3. Results and Discussion
This section presents the outcomes of dynamic simulations performed on a typical Malaysian distribution network with an integrated PV system, focusing on how different mitigation approaches influence system performance. The analysis is organized into three main areas: the initial phase investigates how varying levels of PV penetration affect voltage stability when no corrective measures are in place, revealing critical points in the network that are prone to voltage violations. The second part explores the individual impact of two mitigation techniques, BESS and TOU programs, evaluating how each strategy independently helps in managing voltage fluctuations. The final phase examines a hybrid approach that combines BESS and TOU, aiming to enhance grid resilience through coordinated storage and load-shifting mechanisms. The active and reactive load distribution across the network is illustrated in
Figure 8, where Feeder 8 stands out with the highest load demand, registering 714 kW of active and 357 kVAR of reactive power. This feeder is also the most geographically distant, located about 25 km from the substation, which naturally contributes to elevated line losses and a higher risk of voltage instability, conditions frequently encountered in rural electrical systems. These results underscore the importance of tailoring mitigation strategies to the unique layout and operational characteristics of rural networks. Long feeder spans, limited infrastructure, and uneven load distribution require not just technical upgrades but also demand-side participation to ensure reliable and efficient power delivery. The integration of BESS and TOU, especially in remote areas, offers a promising solution to support voltage regulation, minimize losses, and enhance overall grid performance under high renewable energy penetration.
The primary objective of this study is to manage and stabilize voltage levels when PV systems are connected to the distribution grid.
Figure 9 illustrates the system’s voltage behavior prior to the implementation of any mitigation measures. The modeled network comprises eight feeders, each distinguished by a separate color for clarity in analysis. Notably, Feeder 8 consistently exhibits the most severe voltage irregularities, making it a critical point of concern. These voltage deviations appear to be linked to the feeder’s distance from the substation as well as its load concentration. Interestingly, Feeder 6—despite being the longest in the system—shows stable voltage performance. This can be attributed to its relatively lower number of connected loads, which reduces cumulative voltage drops along the line. The figure also illustrates how voltage levels decrease with increased line length and electrical loading—a condition especially relevant in rural distribution networks, where feeder spans are typically long and serve scattered demand points.
Integrating PV into rural power distribution networks presents a promising approach to improving voltage stability, especially in areas with long feeders and limited grid reinforcement. However, studies examining widespread PV deployment across rural nodes remain limited. In this investigation, each node in the rural test network was assigned a 100 kW PV system unit, and simulations were conducted to determine the impact of various penetration levels under both light and heavy loading conditions. The aim was to identify the most effective PV sizing strategy to support voltage regulation across rural grid segments.
Figure 10 illustrates the voltage profiles for full (100 kW) and partial (20 kW) PV deployment scenarios. The results reveal that during daytime hours, voltage levels peak around 1.02 p.u., benefiting from abundant solar output. However, this excess generation can become problematic if not properly managed, especially when local demand is low. At night—between approximately 8:30 p.m. and 3:00 a.m.—the system exhibits voltage dips, with some nodes experiencing values below acceptable thresholds. This effect is more pronounced in rural grids, where longer feeder distances increase resistance and the risk of voltage drops, particularly in the absence of evening generation. To address these issues, a key recommendation is to shift daytime solar surplus to align with peak demand periods, ensuring that energy is available when it is most needed. This may involve integrating energy storage systems or deploying demand-side strategies to reshape rural consumption patterns. This study highlights that for rural networks, optimal PV deployment must be coupled with intelligent energy management to ensure voltage compliance and improve overall grid performance. Supporting this integration with policies and technologies tailored to rural conditions is essential for long-term sustainability.
