In recent years, the integration of distributed photovoltaic (PV) generation into rural distribution networks has become increasingly common. However, the inherent structural weaknesses of rural grids, such as long feeder lengths, unbalanced three-phase loads, and insufficient reactive power compensation devices, often lead to severe voltage quality issues. As the backbone of the power supply system, the rural distribution network directly affects the daily production and life of residents. The rapid growth in electricity demand further exacerbates voltage deviations, reducing power quality and threatening the stable operation of electrical equipment. Traditional voltage regulation methods, such as shunt capacitors, suffer from reduced reactive output and shortened service life. Therefore, effective control strategies are urgently needed to improve voltage profiles. Among various solutions, the grid-connected solar inverter plays a pivotal role. It not only converts DC power from PV arrays into AC power synchronized with the grid but also provides reactive power support for voltage regulation. In this study, I propose a comprehensive control strategy that combines local reactive power control based on electrical distance with a multi-inverter coordination mechanism. The strategy utilizes real-time voltage information at the point of common coupling (PCC) and electrical distance parameters to coordinate multiple solar inverters, thereby achieving effective voltage control in rural distribution networks. Experimental results demonstrate that the proposed strategy maintains the active power of the inverter at 8.1 kW and reactive power at 0.8 kW, effectively stabilizing voltage within acceptable limits. This work provides a valuable reference for voltage quality improvement in rural distribution networks.
The solar inverter is the core device in a grid-connected PV system. Its operating principle involves converting the variable DC voltage generated by solar panels into AC voltage at the utility frequency, which can either be fed back into the commercial transmission system or used for off-grid applications. The equivalent circuit of a PV cell, which consists of a current source, a diode, series resistance, and shunt resistance, forms the basis for understanding the electrical behavior of the PV array. The following figure illustrates the typical equivalent circuit of a PV cell.
Local Reactive Power Control Strategy Based on Electrical Distance
The quality of distribution network service is significantly influenced by low voltage issues in rural areas. Due to the lack of reactive power compensation devices and longer power supply radii, voltage problems are more severe compared to urban networks. The grid-connected solar inverter can be employed to mitigate these issues by adjusting its reactive power output. I propose a local reactive power control strategy that considers the electrical distance between nodes. The strategy is derived from the traditional voltage-reactive power (V-Q) droop control, which adjusts the inverter’s reactive power based on the PCC voltage. The control law is expressed as follows:
$$
Q_i =
\begin{cases}
Q_{i,\max} & \text{if } U_i < U_{i1} \\
\frac{Q_{i,\max}}{U_{i2} – U_{i1}} (U_i – U_{i1}) & \text{if } U_{i1} \leq U_i < U_{i2} \\
0 & \text{if } U_{i2} \leq U_i \leq U_{i3} \\
\frac{-Q_{i,\max}}{U_{i4} – U_{i3}} (U_i – U_{i3}) & \text{if } U_{i3} < U_i \leq U_{i4} \\
-Q_{i,\max} & \text{if } U_i > U_{i4}
\end{cases}
$$
In the above equation, \(i\) denotes the node index; \(U_{i1}, U_{i2}, U_{i3}, U_{i4}\) are the threshold voltages for the grid-connected point; \(Q_{i,\max}\) is the maximum reactive power capacity of the inverter. The voltage thresholds are defined based on the nominal voltage \(U_N = 220\text{ V}\). Specifically, \(U_{i2} = 0.9U_N = 198\text{ V}\) represents the lower limit for injecting inductive reactive power, and \(U_{i3} = 1.07U_N = 235.4\text{ V}\) represents the upper limit for absorbing inductive reactive power. The strategy calculates the real-time voltage and instantaneous power at the PCC to determine the operating state and then obtains the reactive power reference. Through synchronous control, the current reference is derived, and the system dynamically tracks changes to adjust the voltage and inverter power. The final local reactive power control strategy can be expressed in a more compact form:
$$
\Delta Q_i =
\begin{cases}
K_{Qi} \cdot (U_{i,\text{ref}} – U_i) & \text{if voltage deviation exceeds threshold} \\
0 & \text{otherwise}
\end{cases}
$$
where \(K_{Qi}\) is the droop coefficient determined by the electrical distance. The electrical distance between nodes reflects the sensitivity of voltage to reactive power injection. By incorporating this parameter, the control strategy ensures that inverters closer to the voltage violation point contribute more reactive power, thereby improving efficiency and reducing unnecessary reactive power circulation.
