Pub Date : 2026-01-28DOI: 10.1016/j.egyr.2025.11.075
Mostafa Mohammadpourfard , Fateme Ghanaatpishe , Yang Weng , Anurag Srivastava , Chin-Woo Tan
The smart grid, as a critical cyber–physical system, is highly susceptible to False Data Injection Attacks (FDIAs), which pose significant threats to its stability and security. This paper introduces an advanced deep learning framework designed to generate stealthier FDIAs targeting state estimation (SE) in power systems. Our approach incorporates enhanced Autoencoders (AE), Variational Autoencoders (VAE), and Conditional Generative Adversarial Networks (cGANs). These models are optimized and enhanced with physics-informed constraints specific to the power system’s SE process. The developed models are evaluated based on bypass rates, convergence rates, and data diversity, highlighting their ability to evade detection mechanisms, such as bad data detectors (BDD) and similarity-based metrics like Jensen–Shannon Divergence (JSD). Simulations on IEEE 14-bus and 57-bus systems using real-world load data demonstrate the models’ ability to generate highly covert FDIAs while adhering to the physical principles of the grid. The results highlight the substantial risks posed by these advanced attacks and provide critical insights into developing more resilient detection strategies for smart grid security.
{"title":"Optimizing and evaluating deep learning techniques for stealthy false data injection attacks on smart grids","authors":"Mostafa Mohammadpourfard , Fateme Ghanaatpishe , Yang Weng , Anurag Srivastava , Chin-Woo Tan","doi":"10.1016/j.egyr.2025.11.075","DOIUrl":"10.1016/j.egyr.2025.11.075","url":null,"abstract":"<div><div>The smart grid, as a critical cyber–physical system, is highly susceptible to False Data Injection Attacks (FDIAs), which pose significant threats to its stability and security. This paper introduces an advanced deep learning framework designed to generate stealthier FDIAs targeting state estimation (SE) in power systems. Our approach incorporates enhanced Autoencoders (AE), Variational Autoencoders (VAE), and Conditional Generative Adversarial Networks (cGANs). These models are optimized and enhanced with physics-informed constraints specific to the power system’s SE process. The developed models are evaluated based on bypass rates, convergence rates, and data diversity, highlighting their ability to evade detection mechanisms, such as bad data detectors (BDD) and similarity-based metrics like Jensen–Shannon Divergence (JSD). Simulations on IEEE 14-bus and 57-bus systems using real-world load data demonstrate the models’ ability to generate highly covert FDIAs while adhering to the physical principles of the grid. The results highlight the substantial risks posed by these advanced attacks and provide critical insights into developing more resilient detection strategies for smart grid security.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"15 ","pages":"Article 108816"},"PeriodicalIF":5.1,"publicationDate":"2026-01-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146073636","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-10DOI: 10.1016/j.egyr.2025.108989
Di Sha , Arne Johannssen , Xianyi Zeng , Zhenglei He , Hanhan Wu , Kim Phuc Tran
Accurate carbon price prediction is essential for decision-making and risk management. Most existing predictive models produce deterministic results and fail to account for uncertainties in carbon prices. To address this limitation, this study introduces an interval prediction framework that effectively captures uncertainties and enhances predictive performance. The proposed framework integrates eXtreme Gradient Boosting (XGBoost) for feature selection, a Modified Scaling Approach (MSA) to generate asymmetric prediction intervals and an improved Non-Dominated Sorting Genetic Algorithm-II (NSGA-II) to determine the optimal scaling parameters. Finally, a Bidirectional Long Short-Term Memory (BiLSTM) network is employed to generate the final interval prediction results. Experiments show that the presented framework outperforms benchmark models and demonstrates robustness.