Figure 11 provides a comprehensive heat map of the voltage and loading profile across all nodes in the rural distribution network prior to the integration of PV systems or any mitigation strategies. Each feeder (labeled 1 through 8) is represented with color gradients to indicate both voltage magnitude and loading stress. The color scale at the bottom ranges from red (0.9 p.u., indicating critically low voltage) to yellow and green (0.95–1.0 p.u., nominal voltage range) and up to blue (1.05 p.u., potentially over-voltage). Orange and red overlays highlight regions of high loading, with 100% loading depicted in deep orange/red and 80% loading in lighter orange tones. A closer examination reveals that Feeders 3 and 8 are the most vulnerable segments within the network. Large areas within these feeders, especially at their farthest nodes from the substation, are shaded in orange and red, demonstrating both high loading (close to or exceeding 100%) and significant voltage drops—sometimes falling below the 0.95 p.u. threshold. This dual stress can be attributed to two main factors: (1) long feeder lengths, which inherently cause greater voltage drops due to higher impedance, and (2) higher aggregated load at the extremities, which exacerbates voltage depression. Such conditions increase the risk of undervoltage events and could lead to customer supply interruptions or equipment malfunction if left unaddressed. Feeders 1 and 2 also display moderate voltage drops and higher loading, predominantly at their terminal nodes. This reflects typical challenges in rural grids, where the combination of extended feeder lines and dispersed, but significant, loads pushes voltage below optimal levels, especially during periods of peak demand. In contrast, Feeders 4, 5, and 6 maintain relatively stable operation, indicated by consistent green coloration throughout their lengths. This suggests that these feeders benefit from shorter distances, lower aggregated loads, or a more favorable balance between load and supply, resulting in voltage profiles that remain close to the nominal value (1.0 p.u.) and within safe operating margins.
One of the initial strategies explored in this study was the use of the DR program combined with PV system integration to alleviate voltage issues, particularly during evening hours when generation drops. This method involved adjusting consumer demand patterns using TOU pricing, with
Figure 12a illustrating the voltage response of the network under a 20% TOU implementation scenario. In rural power systems where infrastructure is limited and grid support is often weaker, DR programs have strong potential but also face unique challenges. The analysis reveals that TOU alone is insufficient to eliminate voltage violations, especially under high PV penetration. Even with a 20% participation rate, some nodes continue to experience undervoltage, particularly at night when solar output ceases and loads remain active. Interestingly, during daylight hours, voltage levels can spike, reaching as high as 1.02 p.u. due to excess generation not being matched by consumption. These fluctuations highlight the need for a more balanced approach to managing both surplus and deficit periods in rural distribution systems. Given that raising TOU participation beyond a certain point may not be practical or effective, especially in rural areas with limited consumer engagement or technological readiness, it becomes crucial to promote DR adoption more effectively. Outreach programs, incentives, and education can encourage rural consumers to participate in DR schemes. At the same time, combining DR with battery energy storage systems can provide the additional support needed to stabilize voltage and ensure consistent power quality. This underscores the need for hybrid mitigation strategies and greater policy focus on rural energy behavior, where the success of grid stability measures depends not only on technology but also on community participation.
The effect of BESS on voltage stability in rural networks is illustrated in
Figure 13a,b. These figures present the voltage profiles for all 110 nodes in the network over a 24 h period under two different PV penetration scenarios: (a) 100% and (b) 20%, with each node equipped with a 0.4 MWh BESS. As shown in
Figure 13a, at 100% PV penetration, the BESS units significantly help maintain node voltages within the acceptable range (between the red dashed lines indicating 0.95 and 1.05 p.u.). During the daytime, abundant solar generation allows the BESS units to charge, and as solar output declines in the evening, the stored energy is released to support the network. This results in relatively stable voltage profiles, with most nodes remaining above the minimum threshold (0.95 p.u.) throughout the 24 h period. Only a few nodes—typically at the end of long feeders—approach the lower limit after midnight, but widespread voltage violations are avoided. In contrast,
Figure 13b illustrates the situation at 20% PV penetration. Here, the available solar energy is much lower, which limits the amount of energy that can be stored in the BESS units during the day. As a result, after sunset, the BESS units are unable to sufficiently support the evening and nighttime load, leading to more nodes experiencing voltage drops below the 0.95 p.u. threshold, particularly after 10 p.m. This is especially pronounced in areas of the network with higher line impedance and longer feeders. Together, these results clearly show that while BESS can greatly enhance voltage stability at high levels of PV penetration, their effectiveness is constrained by both the total amount of solar energy available for storage and the storage capacity itself. At lower PV penetration levels, BESS alone may not be sufficient to prevent voltage violations during periods of peak demand or low solar generation. Therefore, as indicated by the comparison of
Figure 13a,b, an integrated approach—combining BESS with additional demand-side management strategies such as TOU load shifting—may be necessary to ensure reliable voltage regulation across the entire rural network, especially when solar generation is limited.
