The effective integration of photovoltaic (PV) systems with battery storage is essential for advancing sustainable energy adoption yet translating forecasts into adaptive and interpretable control remains a key challenge. This study introduces a SHapley Additive exPlanations–Guided Energy Management System (SHAP-EMS) that directly embeds model interpretability into real-time control for residential solar-battery systems. A hybrid Linear Regression–eXtreme Gradient Boost (LR-XGBoost) model provides one-hour-ahead PV forecasts, while a SHAP-weighted rule-based controller dynamically adjusts decision priorities based on feature importance, system state, and temporal interactions. Results demonstrate that SHAP-EMS achieved an 18.3% reduction in peak grid imports (63.2% to 44.9%), a 4.5% decrease in total imports compared with Mixed-Integer Linear Programming optimization, and consistently high self-consumption ratios under polycrystalline PV conditions. By efficiently adapting to temperature fluctuations and generation variability, the framework illustrates how SHAP values can be leveraged to transform black-box forecasts into transparent, computationally efficient, and adaptive control strategies, establishing a novel paradigm for explainable energy management.
{"title":"SHapley Additive exPlanations-guided rule-based energy management: bridging machine learning interpretability and adaptive control strategies","authors":"Abdallah Abdellatif , Hamza Mubarak , Harikrishnan Ramiah , Hazlie Mokhlis , Saad Mekhilef , Hassan Muwafaq Gheni , Jeevan Kanesan","doi":"10.1016/j.ref.2025.100779","DOIUrl":"10.1016/j.ref.2025.100779","url":null,"abstract":"<div><div>The effective integration of photovoltaic (PV) systems with battery storage is essential for advancing sustainable energy adoption yet translating forecasts into adaptive and interpretable control remains a key challenge. This study introduces a SHapley Additive exPlanations–Guided Energy Management System (SHAP-EMS) that directly embeds model interpretability into real-time control for residential solar-battery systems. A hybrid Linear Regression–eXtreme Gradient Boost (LR-XGBoost) model provides one-hour-ahead PV forecasts, while a SHAP-weighted rule-based controller dynamically adjusts decision priorities based on feature importance, system state, and temporal interactions. Results demonstrate that SHAP-EMS achieved an 18.3% reduction in peak grid imports (63.2% to 44.9%), a 4.5% decrease in total imports compared with Mixed-Integer Linear Programming optimization, and consistently high self-consumption ratios under polycrystalline PV conditions. By efficiently adapting to temperature fluctuations and generation variability, the framework illustrates how SHAP values can be leveraged to transform black-box forecasts into transparent, computationally efficient, and adaptive control strategies, establishing a novel paradigm for explainable energy management.</div></div>","PeriodicalId":29780,"journal":{"name":"Renewable Energy Focus","volume":"56 ","pages":"Article 100779"},"PeriodicalIF":5.9,"publicationDate":"2025-10-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145473520","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-27DOI: 10.1016/j.ref.2025.100777
Ismallianto Isia , Bee Huah Lim , Noor Fifinatasha Shahedan , Masli Irwan Rosli , Wai Yin Wong
The global shift toward clean and sustainable energy systems has highlighted the importance of understanding socio-economic barriers to hydrogen technology acceptance, especially in emerging economies like Malaysia. This systematic review aims to assess the socio-economic vulnerability factors that influence the readiness and capacity of Malaysian communities to transition toward hydrogen technology such as fuel cell applications. A comprehensive literature search across databases was conducted, applying PRISMA guidelines to identify, screen, and analyze peer-reviewed studies published between 2010 and 2025. Out of 427 initially identified articles, 8 were selected for detailed evaluation. Findings reveal that financial constraints, infrastructure disparities, policy gaps, and educational inequalities disproportionately hinder vulnerable communities, exacerbating urban-rural divides and perpetuating energy inequities. Moreover, factors like property ownership, employment status, and digital literacy intersect with demographic and spatial variables, reinforcing systemic inequalities in technology access. This review reveals that conventional, additive vulnerability frameworks are insufficient to address the multifactorial nature of energy adoption challenges. Therefore, we advocate for the integration of context-specific, multidimensional indicators tailored to Malaysia’s socio-political and geographic heterogeneity to ensure an inclusive and equitable clean energy transition. The study concludes with strategic recommendations for policymakers to design data-driven interventions targeting vulnerable groups in the hydrogen energy ecosystem.