准确的碳价预测对决策和风险管理至关重要。大多数现有的预测模型产生的是确定性的结果，无法解释碳价格的不确定性。为了解决这一限制，本研究引入了一个有效捕获不确定性并提高预测性能的区间预测框架。该框架集成了用于特征选择的极限梯度增强（XGBoost）、用于生成非对称预测区间的改进缩放方法（MSA）和用于确定最优缩放参数的改进非支配排序遗传算法- ii （NSGA-II）。最后，利用双向长短期记忆（BiLSTM）网络生成最终的区间预测结果。实验表明，该框架优于基准模型，具有较强的鲁棒性。
{"title":"Carbon price interval prediction by bidirectional long short-term memory and multi-objective optimization with an asymmetric scaling approach","authors":"Di Sha , Arne Johannssen , Xianyi Zeng , Zhenglei He , Hanhan Wu , Kim Phuc Tran","doi":"10.1016/j.egyr.2025.108989","DOIUrl":"10.1016/j.egyr.2025.108989","url":null,"abstract":"<div><div>Accurate carbon price prediction is essential for decision-making and risk management. Most existing predictive models produce deterministic results and fail to account for uncertainties in carbon prices. To address this limitation, this study introduces an interval prediction framework that effectively captures uncertainties and enhances predictive performance. The proposed framework integrates eXtreme Gradient Boosting (XGBoost) for feature selection, a Modified Scaling Approach (MSA) to generate asymmetric prediction intervals and an improved Non-Dominated Sorting Genetic Algorithm-II (NSGA-II) to determine the optimal scaling parameters. Finally, a Bidirectional Long Short-Term Memory (BiLSTM) network is employed to generate the final interval prediction results. Experiments show that the presented framework outperforms benchmark models and demonstrates robustness.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"15 ","pages":"Article 108989"},"PeriodicalIF":5.1,"publicationDate":"2026-01-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-08DOI: 10.1016/j.egyr.2025.108952
Eldad Appiah, Yaw Opoku Mensah Sekyere, Francis Boafo Effah
Congestion in deregulated power markets poses a significant challenge to power system reliability, economic dispatch, and market efficiency. This paper proposes a novel multi-stage congestion management framework that integrates generator rescheduling using Particle Swarm Optimisation (PSO) with a severity-based, hierarchical application of demand-side strategies, specifically Real-Time Pricing (RTP) and Adaptive Load Shedding. The methodology is designed to progressively relieve congestion by activating increasingly stringent measures only when preceding steps prove insufficient. Simulations conducted on the IEEE 30-bus and IEEE 118-bus test systems under artificially induced congestion conditions demonstrate the effectiveness and scalability of the proposed framework. In the IEEE 30-bus case, the total generation cost decreases from 2280.90 USD/hr (post-congestion) to 1633.09 USD/hr, with all line flows restored within thermal limits. Application to the larger IEEE 118-bus system further validates the approach, reducing the generation cost from 335,965 USD/hr (congested) to 193,681 USD/hr after the complete three-stage process, i.e., a 42.35 % reduction, while total demand falls moderately from 6055 MW to 5579 MW, i.e., a 7.86 % as congestion is fully eliminated. Results show that although PSO-based generator rescheduling significantly reduces overloads, it is insufficient for full congestion clearance, therefore necessitating the successive deployment of RTP and adaptive load shedding. Compared to conventional single-method solutions, the proposed strategy achieves enhanced technical efficiency, demand-side flexibility, and operational robustness. This work contributes a scalable and adaptive solution for congestion management in emerging electricity markets, particularly in environments transitioning to market-based operation.