Conducting a detailed sensitivity analysis is essential when evaluating the performance of mitigation strategies under varying levels of PV integration—especially in rural power systems, where grid stability is inherently more vulnerable. Rural networks typically involve long feeders, uneven load distribution, and limited control infrastructure, making voltage regulation more complex. The analysis summarized in
Table 4 explores how different configurations of BESS and TOU participation influence voltage behavior across the system. The findings indicate that higher TOU engagement, particularly at 20%, significantly reduces voltage violations and also lowers the required BESS capacity. For instance, under full PV penetration (100%) with a 0.6 MWh BESS, only a small number of nodes still experienced undervoltage conditions. When just 10% TOU participation was added, many of these violations were resolved. Furthermore, the analysis reveals that even with a smaller BESS capacity (0.4 MWh), incorporating TOU at modest levels (up to 20%) under full PV penetration can shift a considerable portion of violations into the acceptable range. This is particularly beneficial for rural systems, where upgrading infrastructure can be costly and slow. Ultimately, this sensitivity assessment demonstrates the importance of tailoring mitigation solutions to grid-specific conditions. In rural environments, hybrid approaches—leveraging both demand-side and storage-based strategies—offer a practical pathway to maintaining voltage within regulatory limits, even under variable PV penetration scenarios.
Previous analysis indicated that relying on a single mitigation method is often insufficient to address voltage irregularities under varying levels of distributed generation penetration—particularly in rural electrical networks. These networks are typically characterized by long feeder distances and uneven load distribution, making them more susceptible to voltage instability. This section explores how combining two approaches—BESS and TOU demand management—can better support voltage control under these challenging conditions.
Figure 14 presents the voltage behavior of the network with both techniques applied at DG penetration levels of 100% and 20%. Results from
Table 4 and the visual data clearly show that a hybrid approach significantly improves voltage levels, especially with a 0.4 MWh BESS paired with a 20% TOU participation rate. In lower penetration scenarios, such as 20%, the system still shows minor violations, although improved, as illustrated in
Figure 14b. This improvement can be attributed to TOU’s ability to shift demand away from peak periods, smoothing out the load curve, and reducing stress on the grid. Meanwhile, BESS serves as a buffer, capturing excess energy when generation exceeds demand and releasing it when needed. However, in rural setups, planners must also consider factors such as weaker grid infrastructure, limited monitoring capabilities, and delayed system response times, which can further complicate mitigation efforts. Ultimately, this investigation highlights that a hybrid strategy is more suitable for rural power systems, offering improved voltage stability compared to standalone methods. Yet, successful deployment also requires addressing rural-specific limitations, including grid accessibility, economic feasibility, and consumer participation in smart energy programs.
In rural power distribution networks, maintaining voltage quality across extended feeder lines presents a significant challenge due to long distances and dispersed load centers. Unlike urban systems, where substations and loads are more compactly arranged, rural feeders must span wide geographical areas to supply power to scattered consumers. This investigation specifically examines how feeder length and load density impact voltage profiles, even after mitigation strategies have been applied.
Figure 15 illustrates voltage trends across several feeders within a rural grid configuration. Feeders positioned furthest from the substation—namely Feeder 6 and Feeder 8, located approximately 24 km and 20 km away—displayed the most noticeable voltage drops, particularly near their terminal nodes. Despite Feeder 6 being the longest, it maintained a better voltage profile than Feeder 8, largely due to having fewer connected loads (18 versus 21). This observation highlights that voltage deviation is not solely a function of distance, as well as how electrical loads are distributed along the feeder. In contrast, Feeder 5, which is only 4 km in length and serves a small cluster of six loads, experienced much smaller voltage variations, maintaining a minimum value of 0.987 p.u. compared to 0.978 p.u. for Feeder 6 and 0.958 p.u. for Feeder 8. This study is particularly valuable for rural grid planners and operators, as it demonstrates that even with advanced hybrid mitigation techniques—such as the integration of BESS and TOU pricing—voltage stability in rural networks remains highly sensitive to feeder length and node configuration. These insights are essential for designing more resilient rural electrical infrastructures, where grid reinforcement and targeted mitigation must be prioritized based on geographical and load characteristics.
To enhance the system’s ability to manage surplus solar generation, BESS units were strategically deployed at each node within the network. The control logic, implemented via DPL scripting (see
Appendix B), actively monitors the state of charge (SOC) of the batteries. When excess power is detected at a node, the control system assesses the SOC and initiates charging if capacity is available.
