In the rapidly evolving landscape of renewable energy integration, the stability of photovoltaic (PV) grid-connected systems under varying grid conditions remains a critical challenge. The proliferation of distributed generation systems based on clean energy sources such as solar and wind has led to a gradual weakening of the grid, where wide variations in grid impedance significantly affect system stability margins. In multi-inverter parallel systems, this issue is exacerbated by potential resonance interactions between grid-connected inverters and the grid itself. To address these challenges, we propose a multi-functional PV grid-connected inverter based on an energy-stored quasi-Z-source inverter, which simultaneously achieves grid-connected power generation and resonance suppression. This approach integrates the functionality of an active damper into the inverter itself, reducing system volume and cost while enhancing reliability.
Among various types of solar inverters, the quasi-Z-source inverter stands out due to its ability to perform single-stage buck-boost conversion, its tolerance for shoot-through states, and its high reliability. The energy-stored variant adds the capability to smooth PV power fluctuations without requiring an additional bidirectional DC-DC converter. This paper presents a comprehensive design methodology for the multi-functional inverter, including circuit parameter design criteria, control strategy development considering phase-locked loop (PLL) dynamics, and control parameter optimization. Experimental results validate the feasibility and effectiveness of the proposed approach.
System Architecture and Circuit Parameter Design
Multi-Functional Inverter Structure
The proposed multi-functional PV grid-connected inverter is connected in parallel with multiple PV inverters in a distributed generation system. The circuit topology of the multi-functional inverter is identical to that of the energy-stored quasi-Z-source inverter. The inverter operates in both shoot-through and non-shoot-through states. During the shoot-through state, the diode D1 is reverse-biased, the energy storage inductors L1 and L2 are magnetized, and the energy storage capacitors C1 and C2 are discharged. During the non-shoot-through state, the diode conducts, the capacitors are charged, and the inductors are demagnetized.
The voltage relationships under steady-state conditions are given by:
$$
u_{C1} = U_{PV} \frac{1 – D_0}{1 – 2D_0}
$$
$$
u_{C2} = U_{PV} \frac{D_0}{1 – 2D_0}
$$
$$
U_{PN} = u_{C1} + u_{C2} = \frac{U_{PV}}{1 – 2D_0}
$$
where \(D_0\) is the shoot-through duty ratio, \(U_{PV}\) is the PV input voltage, and \(U_{PN}\) is the DC bus voltage amplitude.
The capacity of the multi-functional grid-connected inverter must satisfy the damping requirements for resonance suppression. The rated apparent power \(S_N\) is determined by:
$$
S_N \geq \frac{3 \lambda_{kmax} U_g^2}{R_{V\_min}}
$$
where \(\lambda_{kmax}\) is the ratio of the maximum resonant component to the fundamental frequency in the PCC voltage, \(U_g\) is the RMS value of the grid phase voltage, and \(R_{V\_min}\) is the minimum virtual resistance simulated by the active damping function.
The maximum grid inductance \(L_{gmax}\) is related to the total output power \(P_{out}\) of the multi-inverter system:
$$
L_{gmax} \geq \frac{3U_g^2}{K_{SCR} \cdot 2\pi f_0 \cdot P_{out}}
$$
where \(K_{SCR}\) is the short-circuit ratio and \(f_0\) is the fundamental frequency.
Table 1 summarizes the key parameters of different types of solar inverters considered in this study, including the proposed multi-functional quasi-Z-source inverter and conventional inverter topologies.
| Parameter | Conventional Voltage-Source Inverter | Conventional Z-Source Inverter | Energy-Stored Quasi-Z-Source Inverter (Proposed) |
|---|---|---|---|
| Buck-Boost Capability | No | Yes | Yes |
| Shoot-Through Tolerance | No | Yes | Yes |
| Additional DC-DC Converter Required for Energy Storage | Yes | Yes | No |
| Component Count | Moderate | High | Moderate |
| Active Damper Function Integration | Requires External Hardware | Requires External Hardware | Inherently Supported |
| Reliability | Moderate | High | High |
Impedance Network Parameter Design
The design of the energy storage inductor \(L\) must consider high-frequency current ripple suppression, avoidance of impedance network resonance, and maintenance of continuous inductor current. Using the ZSVM2 modulation strategy, the maximum inductor current ripple \(\Delta i_L\) occurs at the midpoint of the sector. To suppress high-frequency current ripple:
$$
L \geq \frac{3 D_0 U_m}{2 \Delta i_L f_s}
$$
where \(U_m\) is the peak phase voltage and \(f_s\) is the switching frequency. To avoid resonance between the energy storage inductor \(L\) and capacitor \(C\):
$$
L \gg \frac{1}{4\pi^2 f_s^2 C}
$$
To maintain continuous inductor current:
$$
L > \frac{3 D_0 U_m U_{PV} \eta}{2 f_s (6 U_m I_o – \sqrt{2} I_o U_{PV} \eta)}
$$
where \(\eta\) is the inverter efficiency and \(I_o\) is the rated RMS output current. The final value of \(L\) must satisfy all three constraints.