{"title":"A multi-stage congestion management strategy in deregulated power markets using generator rescheduling and demand-side interventions","authors":"Eldad Appiah, Yaw Opoku Mensah Sekyere, Francis Boafo Effah","doi":"10.1016/j.egyr.2025.108952","DOIUrl":"10.1016/j.egyr.2025.108952","url":null,"abstract":"<div><div>Congestion in deregulated power markets poses a significant challenge to power system reliability, economic dispatch, and market efficiency. This paper proposes a novel multi-stage congestion management framework that integrates generator rescheduling using Particle Swarm Optimisation (PSO) with a severity-based, hierarchical application of demand-side strategies, specifically Real-Time Pricing (RTP) and Adaptive Load Shedding. The methodology is designed to progressively relieve congestion by activating increasingly stringent measures only when preceding steps prove insufficient. Simulations conducted on the IEEE 30-bus and IEEE 118-bus test systems under artificially induced congestion conditions demonstrate the effectiveness and scalability of the proposed framework. In the IEEE 30-bus case, the total generation cost decreases from 2280.90 USD/hr (post-congestion) to 1633.09 USD/hr, with all line flows restored within thermal limits. Application to the larger IEEE 118-bus system further validates the approach, reducing the generation cost from 335,965 USD/hr (congested) to 193,681 USD/hr after the complete three-stage process, i.e., a 42.35 % reduction, while total demand falls moderately from 6055 MW to 5579 MW, i.e., a 7.86 % as congestion is fully eliminated. Results show that although PSO-based generator rescheduling significantly reduces overloads, it is insufficient for full congestion clearance, therefore necessitating the successive deployment of RTP and adaptive load shedding. Compared to conventional single-method solutions, the proposed strategy achieves enhanced technical efficiency, demand-side flexibility, and operational robustness. This work contributes a scalable and adaptive solution for congestion management in emerging electricity markets, particularly in environments transitioning to market-based operation.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"15 ","pages":"Article 108952"},"PeriodicalIF":5.1,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921295","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-08DOI: 10.1016/j.egyr.2025.108973
Nhu Anh Phan, Lisa Göransson, Filip Johnsson
Electrification of transport and industry, a crucial pathway for emission mitigation, may result in a large increase of electricity demand in Sweden. In this study, we investigate the transition bottlenecks for Sweden's electrification using a mixed-methods approach. We first use energy systems modeling to identify cost-efficient combinations of generation, storage, and demand-side flexibility that can meet the projected demand from electrification. Three cases are applied that differ in predetermined investments in offshore wind power and nuclear power. We then apply a multi-level perspective analysis on the three cases with the aim to map out the main characteristics of the Swedish electricity system. We base this on historical development, as well as the impacting landscape, indicating broad, long-term trends external to the system, and niche factors, referring to technological and social innovations. Drawing on these characteristics and modeling insights, we identify transition bottlenecks to Swedish electrification. We find that changes at the landscape level have been insufficient to enable a shift to an electricity system that has a high share of wind and solar power. Instead, the operational and regulatory regimes are strongly influenced by the existing system, which is dominated by synchronous electricity generation from hydropower and nuclear power. Yet, new nuclear power struggles to become cost-competitive in the deregulated electricity market. Thus, transition bottlenecks exist across all modeled futures.
{"title":"Mind the gap: Mixed-methods approach to investigate transition bottlenecks to low-carbon energy futures","authors":"Nhu Anh Phan, Lisa Göransson, Filip Johnsson","doi":"10.1016/j.egyr.2025.108973","DOIUrl":"10.1016/j.egyr.2025.108973","url":null,"abstract":"<div><div>Electrification of transport and industry, a crucial pathway for emission mitigation, may result in a large increase of electricity demand in Sweden. In this study, we investigate the transition bottlenecks for Sweden's electrification using a mixed-methods approach. We first use energy systems modeling to identify cost-efficient combinations of generation, storage, and demand-side flexibility that can meet the projected demand from electrification. Three cases are applied that differ in predetermined investments in offshore wind power and nuclear power. We then apply a multi-level perspective analysis on the three cases with the aim to map out the main characteristics of the Swedish electricity system. We base this on historical development, as well as the impacting landscape, indicating broad, long-term trends external to the system, and niche factors, referring to technological and social innovations. Drawing on these characteristics and modeling insights, we identify transition bottlenecks to Swedish electrification. We find that changes at the landscape level have been insufficient to enable a shift to an electricity system that has a high share of wind and solar power. Instead, the operational and regulatory regimes are strongly influenced by the existing system, which is dominated by synchronous electricity generation from hydropower and nuclear power. Yet, new nuclear power struggles to become cost-competitive in the deregulated electricity market. Thus, transition bottlenecks exist across all modeled futures.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"15 ","pages":"Article 108973"},"PeriodicalIF":5.1,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921608","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-07DOI: 10.1016/j.egyr.2025.109031
Kacper Mańkowski , Bartłomiej Bajan , Aldona Mrówczyńska-Kamińska
The European Union’s (EU) renewable energy targets for 2030 require a substantial acceleration in the adoption of renewable energy sources (RES) across all sectors. While macro-level progress has been notable, agriculture and agribusiness continue to lag in RES integration, thereby slowing down the overall pace of the energy transition. This study presents the first EU-wide assessment of RES uptake in these sectors, using an input–output model informed by RES targets extracted from the updated 2023 National Energy and Climate Plans (NECPs), prepared under the Renewable Energy Directive III (RED III). Unlike previous studies based on outdated RED II assumptions, this analysis reflects the revised 2023 policy landscape, providing a timely and policy-relevant perspective. Convergence toward the 42.5 % RES target was estimated using two historical trends: 2014–2022 and 2018–2022. Under the first trend, the overall economy is projected to reach the target by 2045, with agribusiness and agriculture lagging by 6 and 27 years, respectively. Under the second, more recent trend, convergence could occur by 2040 for the overall economy, with delays of 7 years for agribusiness and 21 years for agriculture. Although the 2030 RES target appears achievable at the aggregate level, deep structural disparities persist. Accelerating the transition in lagging sectors will require targeted incentives, investments in decentralized energy systems, and geographically differentiated policies aligned with national and regional resource conditions. These findings indicate that strengthened rural investment frameworks in biogas or the electrification of farm machinery could help close the sectoral gaps in RES adoption.
