Search Results (78)

Reliable and efficient low-voltage distribution networks are critical for scaling mini-grid deployment and advancing universal electricity access, yet prevailing design practices remain manual, heuristic, and difficult to scale. This study presents a fully automated workflow that integrates geospatial feature extraction, distribution network layout, conductor sizing, mixed-integer linear programming-based phase balancing, nonlinear AC power flow validation, and system costing to generate rapid, standard-compliant techno-economic designs for greenfield mini-grid sites. The methodology is demonstrated across 62 rural sites to confirm practicality for large-scale rural electrification planning. Designs were evaluated for single-phase, three-phase, and hybrid low-voltage configurations. When design constraints were relaxed, single-phase networks achieved the lowest median voltage drop (~0.8%) and technical losses (~0.6%); however, under realistic voltage-drop and ampacity limits, compliance relied on conductor oversizing, resulting in low utilization (median loading <20%) and substantially higher costs. Fewer than half of the sites met construction feasibility limits for parallel conductors, and single-phase designs were typically 3–4× more expensive than multi-phase alternatives. Multi-phase layouts delivered comparable technical performance at significantly lower cost. Phase-balancing optimization reduced voltage drop by 15–20% and current unbalance by ~50%, enabling loss reduction and increased load accommodation. Overall, the results demonstrate that automated low-voltage network design can replace manual drafting with scalable, data-driven workflows that reduce soft costs while improving technical performance, constructability, and investment readiness. Full article

Mini-grids are increasingly deployed to expand rural electrification in developing countries, yet evidence on service-quality performance remains uneven. This systematic review synthesises empirical evidence from 22 peer-reviewed studies (2005–2025) on rural mini-grid performance across six energy-access indicators: electrification rate, availability of supply, hours of supply, affordability, reliability, and consistency (power quality). Using PRISMA-guided database searches in Scopus and Web of Science, 138 records were identified; following de-duplication and screening, 22 studies met the inclusion criteria. The evidence base is concentrated in Africa and Asia, and most studies adopt mixed-methods approaches combining household- and/or enterprise-level evidence with system or operational data. Across indicators, electrification outcomes are frequently positive but reported using heterogeneous metrics, often relying on connection counts rather than population-referenced rates (10/22 studies report electrification outcomes). Service availability and hours of supply vary widely, ranging from evening-only provision (~5 h/day) to near-continuous service (24 h/day), with several studies documenting demand–capacity mismatch and load shedding (9/22 quantify availability; 12/22 quantify hours). Affordability is most frequently reported (16/22 studies), spanning substantial household cost reductions in some settings to high tariffs that constrain uptake in remote contexts. Reliability is seldom quantified using extractable outage/downtime metrics (4/22 studies). No study reports standardised voltage/frequency power-quality measures; only proxy evidence relates to consistency, leaving power quality as a major evidence gap. Mini-grids can deliver meaningful improvements in rural electricity access, but the literature remains constrained by inconsistent indicator definitions, limited standardised reliability/power-quality measurement, and short monitoring horizons. Future research and regulation should prioritise harmonised service-quality metrics and longer-term, field-based performance evaluation. Full article

Low-voltage mini-grids play a crucial role in expanding electricity access for rural and remote communities. However, they continue to face technical and operational barriers that hinder their performance and reliability. This study reviewed the evolution of research on technical challenges in low-voltage mini-grids from 2005 to 2025. Using the PRISMA approach, data were extracted from the Scopus database, yielding 155 publications for bibliometric analysis. Bibliometrix in R Studio was used to examine publication trends, geographical contributions, and thematic evolution, while qualitative synthesis identified key engineering and operational constraints. The findings revealed a steady increase in research outputs since 2020, driven by global policy commitments, including Sustainable Development Goal 7 and the Paris Agreement. Persistent technical barriers include voltage and frequency instability, inadequate power quality monitoring, inefficient integration of energy storage, poor control coordination, and limited system design optimisation. African nations contribute less to global research despite being most affected by energy poverty, highlighting capacity and funding gaps. The study highlights the need for integrated solutions combining smart control, hybrid storage, and grid-interconnection technologies to enhance resilience and reliability. For policymakers and practitioners, the findings advocate for investment in research, capacity building, and locally tailored technical standards designed for resource-constrained contexts. This review provides a comprehensive evidence base to guide future research and policy directions aimed at achieving sustainable, technically robust, and financially viable mini-grid systems for universal energy access. Full article

