlight the advantages of our multifunctional approach. A standard grid-tied inverter only transfers active power, requiring separate APF units for harmonic mitigation, which increases cost and complexity. Our integrated solution reduces component count and improves overall efficiency. For instance, the multifunctional grid-tied inverter achieves a combined efficiency of 96.5% for power conversion and filtering, compared to 94% for separate systems. Moreover, the use of repetitive control enhances harmonic suppression beyond typical PI-based methods, as shown in Table 4. The grid-tied inverter’s ability to adapt to various grid conditions makes it suitable for modern microgrids with high renewable penetration.
| Control Method | THD Reduction (%) | Dynamic Response Time (ms) | Complexity | Suitability for Multifunction |
|---|---|---|---|---|
| PI Only | 3.73 | 50 | Low | Moderate |
| Repetitive Only | 2.50 | 200 | High | High |
| PI + Repetitive (Proposed) | 2.45 | 60 | Medium | Excellent |
| QPR Control | 2.80 | 70 | Medium | Good |
Future work could explore advanced features for the grid-tied inverter, such as adaptive repetitive control that adjusts to grid frequency variations, or integration with machine learning for predictive energy management. Additionally, hardware-in-the-loop testing would validate real-time performance. The scalability of the multifunctional grid-tied inverter to larger systems or DC microgrids also merits investigation.
In conclusion, we have presented a comprehensive study on a multifunctional grid-tied inverter for PV-energy storage microgrids. The system combines power injection and active power filtering capabilities, addressing both energy transfer and power quality concerns. Through a composite control strategy of PI and repetitive controllers, the grid-tied inverter achieves precise current tracking, harmonic suppression, and reactive power compensation. Simulation results across four operational modes confirm that the grid-tied inverter maintains low THD (below 5%), high power factor (near unity), and stable DC voltage under dynamic conditions. The integration of energy storage enables flexible power management, enhancing system reliability. This work demonstrates that the multifunctional grid-tied inverter is a viable solution for modern renewable-based microgrids, offering economic and technical benefits over conventional approaches. As grid-tied inverter technology evolves, such multifunctional designs will play a pivotal role in achieving sustainable and high-quality power systems.