{"title":"Renewable energy in EU agribusiness: Review of progress in meeting 2030 renewable energy directive III targets","authors":"Kacper Mańkowski , Bartłomiej Bajan , Aldona Mrówczyńska-Kamińska","doi":"10.1016/j.egyr.2025.109031","DOIUrl":"10.1016/j.egyr.2025.109031","url":null,"abstract":"<div><div>The European Union’s (EU) renewable energy targets for 2030 require a substantial acceleration in the adoption of renewable energy sources (RES) across all sectors. While macro-level progress has been notable, agriculture and agribusiness continue to lag in RES integration, thereby slowing down the overall pace of the energy transition. This study presents the first EU-wide assessment of RES uptake in these sectors, using an input–output model informed by RES targets extracted from the updated 2023 National Energy and Climate Plans (NECPs), prepared under the Renewable Energy Directive III (RED III). Unlike previous studies based on outdated RED II assumptions, this analysis reflects the revised 2023 policy landscape, providing a timely and policy-relevant perspective. Convergence toward the 42.5 % RES target was estimated using two historical trends: 2014–2022 and 2018–2022. Under the first trend, the overall economy is projected to reach the target by 2045, with agribusiness and agriculture lagging by 6 and 27 years, respectively. Under the second, more recent trend, convergence could occur by 2040 for the overall economy, with delays of 7 years for agribusiness and 21 years for agriculture. Although the 2030 RES target appears achievable at the aggregate level, deep structural disparities persist. Accelerating the transition in lagging sectors will require targeted incentives, investments in decentralized energy systems, and geographically differentiated policies aligned with national and regional resource conditions. These findings indicate that strengthened rural investment frameworks in biogas or the electrification of farm machinery could help close the sectoral gaps in RES adoption.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"15 ","pages":"Article 109031"},"PeriodicalIF":5.1,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921605","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-07DOI: 10.1016/j.egyr.2025.108964
Kwonye Kim , Kwanwoo Kim , Sehoon Hwang , Yura Choi , Soowon Chae , Changbae Park , Yujin Nam
Globally, seafood consumption is steadily increasing, driven by population growth and rising health awareness, with South Korea recording some of the highest per capita seafood consumption rates worldwide. This highlights the importance of sustainable aquaculture systems, such as the Recirculating Aquaculture System (RAS), which can recycle 90–99 % of aquaculture water, thereby reducing water use and mitigating disease and pollution risks. However, high initial investment and significant energy costs for maintaining optimal water temperatures hinder widespread adoption. To improve economic and energy efficiency, this study evaluates heat pump systems utilizing various natural heat sources (seawater, air, ground) in comparison to conventional boiler–chiller systems for commercial-scale RAS facilities. Using TRNSYS dynamic energy simulations, the research assesses each system’s thermal performance, energy consumption, and operational cost. In addition to energy and maintenance savings, this study introduces a profitability assessment framework by estimating annual revenue from fish production and calculating key economic indicators such as payback period and net profit. Results show that while the ground-source heat pump (GSHP) offers the highest energy efficiency, the seawater-source heat pump (WSHP) achieves the best overall balance between initial investment and operational profit. The findings aim to provide practical insights for selecting optimal heat source systems, reducing the total cost of ownership, and supporting sustainable growth and renewable energy integration in the aquaculture industry.