{"title":"Assessing socio-economic vulnerability factors for hydrogen technology adoption in Malaysia: a systematic review","authors":"Ismallianto Isia , Bee Huah Lim , Noor Fifinatasha Shahedan , Masli Irwan Rosli , Wai Yin Wong","doi":"10.1016/j.ref.2025.100777","DOIUrl":"10.1016/j.ref.2025.100777","url":null,"abstract":"<div><div>The global shift toward clean and sustainable energy systems has highlighted the importance of understanding socio-economic barriers to hydrogen technology acceptance, especially in emerging economies like Malaysia. This systematic review aims to assess the socio-economic vulnerability factors that influence the readiness and capacity of Malaysian communities to transition toward hydrogen technology such as fuel cell applications. A comprehensive literature search across databases was conducted, applying PRISMA guidelines to identify, screen, and analyze peer-reviewed studies published between 2010 and 2025. Out of 427 initially identified articles, 8 were selected for detailed evaluation. Findings reveal that financial constraints, infrastructure disparities, policy gaps, and educational inequalities disproportionately hinder vulnerable communities, exacerbating urban-rural divides and perpetuating energy inequities. Moreover, factors like property ownership, employment status, and digital literacy intersect with demographic and spatial variables, reinforcing systemic inequalities in technology access. This review reveals that conventional, additive vulnerability frameworks are insufficient to address the multifactorial nature of energy adoption challenges. Therefore, we advocate for the integration of context-specific, multidimensional indicators tailored to Malaysia’s socio-political and geographic heterogeneity to ensure an inclusive and equitable clean energy transition. The study concludes with strategic recommendations for policymakers to design data-driven interventions targeting vulnerable groups in the hydrogen energy ecosystem.</div></div>","PeriodicalId":29780,"journal":{"name":"Renewable Energy Focus","volume":"56 ","pages":"Article 100777"},"PeriodicalIF":5.9,"publicationDate":"2025-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145473517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Renewable energy plays a crucial role in achieving sustainable development goal 7 (SDG-7), which aims to ensure access to affordable, reliable, sustainable and modern energy for all. This study presents a bibliometric analysis of renewable energy research in ASEAN countries from 2000 to 2024, highlighting key trends, influential authors, leading research institutions and dominant sources of publication using Scopus database. VOSviewer and Bibliometrix are used to perform the analysis including the evolution of research topics, collaboration networks, the impact of scholarly contributions. The findings indicate a significant rise in publications in the region with Malaysia and Indonesia leading in total research output. However, the per capita productivity metric highlights Singapore and Brunei as the most research-productive nations in the region. Topics such as solar energy, biomass and wind energy emerge as the most studied topics reflecting regional priorities and resource availability. Moreover, the keyword analysis reveals a shift toward emerging technologies such as microgrids, energy storage and smart grids, signalling future research directions. The study underscores the growing emphasis on sustainability, energy efficiency and policy frameworks necessary for advancing renewable energy adoption in ASEAN. In terms of publications source, there was a diverse range of journals and conference proceedings that serve as primary dissemination platforms for ASEAN-based renewable energy researchers. By mapping past and present research landscapes, this study provides insights for policymakers, researchers and industry stakeholders to foster collaboration and drive the transition of the region towards a sustainable energy future in alignment with SDG-7.