Planning in robotics represents an ongoing research challenge, as it requires the integration of sensing, reasoning, and execution. Although large language models (LLMs) provide a high degree of flexibility in planning, they often introduce hallucinated goals and actions and consequently lack the formal reliability of deterministic methods. In this paper, we address this limitation by proposing a hybrid Sense–Plan–Code–Act (SPCA) framework that combines perception, LLM-based reasoning, and symbolic planning. Within the proposed approach, sensory information is first transformed into a symbolic description of the world in Planning Domain Definition Language (PDDL) using an LLM. A heuristic planner is then used to generate a valid plan, which is subsequently converted to code by a second LLM. The generated code is first validated syntactically through compilation and then semantically in simulation. When errors are detected, local corrections can be applied and the process is repeated as necessary. The proposed method is evaluated in the OpenAI Gym MiniGrid reinforcement learning environment and in a Gazebo simulation on a UR5 robotic arm using a curriculum of tasks with increasing complexity. The system successfully completes approximately 71–75% of tasks across environments with a relatively low number of simulation iterations. Full article

Universal electricity access remains elusive in Lesotho, with only a 53% connection rate. This statistic highlights a significant urban–rural gap of 60% to 18%, favouring urban areas mainly served by the main grid. The rugged terrain renders extending the grid to most rural areas impractical. To address this, the energy policy and electrification master plans aim to leverage abundant renewable energy resources and deploy mini-grids in rural regions. However, progress has been slow since the first advanced mini-grid projects began in 2018. The paper reviewed policy and framework documents from 2010 to 2025 that are pertinent to the deployment of mini-grids. It employed a hybrid qualitative-quantitative approach of SWOT-TOWS-AHP, which is rarely applied in energy policy analysis. It used the SWOT analysis tool to identify the Strengths, Weaknesses, Opportunities, and Threats faced in implementing sustainable renewable energy mini-grids. This was followed by the TOWS-AHP (Threats, Opportunities, Weaknesses, and Strengths-Analytical Hierarchy Process) method to develop strategies that utilize strengths and seize opportunities while tackling weaknesses and mitigating threats. These strategies were ranked based on their potential impact on mini-grid deployment. Despite supporting policies for mini-grids, the lack of political will from the government has emerged as a major obstacle. The three top strategies suggested to accelerate the deployment of sustainable mini-grids and advance efforts to achieve Sustainable Development Goal no. 7 by 2030 are establishing a mini-grid financing fund, reviewing the mini-grid regulatory framework, and reforming rural electrification institutions to improve coordination and collaboration. The top strategies carry weights of 8.5%, 7.8%, and 7.7%, respectively. Full article

This paper links the UN Sustainable Development Goal (SDG) of “Affordable and Clean Energy” (nr. 7) to “Partnerships” (nr. 17). These partnerships refer to stakeholders’ participation in renewable energy networks. Given that renewable energy is environmentally superior to fossil fuels and the participatory approaches foster well-being, this paper addresses economic sustainability. Therefore, the costs and benefits of electric power on the grid are compared to the distributed power networks in the EU, the USA, and India. Firstly, the present (dis)incentives for distributed energy networks are identified, concerning power generation, transmission, distribution, and consumption on the grid. Second, the costs of mini-grids and microgrids are assessed based on the existing literature. Thirdly, the benefits of such networks for individual and collective interests of producers and consumers of power are indicated. Although these partnerships are often as yet costly, incorporating those benefits into electricity prices enables price parity with the grid. Policies that pursue those benefits foster the realization of SDGs and improve the balance on the grid. Full article