For the energy storage capacitors \(C_1\) and \(C_2\), which are connected in series to form the DC bus capacitor \(C_{dc} = C_1 // C_2\), the design must satisfy both high-frequency voltage ripple suppression and DC bus voltage ripple limits imposed by the active damping function. The maximum capacitor voltage ripple occurs when the battery current \(i_B = 0\):
$$
C_k \geq \frac{I_{L2} \left| 3U_m – U_{PV}(1 – 2D_0) \right|}{2\gamma f_s U_{Ck} U_{PV}}
$$
where \(\gamma\) is the capacitor voltage ripple coefficient. The instantaneous power pulsation \(p͂(t)\) absorbed from the grid creates energy variations \(\Delta E\) that must be accommodated by the DC bus capacitor:
$$
p͂(t) =
\begin{cases}
3U_1 I_k \cos[(k+1)\omega_0 t + \phi_k], & k = 6m – 1, m = 1,2,3,\ldots \\
-3U_1 I_k \cos[(k-1)\omega_0 t + \phi_k], & k = 6m + 1, m = 1,2,3,\ldots
\end{cases}
$$
The energy variation must satisfy:
$$
|\Delta E| < \frac{1}{2} C_{dc} U_{PN\_max}^2 – \frac{1}{2} C_{dc} U_{PN\_min}^2 = C_{dc} U_{PN} \Delta U_{PN}
$$
Table 2 provides the designed impedance network parameters for the multi-functional inverter.
| Parameter | Symbol | Value |
|---|---|---|
| Energy Storage Inductor 1 | \(L_1\) | 1.6 mH |
| Energy Storage Inductor 2 | \(L_2\) | 1.6 mH |
| Energy Storage Capacitor 1 | \(C_1\) | 110 μF |
| Energy Storage Capacitor 2 | \(C_2\) | 120 μF |
| DC Bus Capacitance (Equivalent) | \(C_{dc}\) | \(C_1 // C_2\) |
| Shoot-Through Duty Ratio | \(D_0\) | 0.25 |
| PV Maximum Power Point Voltage | \(U_{PV}\) | 576 V |
| DC Bus Voltage | \(U_{PN}\) | 1152 V |
LCL Filter Parameter Design
The inverter-side inductor \(L_{f1}\) must limit the current ripple. The maximum ripple \(\Delta i_{max}\) occurs at the grid voltage zero-crossing point:
$$
L_{f1} \geq \frac{U_m T_s}{2\sqrt{3} \varepsilon_{Lf1} I_o}
$$
where \(\varepsilon_{Lf1}\) is the current ripple coefficient. The total inductance \(L_{sum} = L_{f1} + L_{f2}\) must satisfy the current slew rate requirements for the active damping function:
$$
L_{sum} \leq \min\left( \frac{U_L}{2\omega_{res} I_k}, \frac{U_L}{2\omega_0 I_o} \right)
$$
where \(U_L = U_{PN}/\sqrt{3} – 1.1U_m\) and \(\omega_{res}\) is the highest angular frequency of the target resonance frequency band. The filter capacitor \(C_f\) is limited by reactive power constraints:
$$
C_{f\_max} = \frac{5\% S_N}{3\omega_0 U_g^2}
$$
The grid-side inductor \(L_{f2}\) must satisfy both the system target resonance frequency band suppression requirements and the switching harmonic content limits. The natural resonance frequency \(f_r\) of the LCL filter is:
$$
f_r = \frac{1}{2\pi} \sqrt{\frac{L_{f1} + L_{f2}}{L_{f1} L_{f2} C_f}}
$$
Using SVPWM modulation, the output voltage of the inverter can be expressed as a Fourier series. The grid-side current magnitude at the switching harmonic angular frequency \(\omega_k\) is:
$$
|i_{2a}(j\omega_k)| = \frac{|u_{AN}(j\omega_k)|}{\omega_k |(L_{f1} + L_{f2}) – L_{f1} L_{f2} C_f \omega_k^2|}
$$
Table 3 summarizes the LCL filter parameters designed for the proposed multi-functional inverter.