{"title":"Renewable energy research in ASEAN countries: bibliometric analysis of past, present and future trends","authors":"Djamal Hissein Didane , Bukhari Manshoor , Mohammad Sukri Mustapa , Abdulrahman Aljabri , Abba Lawan Bukar , Mahmoud Kassas","doi":"10.1016/j.ref.2025.100776","DOIUrl":"10.1016/j.ref.2025.100776","url":null,"abstract":"<div><div>Renewable energy plays a crucial role in achieving sustainable development goal 7 (SDG-7), which aims to ensure access to affordable, reliable, sustainable and modern energy for all. This study presents a bibliometric analysis of renewable energy research in ASEAN countries from 2000 to 2024, highlighting key trends, influential authors, leading research institutions and dominant sources of publication using Scopus database. VOSviewer and Bibliometrix are used to perform the analysis including the evolution of research topics, collaboration networks, the impact of scholarly contributions. The findings indicate a significant rise in publications in the region with Malaysia and Indonesia leading in total research output. However, the per capita productivity metric highlights Singapore and Brunei as the most research-productive nations in the region. Topics such as solar energy, biomass and wind energy emerge as the most studied topics reflecting regional priorities and resource availability. Moreover, the keyword analysis reveals a shift toward emerging technologies such as microgrids, energy storage and smart grids, signalling future research directions. The study underscores the growing emphasis on sustainability, energy efficiency and policy frameworks necessary for advancing renewable energy adoption in ASEAN. In terms of publications source, there was a diverse range of journals and conference proceedings that serve as primary dissemination platforms for ASEAN-based renewable energy researchers. By mapping past and present research landscapes, this study provides insights for policymakers, researchers and industry stakeholders to foster collaboration and drive the transition of the region towards a sustainable energy future in alignment with SDG-7.</div></div>","PeriodicalId":29780,"journal":{"name":"Renewable Energy Focus","volume":"56 ","pages":"Article 100776"},"PeriodicalIF":5.9,"publicationDate":"2025-10-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145424515","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-26DOI: 10.1016/j.ref.2025.100769
Philip Stelling , Alan C Brent , Daniel Burmester
Aotearoa New Zealand aims to achieve 100% renewable electricity by 2030, currently standing at over 85% from hydro, geothermal, wind, and solar resources. The country’s isolated geography currently necessitates dispatchable hydropower and fossil fuels to manage intermittency and maintain grid stability. A literature review of countries also with high renewable penetrations – Norway, Iceland, Austria, Canada, and Brazil – revealed challenges including price volatility, operational flexibility requirements, dry year risks, and increasing electricity demand from economic growth and electrification. The objective of this paper is to understand the potential consequences for Aotearoa New Zealand by comparing the projected 2030 electricity demand, based on scenarios developed by the government, against anticipated renewable generation capacity, using data on the current generation fleet and the near-term investment pipeline. The method assumed that added capacity of renewables would follow similar generation profiles to existing generators. It is concluded that the 100% renewable electricity target by 2030 is feasible, but only if the committed and actively pursued projects, including offshore wind, are commissioned. Then there would be sufficient generation for all scenarios, maintaining nearly full hydro storage year-round. Minor shortfalls occur during low wind/solar periods (0 to 1% of the year), but with significant excess generation (55 to 65% of the year) where 27 to 42% would be available for effective storage utilisation in the power system. To this end, the shortfalls can be addressed, to some extent, with committed and actively pursued battery storage, which was not included in the analysis due to the uncertainty of how they will be participating in the future electricity market.
{"title":"The impact of 100% renewable electricity on hydropower generation in Aotearoa New Zealand","authors":"Philip Stelling , Alan C Brent , Daniel Burmester","doi":"10.1016/j.ref.2025.100769","DOIUrl":"10.1016/j.ref.2025.100769","url":null,"abstract":"<div><div>Aotearoa New Zealand aims to achieve 100% renewable electricity by 2030, currently standing at over 85% from hydro, geothermal, wind, and solar resources. The country’s isolated geography currently necessitates dispatchable hydropower and fossil fuels to manage intermittency and maintain grid stability. A literature review of countries also with high renewable penetrations – Norway, Iceland, Austria, Canada, and Brazil – revealed challenges including price volatility, operational flexibility requirements, dry year risks, and increasing electricity demand from economic growth and electrification. The objective of this paper is to understand the potential consequences for Aotearoa New Zealand by comparing the projected 2030 electricity demand, based on scenarios developed by the government, against anticipated renewable generation capacity, using data on the current generation fleet and the near-term investment pipeline. The method assumed that added capacity of renewables would follow similar generation profiles to existing generators. It is concluded that the 100% renewable electricity target by 2030 is feasible, but only if the committed and actively pursued projects, including offshore wind, are commissioned. Then there would be sufficient generation for all scenarios, maintaining nearly full hydro storage year-round. Minor shortfalls occur during low wind/solar periods (0 to 1% of the year), but with significant excess generation (55 to 65% of the year) where 27 to 42% would be available for effective storage utilisation in the power system. To this end, the shortfalls can be addressed, to some extent, with committed and actively pursued battery storage, which was not included in the analysis due to the uncertainty of how they will be participating in the future electricity market.</div></div>","PeriodicalId":29780,"journal":{"name":"Renewable Energy Focus","volume":"56 ","pages":"Article 100769"},"PeriodicalIF":5.9,"publicationDate":"2025-10-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145424516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-25DOI: 10.1016/j.ref.2025.100774
Vahid Parvaz
The increasing penetration of Electric Vehicles (EVs) and Photovoltaic (PV) systems has introduced new energy management and financial challenges in the electricity market. The inherent fluctuations in PV generation and the unpredictable behavior of the EV charging load carry a significant risk of imbalance and heavy financial penalties. This paper proposes a comprehensive risk management framework for the optimal energy procurement decision of a PV-EV facility in the electricity market, aiming to minimize operational costs. The methodology was developed along two parallel paths: (1) heterogeneity analysis of EV charging load using K-Means clustering and (2) PV generation uncertainty modeling.