State-of-charge (SoC) balancing across multiple battery energy storage systems (BESS) is a central challenge in renewable-rich mini-grids. Heterogeneous battery capacities, differing states of health, stochastic renewable generation, and variable loads create a high-dimensional uncertain control problem. Conventional droop-based SoC balancing strategies are decentralized and computationally light but fundamentally reactive and limited, whereas model predictive control (MPC) is insightful but computationally intensive and prone to modeling errors. This paper proposes a Hierarchical Predictive–Adaptive Control (HPAC) framework for SoC balancing in mini-grids using deep reinforcement learning. The framework consists of two synergistic layers operating on different time scales. A long-horizon Predictive Engine, implemented as a federated Transformer network, provides multi-horizon probabilistic forecasts of net load, enabling multiple mini-grids to collaboratively train a high-capacity model without sharing raw data. A fast-timescale Adaptive Controller, implemented as a Soft Actor-Critic (SAC) agent, uses these forecasts to make real-time charge/discharge decisions for each BESS unit. The forecasts are used both to augment the agent’s state representation and to dynamically shape a multi-objective reward function that balances SoC, economic performance, degradation-aware operation, and voltage stability. The paper formulates SoC balancing as a Markov decision process, details the SAC-based control architecture, and presents a comprehensive evaluation using a MATLAB-(R2025a)-based digital-twin simulation environment. A rigorous benchmarking study compares HPAC against fourteen representative controllers spanning rule-based, MPC, and various DRL paradigms. Sensitivity analysis on reward weight selection and ablation studies isolating the contributions of forecasting and dynamic reward shaping are conducted. Stress-test scenarios, including high-volatility net-load conditions and communication impairments, demonstrate the robustness of the approach. Results show that HPAC achieves near-minimal operating cost with essentially zero SoC variance and the lowest voltage variance among all compared controllers, while maintaining moderate energy throughput that implicitly preserves battery lifetime. Finally, the paper discusses a pathway from simulation to hardware-in-the-loop testing and a cloud-edge deployment architecture for practical, real-time deployment in real-world mini-grids. Full article

This paper proposes an analysis of different logics (heuristic and linear) of managing renewables scenarios including two different operating conditions and their relative degradation: fixed and variable point. The synergy between two storage technologies, such as Li-ion batteries and the hydrogen power-to-power solution (electrolyzer, H₂ tank, and fuel cells), is evaluated to ensure the balance of the power grid. This paper presents a numerical model of the smart grid developed in MATLAB/Simulink. A detailed performance evaluation of each component was performed to meet an electrical load (30 kW-peak) of a smart renewable energy community. From the optimization process, a fuel cell of 6 kW, an electrolyzer of 18 kW, a tank of 40 m³ at 200 bars, as well as a battery of 75 kWh were selected. The fuel cell operates during autumn and winter due to the lack of photovoltaic power generation, while its contribution is reduced during the summer period. In the heuristic logic, the minimum and maximum hydrogen levels are 18% and 60% of the tank volume (40 m³), respectively, while in the linear logic, they are 33% and 65%. The average value of the state of charge (SOC) of the battery is similar in both logics (0.51 vs. 0.53). Regarding hydrogen produced from the electrolyzer, the linear logic allows it to produce a quantity 7% higher than the heuristic one; therefore, the linear logic allows it to properly manage the electrochemical systems. The dynamic operation results in more significant degradation of hydrogen systems, making them less suitable; thus, to preserve the devices (up to 25% of lifetime more), a fixed-point operation is recommended. The cost comparison does not show relevant differences between the two scenarios, while a steep increase in the costs is shown when the fuel cell is operated in dynamic mode. Finally, the total emissions associated with renewable microgrids are 30 times lower than the traditional grid scenario, demonstrating the potential of renewable energy communities. Full article

Reinforcement learning (RL) agents face significant challenges in sparse-reward environments, as insufficient exploration of the state space can result in inefficient training or incomplete policy learning. To address this challenge, this work proposes a teacher–student framework for RL that leverages the inherent knowledge of large language models (LLMs) to decompose complex tasks into manageable subgoals. The capabilities of LLMs to comprehend problem structure and objectives, based on textual descriptions, can be harnessed to generate subgoals, similar to the guidance a human supervisor would provide. For this purpose, we introduce the following three subgoal types: positional, representation-based, and language-based. Moreover, we propose an LLM surrogate model to reduce computational overhead and demonstrate that the supervisor can be decoupled once the policy has been learned, further lowering computational costs. Under this framework, we evaluate the performance of three open-source LLMs (namely, Llama, DeepSeek, and Qwen). Furthermore, we assess our teacher–student framework on the MiniGrid benchmark—a collection of procedurally generated environments that demand generalization to previously unseen tasks. Experimental results indicate that our teacher–student framework facilitates more efficient learning and encourages enhanced exploration in complex tasks, resulting in faster training convergence and outperforming recent teacher–student methods designed for sparse-reward environments. Full article

