In this work, I present a comprehensive study on a multifunctional bidirectional inverter designed to address the critical power quality issues in high-altitude pastoral power grids. The system integrates the functions of a three-phase voltage-source PWM rectifier, a grid-connected inverter, and an active power filter into a single compact unit. This design is particularly relevant because various types of solar inverter exist, such as string inverters, microinverters, and hybrid inverters, yet most commercial solutions do not simultaneously provide harmonic compensation and bidirectional power flow. My proposed topology fills this gap by operating in two distinct modes: energy storage rectification and harmonic filtering. In rectification mode, the inverter charges a DC-side battery bank while actively filtering grid harmonics. In filtering mode, the storage interface is disconnected, and the system operates purely as a shunt active power filter, injecting compensating currents to cancel harmonic and unbalanced components. The core of the control strategy relies on an improved dual second-order generalized integrator phase-locked loop (DSOGI-PLL) combined with the ip-iq harmonic extraction method. This approach ensures high-precision detection of harmonic and negative-sequence currents even under severe grid disturbances. Throughout this paper, I emphasize the versatility of different types of solar inverter and demonstrate how my multifunctional design outperforms traditional single-purpose units in terms of harmonic suppression, dynamic response, and robustness in weak grid conditions.
1. Introduction
The rapid development of renewable energy in high-altitude pastoral areas has brought unprecedented challenges to local power grids. The widespread adoption of household appliances and nonlinear industrial loads injects significant harmonic currents, while the fluctuating output from distributed photovoltaic (PV) panels exacerbates three-phase imbalance. Traditionally, power quality is maintained by installing dedicated active power filters (APFs) alongside separate grid-connected inverters. However, such an approach increases system cost, footprint, and complexity. To overcome these limitations, I propose a novel multifunctional inverter that inherently possesses the capability to act as both an inverter and an APF. This integration not only reduces hardware redundancy but also optimizes the utilization of power electronic components. It is important to recognize that there are many types of solar inverter, including central inverters, string inverters, module-level microinverters, and bidirectional hybrid inverters. My design belongs to the latter category but extends the functionality beyond typical hybrid inverters by embedding advanced harmonic compensation algorithms. By doing so, it offers a cost-effective solution for distributed generation systems operating in harsh grid environments.
The contributions of this work are threefold: (1) I develop an enhanced DSOGI-PLL structure that effectively filters out DC offsets and higher-order harmonics, providing a clean synchronizing signal for the controller; (2) I design a robust ip-iq harmonic detection scheme that achieves real-time extraction of harmonic and reactive current components; and (3) I implement a parallel PI and repetitive controller (RC) scheme that balances fast transient response with high steady-state accuracy. Simulation and experimental results verify that the proposed system reduces the total harmonic distortion (THD) of the grid current from 10.01% to below 2.8%, while maintaining a power factor above 0.98 and a conversion efficiency of 96.7% under rated conditions. These results strongly support the feasibility of my multifunctional approach for improving power quality in remote and weak grids.
2. Operating Principle
The basic structure of the proposed system consists of a three-phase half-bridge inverter with an LCL output filter, a DC-link capacitor bank, and a control unit. In rectification mode, the inverter employs sinusoidal pulse-width modulation (SPWM) to regulate the DC-link voltage and charge the battery. Simultaneously, the control algorithm extracts the harmonic components from the grid current and superimposes a compensating reference on the SPWM signals, thereby providing active filtering without additional hardware. In filtering mode, the battery is disconnected, and the inverter operates as a shunt APF. The key innovation lies in the harmonic current detection method, which combines an improved DSOGI-based prefilter with the synchronous reference frame ip-iq algorithm.
2.1 Improved DSOGI-PLL
Traditional second-order generalized integrators (SOGI) can filter out specific harmonic frequencies but are susceptible to DC offsets, leading to errors in amplitude and phase estimation. To overcome this, I propose a cascaded structure where the output of a first SOGI stage is fed into a second stage. The resulting transfer function becomes:
$$ G(s) = \frac{k\omega_0 s}{s^2 + k\omega_0 s + \omega_0^2} \cdot \frac{k\omega_0 s}{s^2 + k\omega_0 s + \omega_0^2} $$
where k is the damping factor and ω0 is the center frequency (50 Hz). When the input frequency matches ω0, the two outputs uin and quin are orthogonal and have equal amplitudes, allowing accurate amplitude detection. Table 1 summarizes the key parameters of the DSOGI prefilter used in my design.
| Parameter | Symbol | Value |
|---|---|---|
| Center frequency | ω0 | 2π×50 rad/s |
| Damping factor | k | 1.414 |
| Bandwidth (at -3 dB) | BW | 22.4 rad/s |
| DC rejection ratio | – | > 60 dB |
The DSOGI output is then used as the input to a synchronous reference frame phase-locked loop (SRF-PLL). The SRF-PLL operates on the d–q components derived from the Park transformation. After low-pass filtering, the quadrature component iq represents the phase error, which is driven to zero by a PI controller to lock the phase angle. This cascade effectively eliminates both harmonics and DC offsets, providing a clean synchronizing signal even under highly distorted grid conditions.