| Parameter | Symbol | Value |
|---|---|---|
| Inverter-Side Inductor | \(L_{f1}\) | 3 mH |
| Grid-Side Inductor | \(L_{f2}\) | 0.8 mH |
| Filter Capacitor | \(C_f\) | 2.2 μF |
| LCL Resonance Frequency | \(f_r\) | 4.27 kHz |
| Target Damping Frequency Band | — | 1–2 kHz |
| Inverter Switching Frequency | \(f_s\) | 20 kHz |
Control Strategy Design
Overall Control Architecture
The control strategy for the multi-functional PV grid-connected inverter is shown in the block diagram. The PCC voltage signals \(u_{pcca}\), \(u_{pccb}\), and \(u_{pccc}\) are sampled and processed through a PLL to obtain the grid voltage phase angle \(\theta\) and the \(\alpha\beta\)-frame PCC voltage signals \(u_{pcc\_\alpha}\) and \(u_{pcc\_\beta}\). A harmonic detection unit extracts the resonant voltage components \(u_{pcch\_\alpha}\) and \(u_{pcch\_\beta}\), which are divided by the virtual resistance \(R_V\) and passed through a virtual impedance compensation function \(G_{ca}(s)\) to generate the harmonic reference current signals \(i_{h\_\alpha}^*\) and \(i_{h\_\beta}^*\). The power reference currents \(i_{md}^*\) and \(i_{mq}^*\) are transformed to the \(\alpha\beta\)-frame to obtain \(i_{m\alpha}^*\) and \(i_{m\beta}^*\).
Through the current regulator \(G_i(s)\), the inverter-side inductor currents \(i_\alpha\) and \(i_\beta\) track the reference currents. The filter capacitor current signals, processed by the active damping function \(H_{i1}(s)\), are subtracted from the output of \(G_i(s)\) to obtain the modulation signals \(u_\alpha^*\) and \(u_\beta^*\). These signals, together with the shoot-through signal \(D_0\) and the energy storage capacitor voltage \(u_{C1}\), are processed by the PWM module to generate the gate drive signals for the switches.
The harmonic detection unit uses notch filters with the transfer function:
$$
G_{HD}(s) = \prod_{h=1,5,7} \frac{s^2 + (h\omega_0)^2}{s^2 + s h\omega_0/Q + (h\omega_0)^2}
$$
where \(Q = 0.3\) is the quality factor. The current regulator \(G_i(s)\) employs a proportional-resonant (PR) controller:
$$
G_i(s) = k_p + \frac{2k_r \omega_c s}{s^2 + 2\omega_c s + \omega_0^2}
$$
where \(k_p\) and \(k_r\) are the proportional and resonant coefficients, and \(\omega_c = \pi\) rad/s is the -3 dB cutoff angular frequency of the resonant part.
The capacitor current feedback active damping function \(H_{i1}(s)\) is designed with phase lag compensation:
$$
H_{i1}(s) = K_{il} \frac{1 + bT_c s}{1 + T_c s}
$$
where \(K_{il}\) is the active damping coefficient. The negative value of \(K_{il}\) and the phase lag compensation improve the stability and output impedance phase margin of the inverter. The parameters \(T_c\) and \(b\) determine the maximum phase compensation \(\delta_m\) and its corresponding frequency \(f_m\):
$$
\delta_m = \arcsin\frac{1-b}{1+b}, \quad f_m = \frac{1}{2\pi T_c \sqrt{b}}
$$
Among different types of solar inverters, the control strategy employed here is specifically tailored for the energy-stored quasi-Z-source topology, leveraging its unique characteristics to achieve both power conversion and active damping simultaneously.