A hybrid statistical-machine learning model was adopted for uncertainty estimation, validating the prediction interval reliability with a Prediction Interval Coverage Probability (PICP) of 93.72%. Finally, the optimal energy procurement decision was formulated using a Robust Optimization model to guarantee load supply under the 90% lower bound of PV generation (). The results show that adopting the robust decision, assuming day-ahead market prices (CDA=50 UP/kWh) and imbalance settlement (CRT=70 UP/kWh), leads to a total operational cost of 68.03 million UP over a one-year testing period. Compared to the ideal (deterministic) model, this cost includes a risk cost of 5,546,926.32 UP (equivalent to 7.53% of the deterministic model’s cost). This additional cost quantifies the economic value of uncertainty information and ensures the system’s 93.84% stability in meeting demand. The findings suggest that the proposed framework not only offers technical solutions but can also provide effective guidance for operators and policymakers in electricity markets.
{"title":"A robust framework for financial risk management of PV–EV systems under uncertainty in electricity markets","authors":"Vahid Parvaz","doi":"10.1016/j.ref.2025.100774","DOIUrl":"10.1016/j.ref.2025.100774","url":null,"abstract":"<div><div>The increasing penetration of Electric Vehicles (EVs) and Photovoltaic (PV) systems has introduced new energy management and financial challenges in the electricity market. The inherent fluctuations in PV generation and the unpredictable behavior of the EV charging load carry a significant risk of imbalance and heavy financial penalties. This paper proposes a comprehensive risk management framework for the optimal energy procurement decision of a PV-EV facility in the electricity market, aiming to minimize operational costs. The methodology was developed along two parallel paths: (1) heterogeneity analysis of EV charging load using K-Means clustering and (2) PV generation uncertainty modeling.</div><div>A hybrid statistical-machine learning model was adopted for uncertainty estimation, validating the prediction interval reliability with a Prediction Interval Coverage Probability (PICP) of 93.72%. Finally, the optimal energy procurement decision was formulated using a Robust Optimization model to guarantee load supply under the 90% lower bound of PV generation (<span><math><msubsup><mi>P</mi><mrow><mn>5</mn><mo>%</mo></mrow><mrow><mi>PV</mi></mrow></msubsup></math></span>). The results show that adopting the robust decision, assuming day-ahead market prices (C<sup>DA</sup>=50 UP/kWh) and imbalance settlement (C<sup>RT</sup>=70 UP/kWh), leads to a total operational cost of 68.03 million UP over a one-year testing period. Compared to the ideal (deterministic) model, this cost includes a risk cost of 5,546,926.32 UP (equivalent to 7.53% of the deterministic model’s cost). This additional cost quantifies the economic value of uncertainty information and ensures the system’s 93.84% stability in meeting demand. The findings suggest that the proposed framework not only offers technical solutions but can also provide effective guidance for operators and policymakers in electricity markets.</div></div>","PeriodicalId":29780,"journal":{"name":"Renewable Energy Focus","volume":"56 ","pages":"Article 100774"},"PeriodicalIF":5.9,"publicationDate":"2025-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145424514","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-22DOI: 10.1016/j.ref.2025.100773
Moshood Akanni Alao, Olawale Muhammed Popoola
Optimal allocation of distributed generators (DGs) and capacitor banks (CBs) is crucial for enhancing both the technical performance and environmental sustainability of distribution systems. Uncertainties arising from consumer behaviour, climatic conditions, and time-of-use introduce variability in load demands. Therefore, accounting for load demand uncertainty is essential for effective and sustainable distribution system planning. In this study, a novel fuzzified random number approach is employed to model the uncertainty in voltage-dependent non-linear load (VDL) models. An improved particle swarm optimization (IPSO) algorithm is then applied for the simultaneous allocation of DGs and CBs in radial distribution networks under both constant (CL) and stochastic VDL scenarios. The IPSO method is validated on the IEEE 33-bus system to evaluate the technical, economic, and environmental benefits of DG and CB allocation. Key results indicate that DGs operating at an optimal power factor (OPF-DGs) combined with CBs outperform unity-power-factor (UPF-DGs) and CBs under the CL model, achieving superior outcomes in terms of reduced power loss, lower voltage deviation, and enhanced voltage stability. Specifically, power loss reductions of 94.36% are achieved for UPF-DGs + CBs, compared to 94.89% for OPF-DGs + CBs under CLs. For defuzzified mixed VDLs, a power loss reduction of 93.92% is observed under OPF-DGs + CBs. The proposed approach also demonstrates significant economic benefits, including high net present values and cost savings, alongside substantial reductions in environmental emissions. Furthermore, sensitivity analysis highlights that considering different load mixes, rather than static or single-load configurations, is critical for optimal planning and economic dispatch of distribution systems with DGs and CBs. Incorporating load uncertainty provides a more realistic representation of system performance, emphasizing its importance in sustainable distribution network planning.