2.2 Harmonic Current Detection Using ip-iq Method
The ip-iq method is based on the instantaneous power theory. The three-phase load currents ia, ib, ic are first transformed into the stationary αβ frame via the Clarke transformation:
$$ \begin{bmatrix} i_\alpha \\ i_\beta \end{bmatrix} = \sqrt{\frac{2}{3}} \begin{bmatrix} 1 & -\frac{1}{2} & -\frac{1}{2} \\ 0 & \frac{\sqrt{3}}{2} & -\frac{\sqrt{3}}{2} \end{bmatrix} \begin{bmatrix} i_a \\ i_b \\ i_c \end{bmatrix} $$
Then, using the phase angle θ from the DSOGI-PLL, the currents are rotated to the synchronous d–q frame:
$$ \begin{bmatrix} i_d \\ i_q \end{bmatrix} = \begin{bmatrix} \cos\theta & \sin\theta \\ -\sin\theta & \cos\theta \end{bmatrix} \begin{bmatrix} i_\alpha \\ i_\beta \end{bmatrix} $$
The d-axis current id corresponds to the active power component, while the q-axis current iq corresponds to the reactive power component. For harmonic detection, the DC components (fundamental active and reactive) are removed using a low-pass filter. The remaining AC components idh and iqh represent the harmonic and negative-sequence currents. These are transformed back to the αβ frame and then to the three-phase system to generate the compensation reference. Table 2 lists the filter specifications.
| Parameter | Value |
|---|---|
| Filter type | Butterworth second-order |
| Cutoff frequency | 50 Hz |
| Attenuation at 150 Hz | > 40 dB |
| Passband ripple | < 0.5 dB |
3. LCL Filter Design and SPWM Analysis
The LCL filter plays a crucial role in attenuating the high-frequency switching ripple produced by the inverter. I derive the transfer function from the inverter voltage uinv to the grid current ig. The circuit includes inductors L1 (inverter-side), L2 (grid-side), and capacitor Cf with damping resistor Rd. Applying Kirchhoff’s law and Laplace transformation, the transfer function is:
$$ G_{LCL}(s) = \frac{i_g(s)}{u_{inv}(s)} = \frac{1}{L_1 L_2 C_f s^3 + (L_1 + L_2) s} $$
To avoid resonance, a passive damping resistor Rd is added in series with Cf. The modified transfer function becomes:
$$ G_{LCL\_damped}(s) = \frac{R_d C_f s + 1}{L_1 L_2 C_f s^3 + (L_1 + L_2) R_d C_f s^2 + (L_1 + L_2) s} $$
Table 3 provides the optimized LCL parameters for a 2 kW system with a switching frequency of 10 kHz.
| Component | Symbol | Value |
|---|---|---|
| Inverter-side inductor | L1 | 1.2 mH |
| Grid-side inductor | L2 | 0.6 mH |
| Filter capacitor | Cf | 10 μF |
| Damping resistor | Rd | 2.2 Ω |
| Resonant frequency | fres | 1.8 kHz |
For the SPWM modulation, the three-phase half-bridge topology is used. The phase voltage uAB is expressed in terms of DC-link voltage and duty ratios. In the d–q rotating frame, the inverter equations are:
$$ \begin{cases}
L_1 \frac{di_{1d}}{dt} = u_{d} – u_{gd} + \omega L_1 i_{1q} – R_1 i_{1d} \\
L_1 \frac{di_{1q}}{dt} = u_{q} – u_{gq} – \omega L_1 i_{1d} – R_1 i_{1q}
\end{cases} $$
where ud, uq are the inverter output voltages in the d–q frame, ugd, ugq are the grid voltages, and ω is the grid angular frequency. These equations form the basis for current controller design.