Figure 16 illustrates the cumulative active power of BESS over a 48 h period. At night, when solar output drops to zero, the BESS responds to demand by discharging. During daylight hours, PV generation first supplies the local load; only when a surplus remains and SOC is below the maximum does the BESS begin to charge. In the early hours (6:30 to 8:30 a.m.), all solar output is consumed by the load, leaving no surplus. However, between 8:30 a.m. and 1:00 p.m., available solar exceeds demand, enabling the batteries to gradually reach full charge. By midday, most BESS units reach maximum SOC. Later in the evening, as solar contribution declines (from 6:30 p.m. onward), batteries begin discharging again to meet demand, depending on load conditions. Importantly, when TOU pricing is incorporated, it causes a shift in load behavior, resulting in noticeable early morning activity—particularly around 8:00 a.m. This is reflected in the high level of grid power shown in the figure, indicating that TOU incentives have successfully reshaped demand patterns. The BESS discharges approximately 0.5 kWh by 8:00 a.m., demonstrating its effectiveness in supporting grid operations during shifted peak periods.
The power capacity and operational flexibility of the distribution grid are critical factors in determining its ability to accommodate additional generation from PV systems. An effective grid must be capable of absorbing the variable output from PV installations while maintaining voltage stability and overall reliability. In this study, grid performance was evaluated under three distinct scenarios: (1) a baseline condition without PV integration, (2) full PV penetration without mitigation strategies, and (3) full PV penetration with hybrid mitigation measures, including BESS and TOU pricing. As illustrated in
Figure 17, under the baseline scenario, grid power remained within acceptable operational limits. However, with 100% PV penetration, the system experienced increased voltage violations and power fluctuations, indicating significant stress on the network. The implementation of hybrid mitigation strategies led to a marked improvement in system performance. BESS played a key role by storing excess energy during periods of low demand and discharging it during peak demand, while TOU pricing effectively shifted consumption away from high-stress intervals. These findings underscore the importance of hybrid mitigation techniques in preserving power quality and voltage regulation in distribution grids with high levels of PV integration. Nevertheless, the effectiveness of such strategies is highly dependent on local grid characteristics, load profiles, and solar generation patterns. Therefore, tailored planning and system-specific optimization are essential to ensure the successful deployment of these solutions across diverse grid configurations.
The updated heat map presents the voltage and loading profile of the rural distribution network following the integration of PV systems and the implementation of hybrid mitigation strategies—specifically, BESS and TOU load shifting. Unlike the pre-mitigation scenario, this map shows a significant improvement in voltage regulation across all feeders, now clearly labeled from Feeder 1 through Feeder 8. Notably, areas that previously exhibited undervoltage or overloading—particularly in Feeders 1, 3, and 8—have shifted into a stable operating range, with most nodes now shaded in green, indicating voltage levels close to the nominal 1.0 per unit. Minor yellow hues in parts of Feeders 1 and 8 suggest localized areas of moderate loading, but all values remain within acceptable operational limits. This uniform voltage profile reflects the success of the hybrid mitigation strategy in reducing voltage fluctuations and alleviating peak demand pressure. The coordinated use of BESS to absorb excess daytime PV generation and discharge it during evening hours, combined with TOU-driven load shifting, has optimized energy distribution and improved system stability. As illustrated in
Figure 18, the application of the hybrid mitigation strategy (BESS + TOU) resulted in a well-balanced voltage profile across the network, effectively eliminating undervoltage violations and confirming the enhanced operational reliability of the rural grid under high PV penetration.