{"title":"Optimal allocation of distributed generators and capacitor banks considering load models under stochastic load levels","authors":"Moshood Akanni Alao, Olawale Muhammed Popoola","doi":"10.1016/j.ref.2025.100773","DOIUrl":"10.1016/j.ref.2025.100773","url":null,"abstract":"<div><div>Optimal allocation of distributed generators (DGs) and capacitor banks (CBs) is crucial for enhancing both the technical performance and environmental sustainability of distribution systems. Uncertainties arising from consumer behaviour, climatic conditions, and time-of-use introduce variability in load demands. Therefore, accounting for load demand uncertainty is essential for effective and sustainable distribution system planning. In this study, a novel fuzzified random number approach is employed to model the uncertainty in voltage-dependent non-linear load (VDL) models. An improved particle swarm optimization (IPSO) algorithm is then applied for the simultaneous allocation of DGs and CBs in radial distribution networks under both constant (CL) and stochastic VDL scenarios. The IPSO method is validated on the IEEE 33-bus system to evaluate the technical, economic, and environmental benefits of DG and CB allocation. Key results indicate that DGs operating at an optimal power factor (OPF-DGs) combined with CBs outperform unity-power-factor (UPF-DGs) and CBs under the CL model, achieving superior outcomes in terms of reduced power loss, lower voltage deviation, and enhanced voltage stability. Specifically, power loss reductions of 94.36% are achieved for UPF-DGs + CBs, compared to 94.89% for OPF-DGs + CBs under CLs. For defuzzified mixed VDLs, a power loss reduction of 93.92% is observed under OPF-DGs + CBs. The proposed approach also demonstrates significant economic benefits, including high net present values and cost savings, alongside substantial reductions in environmental emissions. Furthermore, sensitivity analysis highlights that considering different load mixes, rather than static or single-load configurations, is critical for optimal planning and economic dispatch of distribution systems with DGs and CBs. Incorporating load uncertainty provides a more realistic representation of system performance, emphasizing its importance in sustainable distribution network planning.</div></div>","PeriodicalId":29780,"journal":{"name":"Renewable Energy Focus","volume":"56 ","pages":"Article 100773"},"PeriodicalIF":5.9,"publicationDate":"2025-10-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145362640","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-10-17DOI: 10.1016/j.ref.2025.100772
Wiomou Joévin Bonzi , Zhangkai Wu , Sebastian Romuli , Klaus Meissner , Joachim Müller
In resources constrained rural areas, solar-powered oil extraction can be enhanced through recent advances in artificial intelligence for energy optimization. This study introduces SolPrInt, a deep reinforcement-learning (DRL) based controller for a standalone, photovoltaic-battery powered mechanical oil press. A proximal policy optimization (PPO) agent was trained in MATLAB/Simulink using 15 years of PVGIS-SARAH2 radiation data and peanut-oil extraction benchmarks. A primary training phase followed by an adversarial phase on the 5% least-sunny days reinforced robustness under low-irradiance conditions. The developed agent adapts press rotational speed to real-time PV availability, battery state of charge, and system behavior to ensure energy-efficient use of solar resources. In-silico validation achieved stable rewards and simulated throughput of 96 ± 13.5 kg/d under sunny days and 90 ± 20.5 kg/d under cloudy days. Compared with conventional fixed-schedule operation (08:00–18:00) under sunny and cloudy conditions, SolPrInt extends operating time, and reduces power outages, while improves oil yield by 0.7 percentage points. Experimental validation on a PV-simulator bench confirmed real-time deployment feasibility on a low-cost ESP32 microcontroller interfaced with a Kern Kraft KK20 press. These findings demonstrate the potential of PV-sensitive DRL control to improve the performance of standalone renewable energy systems, supporting reliable decentralized energy use and contributing to sustainable energy access. Supplementary materials supporting this work, are available at https://bonjoe.github.io/solprint.demo/.
