Pub Date : 2026-05-01Epub Date: 2026-02-23DOI: 10.1016/j.apenergy.2026.127585
Jonathan Hachez , Nicolas Ghilain , Ali Kök , Diederik Coppitters
This paper introduces a multi-stage stochastic optimization framework based on Stochastic Dual Dynamic Programming (SDDP) to plan long-term District Heating and Cooling (DHC) systems under deep climate uncertainty. Integrating Shared Socioeconomic Pathways (SSPs) with uncertain CO2 prices and thermal demands, the framework provides adaptive investment strategies until 2100. Applied to a Belgian case, results project a fundamental shift from heating to cooling, with heat losses declining by up to and heat gains increasing by up to by 2100 across scenarios. The model consistently converges to robust, electrified configurations dominated by Air-Source Heat Pump (ASHP) and Ground-Source Heat Pump (GSHP) supported by seasonal Borehole Thermal Energy Storage (BTES), with Natural Gas Boiler (NG) relegated to marginal backup roles. While transition mechanisms differ, driven by warming in high-emission pathways and by carbon pricing in mitigation pathways, system costs and emissions converge across scenarios. This work demonstrates that electrified DHC systems with seasonal storage offer a cost-effective, resilient strategy for temperate climates under deep uncertainty, though outcomes are sensitive to regional climate and demand profiles.
{"title":"Climate-driven load shifts and the optimal design of district heating and cooling systems: Planning energy supply for a warming century","authors":"Jonathan Hachez , Nicolas Ghilain , Ali Kök , Diederik Coppitters","doi":"10.1016/j.apenergy.2026.127585","DOIUrl":"10.1016/j.apenergy.2026.127585","url":null,"abstract":"<div><div>This paper introduces a multi-stage stochastic optimization framework based on Stochastic Dual Dynamic Programming (SDDP) to plan long-term District Heating and Cooling (DHC) systems under deep climate uncertainty. Integrating Shared Socioeconomic Pathways (SSPs) with uncertain CO2 prices and thermal demands, the framework provides adaptive investment strategies until 2100. Applied to a Belgian case, results project a fundamental shift from heating to cooling, with heat losses declining by up to <span><math><mo>−</mo><mn>57</mn><mi>%</mi><mo>±</mo><mn>34</mn><mi>%</mi></math></span> and heat gains increasing by up to <span><math><mo>+</mo><mn>291</mn><mi>%</mi><mo>±</mo><mn>25</mn><mi>%</mi></math></span> by 2100 across scenarios. The model consistently converges to robust, electrified configurations dominated by Air-Source Heat Pump (ASHP) and Ground-Source Heat Pump (GSHP) supported by seasonal Borehole Thermal Energy Storage (BTES), with Natural Gas Boiler (NG) relegated to marginal backup roles. While transition mechanisms differ, driven by warming in high-emission pathways and by carbon pricing in mitigation pathways, system costs and emissions converge across scenarios. This work demonstrates that electrified DHC systems with seasonal storage offer a cost-effective, resilient strategy for temperate climates under deep uncertainty, though outcomes are sensitive to regional climate and demand profiles.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":"410 ","pages":"Article 127585"},"PeriodicalIF":11.0,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147386220","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The Offshore Wind Power-Seawater Hydrogen Production-Marine Ranch Integrated System (I-OWHR) addresses the dual challenges of offshore wind power integration and intensive marine space utilization through vertical layered development, demonstrating considerable development potential. However, the project is still at an early stage of development, with limited historical data and practical experience. Moreover, the involvement of multiple interconnected subsystems makes investment decision-making for I-OWHR highly complex. To address these challenges, this study integrates a knowledge-driven Large Language Model (LLM) with a Multi-Criteria Decision-Making (MCDM) framework and proposes a two-stage intelligent robust decision-making framework, termed Large-Language-Model-driven Literature-Frequency-Driven Weighting (LLM-LFDW)-Weighted Perturbation-based Stochastic TOPSIS (WP-STOPSIS). It used the LLM to identify evaluation criteria and further incorporated the risk of fluctuations in criteria importance into the MCDM through WP-STOPSIS. The results indicated that hydrogen blending and co-transportation via existing natural gas pipelines, along with the reutilization of decommissioned offshore platforms, consistently emerged as the preferred investment and construction options for current I-OWHR projects. Comparative experimental analysis confirmed the rationality of the criteria identification and weighting method based on LLM-LFDW, with a correlation coefficient of 0.7 relative to mainstream subjective weighting approaches, indicating strong consensus representativeness. Meanwhile, sensitivity comparison analysis demonstrated that WP-STOPSIS exhibits high robustness under weight perturbations, achieving a robustness index of 0.9903. Compared with conventional MCDM methods, the sensitivity is reduced by 73.18%. Furthermore, the proposed two-stage intelligent robust MCDM approach can be independently applied to support decision-making in other complex energy projects at an early stage of development.