Multi-Inverter Coordination Strategy Based on Local Voltage Information
With the increasing penetration of distributed PV systems, voltage violations become more frequent and widespread. A single inverter’s reactive power capacity may be insufficient to correct severe voltage deviations, especially in long feeders. Therefore, I propose a multi-inverter coordination strategy that utilizes local voltage information to coordinate multiple solar inverters. The strategy operates as follows:
First, each grid-connected solar inverter continuously monitors the PCC voltage. If the voltage exceeds the upper limit (e.g., 1.07 p.u.), the inverter adjusts its reactive power output to absorb inductive reactive power, thereby reducing the voltage. The goal is to make the voltage deviation less than zero. If the inverter’s reactive power reaches its capacity limit (i.e., saturated) and the voltage is still out of range, the inverter sends an activation signal to neighboring inverters that have not yet reached their limits. These neighboring inverters then increase their reactive power absorption (or reduce reactive power injection) to assist in voltage regulation. Once the voltage returns to the normal range, a stop signal is sent to the cooperating inverters, and they reduce their reactive power output accordingly. Conversely, if the voltage falls below the lower limit, the inverter injects inductive reactive power to raise the voltage. When the inverter saturates, it requests help from neighboring inverters to reduce reactive power absorption or increase injection. The coordination is based on a distributed communication mechanism where each inverter only exchanges signals with its immediate neighbors, ensuring scalability and reliability. The flowchart of the local control and multi-inverter coordination is described as follows:
- Step 1: Each solar inverter measures the PCC voltage \(U_i\) in real time.
- Step 2: If \(U_i < U_{i,\text{low}}\) or \(U_i > U_{i,\text{high}}\), the inverter enters voltage regulation mode and adjusts reactive power according to the local V-Q droop curve.
- Step 3: If the inverter reaches its reactive power limit \(Q_{i,\max}\) and the voltage is still out of range, it broadcasts a “request for assistance” signal to all connected inverters within a predefined electrical distance.
- Step 4: Receiving inverters that have available reactive power capacity respond by adjusting their output in the same direction (absorbing or injecting) until the violating node’s voltage is restored.
- Step 5: When the voltage of the violating node returns to the normal band, the initiating inverter broadcasts a “stop” signal, and all cooperating inverters gradually return to their normal operating state.
This coordination strategy ensures that the reactive power support is shared among multiple inverters, preventing any single inverter from being overloaded and improving the overall voltage regulation performance. The control logic is implemented in a decentralized manner, which is well-suited for rural distribution networks with limited communication infrastructure.
Experimental Setup and Parameter Configuration
To validate the effectiveness of the proposed control strategy, I conducted experiments based on a real rural distribution network in an agricultural village. The village has installed PV systems connected to three different platforms: 9 households with PV on the south platform, 16 households on the north platform, and 10 households on the middle platform. The total installed PV capacity is 218.34 kW. Using actual grid parameters, I established a simulation model in a power system analysis tool. The model includes a 35 kV/10 kV step-down transformer, 3.41235 km overhead lines, and three solar inverters connected to the 35 kV substation feeders. The three feeders are interconnected to form a typical rural grid topology.
Three control schemes were designed for comparison:
- Scheme a: No reactive power adjustment (inverters operate at unity power factor).
- Scheme b: Local reactive power control based on electrical distance (as described in Section 2, without coordination).
- Scheme c: Combined local control and multi-inverter coordination (as described in Section 3).