{"title":"Economic and performance analysis of heat pump system with renewable heat sources for recirculating aquaculture system (RAS)","authors":"Kwonye Kim , Kwanwoo Kim , Sehoon Hwang , Yura Choi , Soowon Chae , Changbae Park , Yujin Nam","doi":"10.1016/j.egyr.2025.108964","DOIUrl":"10.1016/j.egyr.2025.108964","url":null,"abstract":"<div><div>Globally, seafood consumption is steadily increasing, driven by population growth and rising health awareness, with South Korea recording some of the highest per capita seafood consumption rates worldwide. This highlights the importance of sustainable aquaculture systems, such as the Recirculating Aquaculture System (RAS), which can recycle 90–99 % of aquaculture water, thereby reducing water use and mitigating disease and pollution risks. However, high initial investment and significant energy costs for maintaining optimal water temperatures hinder widespread adoption. To improve economic and energy efficiency, this study evaluates heat pump systems utilizing various natural heat sources (seawater, air, ground) in comparison to conventional boiler–chiller systems for commercial-scale RAS facilities. Using TRNSYS dynamic energy simulations, the research assesses each system’s thermal performance, energy consumption, and operational cost. In addition to energy and maintenance savings, this study introduces a profitability assessment framework by estimating annual revenue from fish production and calculating key economic indicators such as payback period and net profit. Results show that while the ground-source heat pump (GSHP) offers the highest energy efficiency, the seawater-source heat pump (WSHP) achieves the best overall balance between initial investment and operational profit. The findings aim to provide practical insights for selecting optimal heat source systems, reducing the total cost of ownership, and supporting sustainable growth and renewable energy integration in the aquaculture industry.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"15 ","pages":"Article 108964"},"PeriodicalIF":5.1,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921611","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The aim of this paper is to develop and evaluate a nature-inspired metaheuristic strategy for Maximum Power Point Tracking (MPPT) strategy in Proton Exchange Membrane Fuel Cells (PEMFCs), whose efficiency is highly sensitive to dynamic operating conditions such as cell temperature and the partial pressures of hydrogen and oxygen. These fluctuations continually shift the system’s Maximum Power Point (MPP), necessitating adaptive control methods to maintain optimal power extraction. This study introduces a novel MPPT technique based on the Horse Herd Optimization Algorithm (HOA), a recent bio-inspired metaheuristic modeled on the social behavior of horse populations. To the best of our knowledge, this work presents the first application of HOA to PEMFC systems. A comprehensive dynamic model is constructed, integrating the electrochemical characteristics of a 50 kW PEMFC stack, a DC-DC boost converter, and an adaptive MPPT controller guided by HOA. The algorithm adjusts the converter’s duty cycle by mimicking behavioral mechanisms—such as grazing, hierarchy, sociability, imitation, defense, and roaming—organized across age-based groups to enhance convergence speed and accuracy. The effectiveness of the HOA-based MPPT is benchmarked against the Cuckoo Search Optimization (CSO) method under various conditions, including standard operation, temperature variations (328 K to 348 K), and pressure fluctuations (1.0–2.0 atm). Simulation results using MATLAB/Simulink demonstrate that the HOA algorithm achieves superior performance, with a maximum power point tracking efficiency of 99.7 % compared to 99.64 % for CSO. Additionally, HOA exhibits a significantly faster settling time of 0.0570 s, outperforming CSO's 0.12 s, and maintains comparable rise times () while eliminating voltage and current oscillations. Under varying thermal and pressure conditions, HOA demonstrates exceptional robustness, rapid convergence, and high stability, maintaining optimal power delivery where conventional methods degrade. This work represents the first successful integration of the Horse Herd Optimization Algorithm into MPPT control for PEM fuel cells and demonstrates its superiority over both traditional and intelligent techniques. It offers a highly efficient and adaptive solution, with promising prospects for future scalability and deployment in real-world fuel cell energy management systems.