Current Loop Modeling
The mathematical model of the multi-functional inverter current loop in the stationary \(\alpha\beta\) frame is established. Considering the effects of the PLL, the reference current in the s-domain can be expressed as:
$$
\hat{i}_{1ref\_\alpha}(s) = (i_{md}^* + j i_{mq}^*) G_{pll}(s) \hat{u}_{pcc\_\alpha}(s) – (i_{md}^* – j i_{mq}^*) G_{pll}(s + j2\omega_0) \hat{u}_{pcc\_\alpha}(s + j2\omega_0)
$$
$$
\hat{i}_{1ref\_\beta}(s) = (i_{md}^* + j i_{mq}^*) G_{pll}(s) \hat{u}_{pcc\_\beta}(s) + (i_{md}^* – j i_{mq}^*) G_{pll}(s + j2\omega_0) \hat{u}_{pcc\_\beta}(s + j2\omega_0)
$$
The PLL transfer function is:
$$
G_{pll}(s) = \frac{1}{2} \cdot \frac{k_{p\_pll}(s – j\omega_0) + k_{i\_pll}}{(s – j\omega_0)^2 + [k_{p\_pll}(s – j\omega_0) + k_{i\_pll}] U_m} e^{-j\phi}
$$
The port current \(i_{2\_\alpha\beta}(s)\) is expressed as:
$$
i_{2\_\alpha\beta}(s) = i_{s\_\alpha\beta}(s) – \frac{u_{pcc\_\alpha\beta}(s)}{Z_{p1}(s)} – \frac{u_{pcc\_\alpha\beta}(s)}{Z_v(s)}
$$
where \(Z_{p1}(s)\) is the original port impedance without PLL effects, and \(Z_v(s)\) is the virtual impedance. The equivalent current source \(i_{s\_\alpha\beta}(s)\) and the impedances are given by:
$$
i_{s\_\alpha\beta}(s) = \frac{T_d(s)}{H_{i2}[1 + T_d(s)]} i_{1ref\_\alpha\beta}(s)
$$
$$
Z_{p1}(s) = \frac{1 + T_d(s)}{G_{x2}(s)}
$$
$$
Z_v(s) = \frac{1 + T_d(s)}{T_d(s)} \cdot \frac{R_V}{G_{HD}(s) G_{ca}(s)}
$$
The open-loop transfer function of the current loop is:
$$
T_i(s) = \frac{H_{i2} G_i(s) k_{pwm} e^{-1.5sT_s} (1 + s^2 C_f L_{f2})}{s L_{f1} L_{f2} C_f \left(s^2 + s \frac{H_{i1}(s) k_{pwm} e^{-1.5sT_s}}{L_{f1}} + \omega_r^2\right)}
$$
Control Parameter Design
Current Regulator and Active Damping Design
The design of the current regulator parameters and \(H_{i1}(s)\) must balance system stability under stiff grid conditions and the output impedance phase requirement for impedance-based stability criteria under weak grid conditions. The current loop cutoff frequency must be less than the LCL filter resonance frequency, and the bandwidth should exceed the target resonance frequency band for effective damping.
Under stiff grid conditions, the current loop open-loop transfer function simplifies to:
$$
T_i(s) \approx \frac{H_{i2} G_i(s) k_{pwm} e^{-1.5sT_s}}{s(L_{f1} + L_{f2})}
$$
At the cutoff frequency \(f_c = 2\) kHz, the resonant term of \(G_i(s)\) is negligible, leading to \(k_p = 47\). The fundamental frequency gain requirement \(T_{ifo} \geq 60\) dB yields \(k_r \geq 1146.8\), and we choose \(k_r = 1900\).
The active damping coefficient \(K_{il}\) must be carefully selected. When \(K_{il} < 0\), the stability condition derived from the Routh-Hurwitz criterion applied to the \(\omega\)-domain characteristic equation yields the permissible range:
$$
K_{il\_a0} < K_{il} < 0
$$
where \(K_{il\_a0} = -119.9\) for the designed parameters. Table 4 summarizes the stability conditions for different values of \(K_{il}\).
| Condition | Range of \(K_{il}\) | Stability Status |
|---|---|---|
| \(K_{il} > 0\) | \(0 < K_{il} < K_{il\_b1a}\) | Two right-half-plane poles exist; requires Nyquist verification |
| \(K_{il} < 0\) | \(K_{il\_a0} < K_{il} < 0\) | No right-half-plane poles; stable with proper margins |
| Selected Value | \(K_{il} = -12\) | Phase margin > 0°, system stable |
Table 5 provides the control parameters selected for the multi-functional inverter.