Figure 19 presents a comparative analysis of accumulated power losses across all feeders in the distribution network, highlighting the impact of the hybrid mitigation strategy—which integrates PV generation, BESS, and TOU load shifting. In the pre-mitigation scenario, total system losses peaked at over 320 kW, particularly during nighttime hours (00:00–06:00 and 20:00–24:00), when demand remained high and PV generation was unavailable. These losses were most pronounced in Feeder 8, which contributed more than 60% of the total loss due to its long distance from the substation, high concentration of load, and lack of localized voltage support. This pattern is typical of rural distribution systems, where feeders often extend over long distances to serve scattered loads, resulting in elevated resistance and voltage drops. The lack of voltage regulation infrastructure in such areas further exacerbates energy losses, especially during peak load periods when current flow through long feeder lines is highest. Following the implementation of the hybrid mitigation approach, a substantial reduction in power losses was observed across all feeders. Peak losses decreased to approximately 140–160 kW, representing a 50–55% improvement in system efficiency. This reduction is primarily attributed to the strategic deployment of BESS, which discharged stored energy during high-demand evening hours, reducing reliance on upstream supply and minimizing line losses. In parallel, the TOU load shifting strategy successfully redistributed consumption to mid-day periods when PV system output was at its peak, flattening the load curve and easing pressure on the grid. Feeder 8, which initially exhibited the most critical loss profile, showed the greatest improvement after mitigation—with its peak losses nearly halved. Other rural feeders, such as Feeders 2, 3, and 7, also recorded consistent reductions, although to a lesser extent, due to their moderate baseline loads. The distributed nature of PV systems allowed local generation to directly serve nearby loads, thus reducing the need for long-distance power transfer—a key benefit in rural areas where transmission distances inherently lead to inefficiencies.
Compared to recent approaches that employ advanced reinforcement learning for energy market optimization [
22] and adaptive management in constrained islanded systems [
23], this study presents a more practically deployable solution for rural medium-voltage distribution networks. While those works demonstrate strong performance in environments with access to high-level control infrastructure and centralized coordination, they may face implementation challenges in rural contexts where communication systems and real-time control capabilities are limited. In contrast, our hybrid mitigation strategy—combining BESS with TOU demand response—requires minimal infrastructure changes and leverages existing consumer behavior. By using a real-world Malaysian rural network model and focusing on system-wide voltage stability and power loss reduction, this work offers a robust, scalable alternative that aligns with the operational realities of rural grids.
4. Limitations
This study is based on detailed network models and typical utility data for Malaysian rural feeders; however, the simulation results were not benchmarked against measured feeder voltage profiles or actual utility billing records due to data availability constraints. Likewise, the performance of the modeled 100 kW PV systems was not directly validated with local irradiance and temperature measurements. In addition, the analysis assumes a uniform fixed-tilt PV installation of 100 kW (and alternative scenarios of 60 kW and 20 kW) at each node, using a single derating factor to account for general system losses. Site-specific factors such as available roof area, local shading, and inverter capacity limits were not explicitly modeled. To partially address these uncertainties, a ±20% sensitivity analysis on PV capacity was conducted, which indicated that the main conclusions regarding mitigation strategy effectiveness remain robust despite performance variations. Nevertheless, the results may still be subject to uncertainties due to differences between assumed and real-world operating conditions. Future research will focus on obtaining at least one week of measured feeder voltage and PV performance data, as well as incorporating more detailed modeling of physical installation constraints, to further validate and improve the simulation framework.
Equipping all 110 nodes with 0.4 MWh of BESS implies a total district storage capacity exceeding 44 MWh. Given current market prices for LFP battery systems, this scale of deployment would require tens of millions of dollars in capital expenditure (capex), presenting a significant economic challenge for most rural utilities. The uniform, fully distributed storage model used in this study was selected to explore the maximum technical benefit for voltage stability; however, in practice, economic constraints may favor more centralized or selectively deployed BESS solutions. Future research should include comprehensive techno-economic analyses to determine the optimal balance between technical performance and investment costs, as well as explore strategies such as shared community storage, phased deployment, or integration with other grid-support technologies to improve cost-effectiveness and practical feasibility.
Additionally, the DPL script used for BESS operation in this study employs a simple rule-based strategy: batteries are charged whenever surplus solar generation exists and discharged when local load exceeds PV output. This approach does not consider reserving SOC headroom for anticipated evening peak demands, nor does it account for the economic implications of battery cycling or degradation costs. As a result, the operational performance and cost-effectiveness of the BESS may be overestimated. Future studies should incorporate predictive or optimization-based control strategies and include battery lifetime and cycling costs for a more realistic assessment.
Moreover, this study assumes flat TOU participation rates of 10% and 20% without reference to actual market data, tariff values, or consumer price elasticity curves. The absence of detailed price signals and elasticity assumptions limits the accuracy of projected load-shifting behavior and its impact on grid performance. Future research should aim to incorporate real TOU tariff structures, estimate consumer responsiveness to price signals, and perform sensitivity analyses over a wider range of participation levels to more accurately model demand response effectiveness in rural networks.