Continual reinforcement learning (CRL) agents face significant challenges when encountering distributional shifts. This paper formalizes these shifts into two key scenarios, namely virtual drift (domain switches), where object semantics change (e.g., walls becoming lava), and concept drift (task switches), where the environment’s structure is reconfigured (e.g., moving from object navigation to a door key puzzle). This paper demonstrates that while conventional convolutional neural networks (CNNs) struggle to preserve relational knowledge during these transitions, graph convolutional networks (GCNs) can inherently mitigate catastrophic forgetting by encoding object interactions through explicit topological reasoning. A unified framework is proposed that integrates GCN-based state representation learning with a proximal policy optimization (PPO) agent. The GCN’s message-passing mechanism preserves invariant relational structures, which diminishes performance degradation during abrupt domain switches. Experiments conducted in procedurally generated MiniGrid environments show that the method significantly reduces catastrophic forgetting in domain switch scenarios. While showing comparable mean performance in task switch scenarios, our method demonstrates substantially lower performance variance (Levene’s test, $p<1.0\times {10}^{-10}$), indicating superior learning stability compared to CNN-based methods. By bridging graph representation learning with robust policy optimization in CRL, this research advances the stability of decision-making in dynamic environments and establishes GCNs as a principled alternative to CNNs for applications requiring stable, continual learning. Full article

Reinforcement learning is one of the most prominent research areas in the field of artificial intelligence, playing a crucial role in developing agents that autonomously make decisions in complex environments. This study proposes a method to optimize agent behavior in the MiniGrid-Empty-5x5-v0 environment using large language models (LLMs). By leveraging the natural language processing capabilities of LLMs to interpret environmental states and select appropriate actions, this research explores an approach that differs from traditional reinforcement learning methods. Experimental results confirm that LLM-based agents can effectively achieve their goals, and it is anticipated that maximizing the synergy between LLMs and reinforcement learning will contribute to the development of more intelligent and adaptable AI systems. Full article

Sub-Saharan Africa, especially its rural areas, faces significant challenges in achieving universal electrification despite its abundant renewable energy resources. The region has the highest population without access to electricity, largely due to economic, infrastructural, and geographical barriers. Energy poverty is a critical issue that hinders sustainable development and exacerbates inequalities. Namibia’s sustainable energy policy aligns with the global Sustainable Development Goals (SDGs), particularly SDG 7, which aims to provide affordable and reliable modern energy access for all. The policy emphasizes mini-grids and decentralized power systems as key strategies for rural electrification. However, despite increased deployment of mini-grids, these solutions often struggle with long-term sustainability. This research explores cost-effective electrification strategies through scenario-based modeling to reduce energy poverty and expand energy access in Namibia’s rural communities, focusing on the existing mini-grids in Tsumkwe and Gam. Using a comprehensive methodology that incorporates HOMER Pro for mini-grid capacity expansion and MS Excel for evaluating main-grid extensions, this study aims to identify the most feasible and economical electrification solutions. The analysis compares electricity supply, total net present cost, and the levelized cost of electricity across these systems. The findings will offer insights into addressing energy poverty in Namibia and provide recommendations for sustainable and scalable rural electrification across Sub-Saharan Africa. Full article