| Parameter | Symbol | Value |
|---|---|---|
| Current Regulator Proportional Gain | \(k_p\) | 47 |
| Current Regulator Resonant Gain | \(k_r\) | 1900 |
| Resonant Cutoff Angular Frequency | \(\omega_c\) | \(\pi\) rad/s |
| Active Damping Coefficient | \(K_{il}\) | -12 |
| Phase Lag Compensation Parameter | \(b\) | 0.589 |
| Phase Lag Compensation Time Constant | \(T_c\) | \(1.5 \times 10^{-5}\) s |
| Maximum Phase Compensation | \(\delta_m\) | 15° |
| Maximum Compensation Frequency | \(f_m\) | 13.8 kHz |
| PLL Proportional Gain | \(k_{p\_pll}\) | 4.5 |
| PLL Integral Gain | \(k_{i\_pll}\) | 18 |
| PLL Bandwidth | — | 223 Hz |
The phase lag compensation in \(H_{i1}(s)\) significantly affects the output impedance phase margin. Without compensation (\(\delta_m = 0^\circ\)), the output impedance phase margin is negative for all tested values of \(K_{il}\). With \(\delta_m = 15^\circ\) and \(f_m = 10\) kHz, the phase margin becomes positive for all three values of \(K_{il}\) tested. The PLL affects the low-frequency magnitude and phase of the output impedance, reducing the phase margin. These effects are more pronounced at lower power levels.
Virtual Impedance Parameter Design
For the multi-inverter parallel system, impedance-based stability criteria require that the total equivalent output impedance of the parallel inverters has a phase margin greater than 0° within the target resonance frequency band. The compensation function \(G_{ca}(s)\) is designed to ensure that the virtual impedance \(Z_v(s)\) exhibits purely resistive behavior in the target band:
$$
Z_v(s) = R_V \cdot \frac{1 + T_d(s)}{T_d(s)} \cdot \frac{1}{G_{HD}(s)}
$$
Within the 1-2 kHz target band, the notch filter effect is negligible (\(G_{HD}(s) \approx 1\)), and the current regulator approximates \(G_i(s) \approx k_p\). The compensation function simplifies to:
$$
G_{ca}(s) \approx 1 + \frac{s(L_{f1} + L_{f2})}{H_{i2} k_{pwm} k_p} (1.5 T_s s + 1)
$$
To avoid noise sensitivity from the derivative term, a non-ideal generalized integrator (GI) is used:
$$
G_I(s) = \frac{\omega_m^2 s}{s^2 + \omega_c s + \omega_m^2}
$$
where \(\omega_c = 8000\) rad/s and \(\omega_m = \pi f_s\). The digital implementation uses the Tustin discretization method. Table 6 compares the phase performance of the virtual impedance with and without compensation.
| Frequency (kHz) | Phase Deviation without Compensation (°) | Phase Deviation with Tustin Compensation (°) |
|---|---|---|
| 0.1 | 2 | 0.5 |
| 0.5 | 12 | 2 |
| 1.0 | 30 | 4 |
| 1.5 | 51 | 11 |
| 2.0 | 72 | 18 |
The impedance analysis of the multi-inverter system, considering the main inverter and the multi-functional inverter, shows that the PLL influences the low-frequency output impedance magnitude and phase. The addition of the virtual impedance ensures that the total impedance of the parallel system has a phase margin greater than 0° within the target resonance frequency band, guaranteeing system stability over a wide range of grid impedance variations.
Experimental Validation
To validate the feasibility and effectiveness of the proposed multi-functional PV grid-connected inverter, we constructed a multi-inverter parallel system experimental prototype. The main inverter and the multi-functional inverter each consist of an energy-stored quasi-Z-source inverter with its own DSP-based control circuit. The PV source is emulated using a DC power supply, and the battery is emulated using a DC supply with series and parallel resistors. The grid is simulated by connecting through a variable transformer to the main grid, with grid impedance emulated using magnetic powder core inductors.
For safety considerations, the experimental parameters are scaled to low voltage and low power levels. Table 7 summarizes the experimental parameters.
| Parameter | Main Inverter | Multi-Functional Inverter |
|---|---|---|
| PV Maximum Power Point Voltage | 120 V | 120 V |
| PV Maximum Power | 550 W | 550 W |
| Battery Voltage / Output Power | 150 V / 750 W | 150 V / 750 W |
| Grid Phase Voltage (RMS) | 55 V | 55 V |
| Switching Frequency | 10 kHz | 20 kHz |
| d-axis / q-axis Reference Current | 6.4 A / 0 A | 6.4 A / 0 A |
| Current Regulator Proportional Gain | 8.5 | 44 |
| Current Regulator Resonant Gain | 2.5 | 200 |
| Active Damping Coefficient | -18 | -25 |
| PLL Proportional Gain | 2.0 | 4.5 |
| PLL Integral Gain | 40 | 18 |
| Phase Lag Compensation Parameter \(b\) | 0 | 0.84 |
| Phase Lag Compensation Time Constant \(T_c\) | 0 | \(1.7 \times 10^{-5}\) s |
Experimental results under stiff grid conditions demonstrate stable operation with good dynamic performance during power step changes from 375 W to 750 W for both the main inverter and the multi-functional inverter. The three-phase grid currents and PCC voltage waveforms confirm stable operation. Steady-state measurements of PV source and battery voltages and currents verify that the battery discharges to balance the power.