{"title":"Enhancing PEM fuel cell efficiency through bio-inspired MPPT under variable operating conditions","authors":"Fatima Zohra Kebbab , Mohit Bajaj , Vojtech Blazek , Lukas Prokop","doi":"10.1016/j.egyr.2025.108999","DOIUrl":"10.1016/j.egyr.2025.108999","url":null,"abstract":"<div><div>The aim of this paper is to develop and evaluate a nature-inspired metaheuristic strategy for Maximum Power Point Tracking (MPPT) strategy in Proton Exchange Membrane Fuel Cells (PEMFCs), whose efficiency is highly sensitive to dynamic operating conditions such as cell temperature and the partial pressures of hydrogen and oxygen. These fluctuations continually shift the system’s Maximum Power Point (MPP), necessitating adaptive control methods to maintain optimal power extraction. This study introduces a novel MPPT technique based on the Horse Herd Optimization Algorithm (HOA), a recent bio-inspired metaheuristic modeled on the social behavior of horse populations. To the best of our knowledge, this work presents the first application of HOA to PEMFC systems. A comprehensive dynamic model is constructed, integrating the electrochemical characteristics of a 50 kW PEMFC stack, a DC-DC boost converter, and an adaptive MPPT controller guided by HOA. The algorithm adjusts the converter’s duty cycle by mimicking behavioral mechanisms—such as grazing, hierarchy, sociability, imitation, defense, and roaming—organized across age-based groups to enhance convergence speed and accuracy. The effectiveness of the HOA-based MPPT is benchmarked against the Cuckoo Search Optimization (CSO) method under various conditions, including standard operation, temperature variations (328 K to 348 K), and pressure fluctuations (1.0–2.0 atm). Simulation results using MATLAB/Simulink demonstrate that the HOA algorithm achieves superior performance, with a maximum power point tracking efficiency of 99.7 % compared to 99.64 % for CSO. Additionally, HOA exhibits a significantly faster settling time of 0.0570 s, outperforming CSO's 0.12 s, and maintains comparable rise times (<span><math><mrow><mn>0</mn><mo>.</mo><mn>0016</mn><mi>s</mi></mrow></math></span>) while eliminating voltage and current oscillations. Under varying thermal and pressure conditions, HOA demonstrates exceptional robustness, rapid convergence, and high stability, maintaining optimal power delivery where conventional methods degrade. This work represents the first successful integration of the Horse Herd Optimization Algorithm into MPPT control for PEM fuel cells and demonstrates its superiority over both traditional and intelligent techniques. It offers a highly efficient and adaptive solution, with promising prospects for future scalability and deployment in real-world fuel cell energy management systems.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"15 ","pages":"Article 108999"},"PeriodicalIF":5.1,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921444","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The paper explores the prerequisites for the expansion of energy cooperatives as a market-based form evolutioning from self-sufficiency of electricity supply to the wide range of models of the energy and energy-related business activity, basing on the international and Ukrainian experience. We assume that a cooperative size and energy technologies combination are important parameters that determine their efficiency. The insufficient per capita household income as the barrier of the energy cooperatives development hinders the development of sustainable energy and market transitions. To provide the investments that are able to cover the cost of purchasing and installing of generating units for self-sufficiency and active consumption, it is needed to align the share contribution and the needed installed generation capacity sufficient for the number of participants united with the appropriate amount of the share contribution and to define the optimal cooperative size. Maintaining the individual interest for each participant creates a driving force for increasing the number of energy cooperatives.