The energy transition toward Net Zero Emission by 2060 hinges on the renewable energy power plants in Indonesia. Good practices in several countries suggest a peer-to-peer (P2P) energy trading system using blockchain technology, supported by renewable energy (solar panels), an innovation to provide equal access to sustainable electricity while reducing the impact of climate change. The P2P energy trading concept has a higher social potential than the conventional electricity buying and selling approach, such as that of PLN (the state-owned electricity company in Indonesia), which applies the network management concept but does not have a sharing element. This model implements a solar-powered mini-grid system and produces a smart contract that facilitates electricity network users to buy, sell, and trade electricity in rural areas via smartphones. This study aims to measure the stakeholders’ perceptions of the peer-to-peer (P2P) energy trading model using blockchain technology in the Gumelar District, Banyumas Regency, Central Java Province, Indonesia. The stakeholders in question are representatives of Households (producers and consumers), Government, State Electricity Company (PLN), Non-Governmental Organizations, Private Sector and Academician. Measurement of perception in this study used a questionnaire approach with a Likert scale. The results of filling out the questionnaire were analyzed using four methods: IFE/EFE matrix; IE matrix; SWOT matrix; and SPACE matrix to assess the results and their suitability to each other. The results of the stakeholder perception assessment show that there are 44 internal factors and 33 external factors that can influence this model. We obtained an IFE and EFE score of 2.92 and 2.83 for the internal and external results using the IE matrix. These place the model in quadrant V, meaning the P2P model can survive in the long term to generate profits. Based on the SWOT analysis results, this model is located at the coordinate point −0.40, 0.31, placing it in quadrant II. This means that the P2P model is in a competitive situation and faces threats but still has internal strengths. Based on the SPACE matrix, stakeholder perception states that the P2P model is at coordinate point 1, −0.3. This shows that the P2P model has the potential to be a competitive advantage in its type of activity that continues to grow. In conclusion, our findings show that stakeholders’ perceptions of P2P models using blockchain technology can be implemented effectively and provide social, economic, and environmental incentives. Full article

In mini-grids and marine-isolated grids, power generation gas turbines are subjected to rapid start-up, shutdown, and acceleration/deceleration. This sudden load change can pose a significant impact on the power grid, severely affecting the operational characteristics of gas turbines. To understand the dynamic characteristics of the gas turbine in the transitional processes, this testing takes twin-shaft medium-sized power generation gas turbines as the test object, and goes through the process of startup, acceleration, deceleration, acceleration, shutdown in one hour, and repeats this process 40 times continuously. With fuel flow as the control parameter and power turbine outlet temperature and high-pressure turbine speed as the controlled parameters, the parameter response rate of the gas turbine under various transition processes is analyzed and the effect of thermal inertia on the gas turbine mass temperature as well as speed is studied. Research findings: During the transition processes, the gas temperature exhibited an axial gradient distribution in the channel. In both the acceleration and deceleration processes, the working fluid temperature gradually decreased along the flow direction. And thermal inertia posed different extents of impact on the dynamic characteristics of the gas turbine under different transitional processes. In the same transition process, the impacts of thermal inertia on the response speeds of temperature and rotational speed varied. The results of this study help to more accurately predict the operating state of the gas turbine during the transition process and lay the foundation for the dynamic simulation model of the non-adiabatic gas turbine. Full article

Renewable energy mini-grids are considered a cost-effective way to provide electricity for a large proportion of the population in developing countries who do not have access to it. Compared with standalone home systems and national grid systems, mini-grids can potentially offer a better service. They can be deployed faster, making them essential for sustainable development, especially in rural and semi-urban areas of developing countries. However, mini-grids often face challenges regarding their resilience, and many fail to survive beyond their pilot phases. This paper aims to identify the factors contributing to the success of mini-grids and to identify common themes that can help existing and future mini-grid developments become more resilient and influence policy decision making. To achieve this goal, we developed a database of the status of mini-grids in Malawi, with the energy generation resource(s) of their installed capacity, enabling factors, and challenges. We undertook a more detailed investigation of two hydro mini-grid systems—Bondo and Chipopoma. We collected qualitative and quantitative data through literature reviews, site visits, interviews, and observations. The study identified 19 mini-grids with a combined installed capacity of 26 MW. Of these, seven had been abandoned, and one was under development. Several factors that affect successful mini-grid efficacy in Malawi were identified, including financial resourcefulness, technical resourcefulness, policies and regulations, community engagement and capacity building, cross-sector linkages, and institutional organisational frameworks. These factors need to be integrated into decision making by all stakeholders to ensure the enhancement of resilience and the sustainable development of mini-grids. Full article