Under weak grid conditions, the main inverter exhibits resonance while the multi-functional inverter maintains stable operation, consistent with theoretical analysis. This demonstrates the inherent stability advantage of the proposed multi-functional inverter design.
The critical validation comes from the parallel operation of both inverters under weak grid conditions. Before enabling the active damper function, the parallel system exhibits significant resonance in the three-phase currents and PCC voltage. After enabling the active damper function, the resonance is effectively suppressed, and the system returns to stable operation. Table 8 summarizes the experimental results under different operating conditions.
| Operating Condition | System Configuration | Active Damper | System Status | THD of Grid Current |
|---|---|---|---|---|
| Stiff Grid (\(L_g = 0.12\) mH) | Single Main Inverter | N/A | Stable | 2.1% |
| Stiff Grid (\(L_g = 0.12\) mH) | Single Multi-Functional Inverter | N/A | Stable | 1.8% |
| Weak Grid (\(L_g = 2.3\) mH) | Single Main Inverter | N/A | Resonance (Unstable) | 12.5% |
| Weak Grid (\(L_g = 2.3\) mH) | Single Multi-Functional Inverter | N/A | Stable | 2.2% |
| Weak Grid (\(L_g = 2.3\) mH) | Parallel System | Disabled | Resonance (Unstable) | 15.3% |
| Weak Grid (\(L_g = 2.3\) mH) | Parallel System | Enabled | Stable | 2.5% |
The experimental results conclusively demonstrate that the proposed multi-functional PV grid-connected inverter effectively integrates active damper functionality with grid-connected power generation, achieving stable operation under weak grid conditions where conventional inverters would experience resonance. The seamless transition between power generation and resonance suppression modes, without requiring changes to switching frequency or control loop parameters, represents a significant advantage over existing approaches.
This work contributes to the advancement of types of solar inverters by demonstrating a practical implementation that combines multiple functions in a single power conversion stage. The energy-stored quasi-Z-source topology proves to be an excellent platform for this integration, leveraging its inherent impedance network to buffer reactive power and its shoot-through capability to enhance reliability. The design methodology presented here can be extended to other types of solar inverters seeking to incorporate active damping functionality.
Conclusion
We have proposed and validated a multi-functional PV grid-connected inverter based on an energy-stored quasi-Z-source inverter that simultaneously achieves grid-connected power generation and resonance suppression. The key contributions of this work are as follows:
1. A comprehensive circuit parameter design methodology for the multi-functional grid-connected inverter was developed, covering the impedance network, LCL filter, and DC bus capacitor design. The design constraints ensure proper operation under both normal generation and active damping modes.
2. A control strategy based on inverter-side current feedback was developed, incorporating phase lag compensation in the capacitor current feedback loop to increase the phase margin of the original port impedance. This approach enhances system stability without compromising dynamic performance.
3. A virtual impedance construction method was implemented at the output of the multi-functional inverter to achieve resonance suppression in the target frequency band. The use of a non-ideal generalized integrator with Tustin discretization ensures accurate resistive behavior in the 1-2 kHz damping band.
4. The influence of the phase-locked loop on system stability was explicitly considered in the analysis. The PLL introduces a negative impedance that affects the low-frequency characteristics of the output impedance, reducing the phase margin. The proposed compensation strategy mitigates this effect.
5. Experimental validation confirmed the effectiveness of the proposed approach. The multi-functional inverter maintains stable operation under weak grid conditions where conventional inverters experience resonance. The active damper function effectively suppresses resonance in the parallel inverter system, reducing the grid current THD from 15.3% to 2.5%.
The proposed multi-functional inverter represents a significant advancement in types of solar inverters, demonstrating how power conversion and system stabilization functions can be integrated into a single, cost-effective solution. This approach is particularly valuable for distributed generation systems connected to weak grids, where resonance stability is a critical concern. Future work will explore the application of this concept to other types of solar inverters and the extension to higher power levels and different grid configurations.