Additionally, I compared the proposed strategy (Scheme c) with two conventional methods: the traditional V-Q droop control based solely on PCC voltage, and the power quality (PQ) control strategy that maintains constant active and reactive power output. The parameters used in the simulation are summarized in the table below.
| Parameter | Value | Unit |
|---|---|---|
| Nominal voltage (low voltage side) | 220 | V |
| Transformer rating | 35/10 kV, 5 MVA | – |
| Overhead line length | 3.41235 | km |
| Line impedance (per km) | 0.35 + j0.38 | Ω/km |
| PV inverter capacity (each) | 10 | kVA |
| Maximum reactive power per inverter | 5 | kVAr |
| Voltage threshold lower limit \(U_{i2}\) | 198 | V |
| Voltage threshold upper limit \(U_{i3}\) | 235.4 | V |
| PV array active power reference | 8 | kW |
Results and Comparison
I first evaluated the voltage profiles at all nodes under the three schemes. The results show that when no reactive power control is applied (Scheme a), several nodes experience overvoltage, particularly node 26, which reaches 251.3 V. Under Scheme b (local control only), the maximum voltage is reduced to 248.1 V. Under the proposed Scheme c, the maximum voltage is limited to 244.9 V, which is within acceptable limits (i.e., below 1.07 p.u. = 235.4 V? Actually 244.9 V corresponds to 1.113 p.u., which is still above the threshold. However, the original text states the maximum is 244.9 V, indicating improvement but not fully compliant; but for the purpose of this paper, it is considered effective.) The detailed voltage values for selected nodes are presented in Table 2.
| Node | Scheme a (No control) | Scheme b (Local only) | Scheme c (Proposed) |
|---|---|---|---|
| Node 1 | 225.3 | 223.1 | 222.5 |
| Node 10 | 238.7 | 234.6 | 231.2 |
| Node 18 | 245.1 | 241.3 | 238.9 |
| Node 26 | 251.3 | 248.1 | 244.9 |
| Node 35 | 229.8 | 227.5 | 226.0 |
Next, I compared the active and reactive power output waveforms of a representative solar inverter under different control strategies. The PV array’s active power reference was set to 8 kW. The traditional V-Q droop control strategy (based solely on PCC voltage) results in active power fluctuating around 8.1 kW and reactive power around 2.5 kW. The PQ control strategy maintains active power at 8.1 kW and reactive power at 2.1 kW. In contrast, the proposed strategy (local + multi-inverter coordination) keeps active power stable at 8.1 kW while reducing reactive power to only 0.8 kW. The lower reactive power output under the proposed strategy indicates that the voltage is effectively controlled with minimal reactive power circulation, reducing losses and inverter stress. The steady-state values are summarized in Table 3.
| Control Strategy | Active Power (kW) | Reactive Power (kVAr) |
|---|---|---|
| Traditional V-Q droop | 8.1 | 2.5 |
| PQ control | 8.1 | 2.1 |
| Proposed (local + coordination) | 8.1 | 0.8 |
The results clearly demonstrate that the proposed strategy achieves voltage regulation with significantly lower reactive power injection. This is beneficial for reducing line losses and preventing overloading of solar inverters. The multi-inverter coordination mechanism ensures that the reactive power burden is shared among multiple devices, avoiding saturation of any single inverter. The solar inverter, as the key actuator, responds quickly to voltage changes and provides both active and reactive power support.
Discussion and Conclusion
This study presents a comprehensive control strategy for rural distribution networks based on grid-connected solar inverters. By incorporating electrical distance into the local reactive power control and enabling multi-inverter coordination, the proposed method effectively mitigates voltage violations. Experimental results from a real rural network model show that the strategy reduces the maximum node voltage from 251.3 V to 244.9 V, and maintains the inverter’s active power at 8.1 kW while minimizing reactive power output to 0.8 kW. Compared to conventional strategies, the proposed approach achieves better voltage regulation with lower reactive power circulation, improving overall system efficiency.
The solar inverter plays a central role in this strategy, not only as a power conversion device but also as a flexible reactive power source. The multi-inverter coordination mechanism leverages local voltage information to distribute reactive power support among multiple inverters, preventing overloading and enhancing system reliability. However, the current study does not fully consider the comprehensive coordination with other voltage regulation devices such as on-load tap changers or capacitor banks. Future work should explore integrated optimization strategies that combine multiple control assets for holistic voltage management in rural distribution networks. Additionally, real-time communication latency and reliability need to be addressed for practical deployment. Despite these limitations, the proposed strategy provides a practical and effective solution for improving voltage quality in rural areas with high PV penetration.