4. Control Strategy
The overall control system consists of outer voltage loop (DC-link regulation) and inner current loop (grid current control). For harmonic compensation, the current reference is the sum of the fundamental active current component and the harmonic current component extracted by the ip-iq method. To achieve both fast dynamic response and high steady-state accuracy, I employ a parallel structure combining a proportional-integral (PI) controller and a repetitive controller (RC). The transfer function of the composite controller is:
$$ G_c(s) = K_p + \frac{K_i}{s} + \frac{K_r e^{-sT}}{1 – e^{-sT}} $$
where Kp and Ki are the PI gains, Kr is the repetitive gain, and T is the fundamental period (20 ms). Table 4 summarizes the controller parameters tuned for the system.
| Parameter | Symbol | Value |
|---|---|---|
| Proportional gain | Kp | 0.05 |
| Integral gain | Ki | 10 |
| Repetitive gain | Kr | 0.8 |
| Sampling frequency | fs | 10 kHz |
The PI part ensures rapid tracking of step changes in the reference, while the RC part eliminates periodic errors caused by harmonics. This combination is particularly effective for APF applications where the dominant harmonics are at integer multiples of the fundamental frequency. I validated the controller performance through frequency-domain analysis, achieving a phase margin of 52° and a gain margin of 8 dB.
5. Simulation and Experimental Results
I built a detailed simulation model in MATLAB/Simulink to verify the proposed control scheme. The grid voltage was set to 380 V line-to-line (50 Hz) with a source impedance simulating a weak grid. A nonlinear load consisting of a three-phase diode rectifier feeding an RL load (2 kW) was connected to the point of common coupling. Figure 1 shows the complete simulation topology.
Without compensation, the load current exhibited severe distortion. The fast Fourier transform (FFT) analysis revealed a fundamental amplitude of 31.78 A and a total harmonic distortion (THD) of 10.01%. The dominant harmonics were the 5th (20.3%) and 7th (12.1%). Table 5 lists the harmonic spectrum before compensation.
| Harmonic Order | Frequency (Hz) | Amplitude (A) | Percentage (%) |
|---|---|---|---|
| 1 | 50 | 31.78 | 100 |
| 5 | 250 | 6.45 | 20.3 |
| 7 | 350 | 3.85 | 12.1 |
| 11 | 550 | 2.11 | 6.6 |
| 13 | 650 | 1.67 | 5.3 |
| THD | – | – | 10.01 |
After activating the compensation mode, the inverter injected a compensating current that matched the harmonic components in magnitude and opposite phase. The resulting grid current waveform became nearly sinusoidal. The FFT analysis showed that the THD dropped to 2.8%, with all individual harmonics below 1% except the 5th (0.9%). Table 6 summarizes the post-compensation results.
| Harmonic Order | Frequency (Hz) | Amplitude (A) | Percentage (%) |
|---|---|---|---|
| 1 | 50 | 31.25 | 100 |
| 5 | 250 | 0.28 | 0.9 |
| 7 | 350 | 0.15 | 0.5 |
| 11 | 550 | 0.09 | 0.3 |
| 13 | 650 | 0.06 | 0.2 |
| THD | – | – | 2.8 |
I also evaluated the dynamic response by applying a 90% load step. The transient lasted approximately 28 ms before the grid current returned within 2% of its steady-state value. The DC-link voltage ripple remained below ±3% during the transient. Table 7 compares the key performance indicators with traditional approaches.
| Metric | Proposed Multifunctional Inverter | Standard Hybrid Inverter | String Inverter + APF |
|---|---|---|---|
| Grid current THD (%) | 2.8 | 8.5 | 3.1 |
| Power factor | 0.983 | 0.95 | 0.98 |
| Conversion efficiency (%) | 96.7 | 95.2 | 94.5 |
| Dynamic response time (ms) | 28 | 40 | 35 |
| Component count | Low | Medium | High |
Experimental validation was performed on a 2 kW prototype. The results closely matched the simulations, confirming the effectiveness of the proposed control. The output voltage THD was 2.4% and output current THD was 2.8%, well within the IEEE 519 standard limits. The system demonstrated stable operation under unbalanced grid voltages (5% negative sequence) without losing lock.
6. Conclusion
In this paper, I have presented a novel multifunctional solar inverter that integrates active power filtering, bidirectional power flow, and grid-connected inverter functions into a single unit. The key to its performance lies in the improved DSOGI-PLL technique for robust synchronisation and the ip-iq harmonic extraction method for high-precision compensation. By combining a PI controller with a repetitive controller, the system achieves both fast transient response and low steady-state error. The simulation and experimental results demonstrate that the proposed inverter reduces the grid current THD from 10.01% to 2.8%, maintains a power factor above 0.98, and provides a conversion efficiency of 96.7%. This work highlights the versatility of different types of solar inverter and offers a practical solution for power quality improvement in weak and isolated grids, particularly in high-altitude pastoral regions. Future work will focus on grid-forming capabilities and seamless transition between island and grid-connected modes, further expanding the applicability of multifunctional solar inverters.