{"title":"Smart control of standalone solar-powered oil press: Applying Reinforcement Learning for productivity and energy utilization improvement","authors":"Wiomou Joévin Bonzi , Zhangkai Wu , Sebastian Romuli , Klaus Meissner , Joachim Müller","doi":"10.1016/j.ref.2025.100772","DOIUrl":"10.1016/j.ref.2025.100772","url":null,"abstract":"<div><div>In resources constrained rural areas, solar-powered oil extraction can be enhanced through recent advances in artificial intelligence for energy optimization. This study introduces <em>SolPrInt</em>, a deep reinforcement-learning (DRL) based controller for a standalone, photovoltaic-battery powered mechanical oil press. A proximal policy optimization (PPO) agent was trained in MATLAB/Simulink using 15 years of PVGIS-SARAH2 radiation data and peanut-oil extraction benchmarks. A primary training phase followed by an adversarial phase on the 5% least-sunny days reinforced robustness under low-irradiance conditions. The developed agent adapts press rotational speed to real-time PV availability, battery state of charge, and system behavior to ensure energy-efficient use of solar resources. In-silico validation achieved stable rewards and simulated throughput of 96 ± 13.5<!--> <!-->kg/d under sunny days and 90 ± 20.5<!--> <!-->kg/d under cloudy days. Compared with conventional fixed-schedule operation (08:00–18:00) under sunny and cloudy conditions, <em>SolPrInt</em> extends operating time, and reduces power outages, while improves oil yield by 0.7 percentage points. Experimental validation on a PV-simulator bench confirmed real-time deployment feasibility on a low-cost ESP32 microcontroller interfaced with a Kern Kraft KK20 press. These findings demonstrate the potential of PV-sensitive DRL control to improve the performance of standalone renewable energy systems, supporting reliable decentralized energy use and contributing to sustainable energy access. Supplementary materials supporting this work, are available at <span><span>https://bonjoe.github.io/solprint.demo/</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":29780,"journal":{"name":"Renewable Energy Focus","volume":"56 ","pages":"Article 100772"},"PeriodicalIF":5.9,"publicationDate":"2025-10-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145424517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Nowadays, the industrial sector stands as the major energy consumer globally, simultaneously holding a pivotal role as a significant contributor to greenhouse gas emissions. Therefore, energy system planning and management in these systems are under heightened scrutiny due to concerns over energy, economic, and environmental challenges. This study aims to develop a comprehensive optimal model that integrates renewable potential assessment and utilizes particle swarm optimization for accurate and cost-effective planning and operation of the energy system within an industrial zone. The research proposes a novel strategy for planning and operating industrial energy hubs, offering a robust and adaptable framework tailored to industrial zones. By integrating uncertain renewable energy sources and EVs, the framework effectively manages variability and uncertainty. It holistically connects electricity, heating, cooling, and transportation sectors, enabling cross-sectoral flexibility and enhancing system adaptability. The study compares four scenarios: BAU, BAU CO2-Aware, CO2-Blind, and CO2-Aware, evaluating their impact on energy costs, investment, operational cost, and environmental benefits. The results show that the CO2-Aware and CO2-Blind scenarios reduce overall costs by approximately 15% and 10%, respectively, compared to the BAU. Additionally, the CO2-Aware scenario achieves a 32% reduction in CO2 emissions. Despite higher investment and operational costs, these alternative energy systems provide substantial economic and environmental advantages. Additionally, the implementation of this smart energy system within the industrial zone has addressed certain energy challenges in the studied region, such as mitigating electricity shortages during summer and alleviating natural gas shortages in winter.