{"title":"Development of energy cooperatives in Ukraine: scale, energy mix and business models","authors":"Uliana Pysmenna , Sviatoslav Petrovets , Iryna Sotnyk , Tetiana Kurbatova","doi":"10.1016/j.egyr.2025.108965","DOIUrl":"10.1016/j.egyr.2025.108965","url":null,"abstract":"<div><div>The paper explores the prerequisites for the expansion of energy cooperatives as a market-based form evolutioning from self-sufficiency of electricity supply to the wide range of models of the energy and energy-related business activity, basing on the international and Ukrainian experience. We assume that a cooperative size and energy technologies combination are important parameters that determine their efficiency. The insufficient per capita household income as the barrier of the energy cooperatives development hinders the development of sustainable energy and market transitions. To provide the investments that are able to cover the cost of purchasing and installing of generating units for self-sufficiency and active consumption, it is needed to align the share contribution and the needed installed generation capacity sufficient for the number of participants united with the appropriate amount of the share contribution and to define the optimal cooperative size. Maintaining the individual interest for each participant creates a driving force for increasing the number of energy cooperatives.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"15 ","pages":"Article 108965"},"PeriodicalIF":5.1,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921446","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-06DOI: 10.1016/j.egyr.2025.108993
Daehyuk Kim , SoonSik Jang , Jung-il Lee , Shin Hyung Rhee , Hyukchan Kwon , Hyunhak Jeong
This study develops a data-driven yet physically consistent framework to quantify the real-world fuel efficiency of a PTO-mode shaft generator (S/G) using high-resolution operational data from an 86k CBM LPG carrier. To capture the nonlinear and load-dependent interactions between propulsion and electric power generation, an Input Convex Neural Network (ICNN) is trained under convexity and monotonicity constraints in the S/G power ratio, ensuring both accurate prediction and physically reliable response shapes. Compared with conventional machine-learning models (MLP, gradient boosting, random forest), the ICNN achieves comparable predictive accuracy while eliminating the shape violations frequently observed in unconstrained models. Model performance remains high for both laden and ballast operations (R² = 0.88 – 0.94; RMSE ≈ 1 – 3 g/kWh). On-manifold partial dependence analysis reveals distinct operational patterns: in laden voyages, fuel savings emerge primarily above ∼ 60 % S/G power ratio, reaching ∼3.8 g/kWh at full utilization, whereas in ballast voyages savings increase monotonically across the entire range, peaking at ∼1.65 g/kWh. These behaviors align with main engine BSFC maps and reflect condition-dependent trade-offs between propulsion loading and auxiliary generator displacement. The proposed framework provides actionable guidelines, maximizing S/G use in ballast runs and selectively increasing it in laden runs, and offers a robust foundation for AI-assisted energy-management systems that optimize fuel economy under realistic operating constraints.
本研究开发了一个数据驱动但物理上一致的框架，利用来自86k CBM LPG运输船的高分辨率运行数据，量化pto模式轴发电机（S/G）的实际燃油效率。为了捕获推进和发电之间的非线性和负载相关的相互作用，在S/G功率比的凸性和单调性约束下训练输入凸神经网络（ICNN），以确保准确的预测和物理可靠的响应形状。与传统的机器学习模型（MLP、梯度增强、随机森林）相比，ICNN在消除无约束模型中经常观察到的形状违反的同时，达到了相当的预测精度。模型在载货和压载工况下的性能仍然很高（R²= 0.88 - 0.94;RMSE≈1 - 3 g/kWh）。流形部分依赖分析揭示了不同的运行模式：在满载航行中，燃料节约主要出现在~ 60 % S/G功率比以上，在充分利用时达到~ 3.8 G /kWh，而在压载航行中，节省在整个范围内单调增加，峰值为~ 1.65 G /kWh。这些行为与主机BSFC图一致，并反映了推进负载和辅助发电机排量之间的条件相关权衡。拟议的框架提供了可操作的指导方针，在压载运行中最大化S/G的使用，并在负载运行中选择性地增加S/G的使用，并为人工智能辅助能源管理系统提供了坚实的基础，该系统可以在实际操作限制下优化燃油经济性。
{"title":"Ship operational data driven fuel efficiency assessment for shaft generator using input convex neural network","authors":"Daehyuk Kim , SoonSik Jang , Jung-il Lee , Shin Hyung Rhee , Hyukchan Kwon , Hyunhak Jeong","doi":"10.1016/j.egyr.2025.108993","DOIUrl":"10.1016/j.egyr.2025.108993","url":null,"abstract":"<div><div>This study develops a data-driven yet physically consistent framework to quantify the real-world fuel efficiency of a PTO-mode shaft generator (S/G) using high-resolution operational data from an 86k CBM LPG carrier. To capture the nonlinear and load-dependent interactions between propulsion and electric power generation, an Input Convex Neural Network (ICNN) is trained under convexity and monotonicity constraints in the S/G power ratio, ensuring both accurate prediction and physically reliable response shapes. Compared with conventional machine-learning models (MLP, gradient boosting, random forest), the ICNN achieves comparable predictive accuracy while eliminating the shape violations frequently observed in unconstrained models. Model performance remains high for