{"title":"Mitigating emissions: energy balancing in eco-industrial zones considering renewable energy and electric vehicle uncertainties","authors":"Aminabbas Golshanfard , Younes Noorollahi , Hamed Hashemi-Dezaki , Henrik Lund","doi":"10.1016/j.ref.2025.100768","DOIUrl":"10.1016/j.ref.2025.100768","url":null,"abstract":"<div><div>Nowadays, the industrial sector stands as the major energy consumer globally, simultaneously holding a pivotal role as a significant contributor to greenhouse gas emissions. Therefore, energy system planning and management in these systems are under heightened scrutiny due to concerns over energy, economic, and environmental challenges. This study aims to develop a comprehensive optimal model that integrates renewable potential assessment and utilizes particle swarm optimization for accurate and cost-effective planning and operation of the energy system within an industrial zone. The research proposes a novel strategy for planning and operating industrial energy hubs, offering a robust and adaptable framework tailored to industrial zones. By integrating uncertain renewable energy sources and EVs, the framework effectively manages variability and uncertainty. It holistically connects electricity, heating, cooling, and transportation sectors, enabling cross-sectoral flexibility and enhancing system adaptability. The study compares four scenarios: BAU, BAU CO<sub>2</sub>-Aware, CO<sub>2</sub>-Blind, and CO<sub>2</sub>-Aware, evaluating their impact on energy costs, investment, operational cost, and environmental benefits. The results show that the CO<sub>2</sub>-Aware and CO<sub>2</sub>-Blind scenarios reduce overall costs by approximately 15% and 10%, respectively, compared to the BAU. Additionally, the CO<sub>2</sub>-Aware scenario achieves a 32% reduction in CO<sub>2</sub> emissions. Despite higher investment and operational costs, these alternative energy systems provide substantial economic and environmental advantages. Additionally, the implementation of this smart energy system within the industrial zone has addressed certain energy challenges in the studied region, such as mitigating electricity shortages during summer and alleviating natural gas shortages in winter.</div></div>","PeriodicalId":29780,"journal":{"name":"Renewable Energy Focus","volume":"56 ","pages":"Article 100768"},"PeriodicalIF":5.9,"publicationDate":"2025-10-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
A smart neighborhood (SN) comprising multiple home microgrids (HMGs) can provide cost-efficient electricity to end-users while supporting the main grid through ancillary services. The integration of renewable energy sources (RESs), energy storage systems (ESSs), and electric vehicles (EVs) introduces dynamic challenges, particularly under varying EV charging behaviors. To address these challenges, this study develops a hierarchical energy management system (HEMS) formulated as an optimization problem and solved using the Aquila optimizer (AO). The proposed HEMS enables the SN to operate as a cloud-based energy storage system (cloud-based ESS), minimizing energy imports from the main grid while maximizing local self-consumption and revenue. The performance of AO is benchmarked against the Particle Swarm Optimization (PSO) algorithm under two control architectures: (i) individual operation, where each local EMS (LEMS) optimizes its own HMG, and (ii) coordinated operation, where a central EMS (CEMS) synchronizes all HMGs, enabling the SN to function collectively as a cloud-based ESS. Simulation results highlight the superior performance of AO under the coordinated CEMS framework. For standard operation, AO reduces main grid imports to 30.62 kWh compared to 61.66 kWh, maintains higher SOC levels across ESSs and EVs (up to 90%), delivers greater total revenue (£44.662 vs. £22.907), and minimizes cumulative error (10.2% vs. 18.7%). Under different EV charging behaviors, AO demonstrates robust adaptability, achieving lower grid imports (40.43 kWh vs. 49.97 kWh), maintaining higher SOC across ESSs and EVs (up to 88.5%), delivering greater total revenue (£15.311 vs. £12.101, +26.5%), and reducing cumulative error from 158.19 to 146.25 (7.6% improvement). These results confirm that the AO-based HEMS efficiently coordinates distributed energy resources, enabling the SN to function as a reliable cloud-based ESS. It improves energy efficiency, economic returns, and grid support while maintaining resilience under dynamic EV charging conditions, providing a scalable and adaptive framework for future SN energy management.
由多个家庭微电网（hmg）组成的智能社区（SN）可以为最终用户提供经济高效的电力，同时通过辅助服务支持主电网。可再生能源（RESs）、储能系统（ess）和电动汽车（EV）的整合带来了动态挑战，特别是在不同的电动汽车充电行为下。为了解决这些挑战，本研究开发了一种分层能量管理系统（HEMS），该系统被制定为一个优化问题，并使用Aquila优化器（AO）来解决。拟议的HEMS使SN能够作为基于云的储能系统（cloud-based ESS）运行，最大限度地减少从主电网进口的能源，同时最大限度地提高本地自我消耗和收入。AO的性能在两种控制体系结构下以粒子群优化（PSO）算法为基准进行基准测试：(i)单独操作，其中每个本地EMS （LEMS）优化其自己的HMG； （ii）协调操作，其中中央EMS （CEMS）同步所有HMG，使SN能够作为基于云的ESS共同发挥作用。仿真结果表明，在协调CEMS框架下，AO具有优越的性能。对于标准运行，AO将主电网进口量从61.66千瓦时减少到30.62千瓦时，在ess和电动汽车中保持更高的SOC水平（高达90%），提供更高的总收入（44.662英镑对22.907英镑），并最大限度地减少累积误差（10.2%对18.7%）。在不同的电动汽车充电行为下，AO表现出强大的适应性，实现了更低的电网输入（40.43 kWh vs 49.97 kWh），在ess和电动汽车中保持更高的SOC（高达88.5%），提供更高的总收入（15.311英镑vs 12.101英镑，+26.5%），并将累积误差从158.19降低到146.25（改善7.6%）。这些结果证实，基于ao的HEMS有效地协调分布式能源，使SN能够作为可靠的基于云的ESS。它提高了能源效率、经济回报和电网支持，同时在动态电动汽车充电条件下保持弹性，为未来的SN能源管理提供了可扩展和自适应的框架。
{"title":"Hierarchical energy management system for coordinated operation of multiple grid-tied home microgrids","authors":"Omar Muhammed Neda , Jafar Adabi , Mousa Marzband , Hamidreza Gholinezhadomran","doi":"10.1016/j.ref.2025.100766","DOIUrl":"10.1016/j.ref.2025.100766","url":null,"abstract":"<div><div>A smart neighborhood (SN) comprising multiple home microgrids (HMGs) can provide cost-efficient electricity to end-users while supporting the main grid through ancillary services. The integration of renewable energy sources (RESs), energy storage systems (ESSs), and electric vehicles (EVs) introduces dynamic challenges, particularly under varying EV charging behaviors. To address these challenges, this study develops a hierarchical energy management system (HEMS) formulated as an optimization problem and solved using the Aquila optimizer (AO). The proposed HEMS enables the SN to operate as a cloud-based energy storage system (cloud-based ESS), minimizing energy imports from the main grid while maximizing local self-consumption and revenue. The performance of AO is benchmarked against the Particle Swarm Optimization (PSO) algorithm under two control architectures: (i) individual operation, where each local EMS (LEMS) optimizes its own HMG, and (ii) coordinated operation, where a central EMS (CEMS) synchronizes all HMGs, enabling the SN to function collectively as a cloud-based ESS. Simulation results highlight the superior performance of AO under the coordinated CEMS framework. For standard operation, AO reduces main grid imports to 30.62 kWh compared to 61.66 kWh, maintains higher SOC levels across ESSs and EVs (up to 90%), delivers greater total revenue (£44.662 vs. £22.907), and minimizes cumulative error (10.2% vs. 18.7%). Under different EV charging behaviors, AO demonstrates robust adaptability, achieving lower grid imports (40.43 kWh vs. 49.97 kWh), maintaining higher SOC across ESSs and EVs (up to 88.5%), delivering greater total revenue (£15.311 vs. £12.101, +26.5%), and reducing cumulative error from 158.19 to 146.25 (7.6% improvement). These results confirm that the AO-based HEMS efficiently coordinates distributed energy resources, enabling the SN to function as a reliable cloud-based ESS. It improves energy efficiency, economic returns, and grid support while maintaining resilience under dynamic EV charging conditions, providing a scalable and adaptive framework for future SN energy management.</div></div>","PeriodicalId":29780,"journal":{"name":"Renewable Energy Focus","volume":"56 ","pages":"Article 100766"},"PeriodicalIF":5.9,"publicationDate":"2025-10-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145266713","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号