both laden and ballast operations (R² = 0.88 – 0.94; RMSE ≈ 1 – 3 g/kWh). On-manifold partial dependence analysis reveals distinct operational patterns: in laden voyages, fuel savings emerge primarily above ∼ 60 % S/G power ratio, reaching ∼3.8 g/kWh at full utilization, whereas in ballast voyages savings increase monotonically across the entire range, peaking at ∼1.65 g/kWh. These behaviors align with main engine BSFC maps and reflect condition-dependent trade-offs between propulsion loading and auxiliary generator displacement. The proposed framework provides actionable guidelines, maximizing S/G use in ballast runs and selectively increasing it in laden runs, and offers a robust foundation for AI-assisted energy-management systems that optimize fuel economy under realistic operating constraints.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"15 ","pages":"Article 108993"},"PeriodicalIF":5.1,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921509","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-06DOI: 10.1016/j.egyr.2025.108983
Shuang Zhao , Chengcheng Liu , Ji Ma
The growing global energy demand necessitates more efficient reservoir management, highlighting the need for advanced injection-production optimization strategies. Traditional methods are computationally expensive and scale poorly, especially when new wells are added and nearby agents must be retrained. These limitations hinder their application in large, heterogeneous reservoirs and reduce their effectiveness for adaptive decision-making. To address these challenges, this study introduces an enhanced multi-agent reinforcement learning (MARL) framework with three key innovations. First, an online-updating surrogate model, based on a simplified U-Net architecture, partially replaces costly numerical simulations, significantly reducing computational overhead. Second, a regional observation construction method encodes relative well positions to capture inter-well dependencies and enhance local decision-making. Third, a self-adaptive, graph-based unified agent design eliminates the need for retraining when new wells are added, ensuring scalability. The proposed framework was validated using both a synthetic “Three-Channel” model and a real oilfield case. Experimental results show substantial improvements in displacement efficiency, delayed water breakthrough, and increased Net Present Value (NPV). Additionally, the framework adapts seamlessly to the addition of new wells without retraining, maintaining high computational efficiency. These results underscore the practical potential of the MARL-based approach as a robust and flexible solution for real-time reservoir management in dynamically evolving oil fields.
{"title":"Scalable and adaptive injection-production control in reservoirs via a multi-agent reinforcement learning approach","authors":"Shuang Zhao , Chengcheng Liu , Ji Ma","doi":"10.1016/j.egyr.2025.108983","DOIUrl":"10.1016/j.egyr.2025.108983","url":null,"abstract":"<div><div>The growing global energy demand necessitates more efficient reservoir management, highlighting the need for advanced injection-production optimization strategies. Traditional methods are computationally expensive and scale poorly, especially when new wells are added and nearby agents must be retrained. These limitations hinder their application in large, heterogeneous reservoirs and reduce their effectiveness for adaptive decision-making. To address these challenges, this study introduces an enhanced multi-agent reinforcement learning (MARL) framework with three key innovations. First, an online-updating surrogate model, based on a simplified U-Net architecture, partially replaces costly numerical simulations, significantly reducing computational overhead. Second, a regional observation construction method encodes relative well positions to capture inter-well dependencies and enhance local decision-making. Third, a self-adaptive, graph-based unified agent design eliminates the need for retraining when new wells are added, ensuring scalability. The proposed framework was validated using both a synthetic “Three-Channel” model and a real oilfield case. Experimental results show substantial improvements in displacement efficiency, delayed water breakthrough, and increased Net Present Value (NPV). Additionally, the framework adapts seamlessly to the addition of new wells without retraining, maintaining high computational efficiency. These results underscore the practical potential of the MARL-based approach as a robust and flexible solution for real-time reservoir management in dynamically evolving oil fields.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"15 ","pages":"Article 108983"},"PeriodicalIF":5.1,"publicationDate":"2026-01-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145921517","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","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