{"title":"A robust MCDM framework with LLM for offshore wind power-seawater hydrogen production-marine ranch integrated system investment decision","authors":"Xiaoyu Yu , Xiwen Cui , Dongxiao Niu , Yuchen Diao , Xiaodan Zhang","doi":"10.1016/j.apenergy.2026.127547","DOIUrl":"10.1016/j.apenergy.2026.127547","url":null,"abstract":"<div><div>The Offshore Wind Power-Seawater Hydrogen Production-Marine Ranch Integrated System (I-OWHR) addresses the dual challenges of offshore wind power integration and intensive marine space utilization through vertical layered development, demonstrating considerable development potential. However, the project is still at an early stage of development, with limited historical data and practical experience. Moreover, the involvement of multiple interconnected subsystems makes investment decision-making for I-OWHR highly complex. To address these challenges, this study integrates a knowledge-driven Large Language Model (LLM) with a Multi-Criteria Decision-Making (MCDM) framework and proposes a two-stage intelligent robust decision-making framework, termed Large-Language-Model-driven Literature-Frequency-Driven Weighting (LLM-LFDW)-Weighted Perturbation-based Stochastic TOPSIS (WP-STOPSIS). It used the LLM to identify evaluation criteria and further incorporated the risk of fluctuations in criteria importance into the MCDM through WP-STOPSIS. The results indicated that hydrogen blending and co-transportation via existing natural gas pipelines, along with the reutilization of decommissioned offshore platforms, consistently emerged as the preferred investment and construction options for current I-OWHR projects. Comparative experimental analysis confirmed the rationality of the criteria identification and weighting method based on LLM-LFDW, with a correlation coefficient of 0.7 relative to mainstream subjective weighting approaches, indicating strong consensus representativeness. Meanwhile, sensitivity comparison analysis demonstrated that WP-STOPSIS exhibits high robustness under weight perturbations, achieving a robustness index of 0.9903. Compared with conventional MCDM methods, the sensitivity is reduced by 73.18%. Furthermore, the proposed two-stage intelligent robust MCDM approach can be independently applied to support decision-making in other complex energy projects at an early stage of development.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":"410 ","pages":"Article 127547"},"PeriodicalIF":11.0,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147386221","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2026-05-01Epub Date: 2026-02-24DOI: 10.1016/j.apenergy.2026.127570
Tian Peng , Zhongzheng Mo , Jie Chen , Chenghao Sun , Zhi Wang , Muhammad Shahzad Nazir , Chu Zhang
Accurate prediction of battery remaining useful life (RUL) is crucial for enhancing equipment safety and reliability as well as for sustainable environmental development. This paper proposes a lithium-ion battery RUL prediction model that combines the maximum information coefficient (MIC), the Bayesian optimization algorithm (BOA), kernel density estimation (KDE), and the time-series dense encoder (TiDE). First, using NASA's publicly available lithium-ion battery cycle-life data and the CAL-CE dataset, multiple health indicators including constant-current charge time, discharge time, and IC-curve peak values are extracted and selected via MIC. Next, the TiDE model is employed for accurate RUL prediction, with its key hyperparameters optimized by BOA to boost predictive performance. Finally, KDE is adopted to produce probabilistic RUL forecasts and construct confidence intervals that quantify prediction uncertainty, thereby refining the overall assessment. Comparative experiments demonstrate that the MIC-BOA-TiDE framework reduces the capacity-forecast MAE by 68.9% versus a back-propagation network and by 46.0% versus a GRU baseline, while its RUL error converges to virtually zero, underscoring its superior accuracy and stability. Additionally, the KDE-based interval prediction results show that at the 95% confidence level, the coverage probability on the CS2_35 dataset reaches 95.27% with an average interval width of 0.0731, confirming the model's effectiveness in quantifying predictive uncertainty.
{"title":"A novel MIC-BOA-TiDE fusion framework with kernel density estimation for point and probabilistic remaining useful life prediction of lithium-ion batteries","authors":"Tian Peng , Zhongzheng Mo , Jie Chen , Chenghao Sun , Zhi Wang , Muhammad Shahzad Nazir , Chu Zhang","doi":"10.1016/j.apenergy.2026.127570","DOIUrl":"10.1016/j.apenergy.2026.127570","url":null,"abstract":"<div><div>Accurate prediction of battery remaining useful life (RUL) is crucial for enhancing equipment safety and reliability as well as for sustainable environmental development. This paper proposes a lithium-ion battery RUL prediction model that combines the maximum information coefficient (MIC), the Bayesian optimization algorithm (BOA), kernel density estimation (KDE), and the time-series dense encoder (TiDE). First, using NASA's publicly available lithium-ion battery cycle-life data and the CAL-CE dataset, multiple health indicators including constant-current charge time, discharge time, and IC-curve peak values are extracted and selected via MIC. Next, the TiDE model is employed for accurate RUL prediction, with its key hyperparameters optimized by BOA to boost predictive performance. Finally, KDE is adopted to produce probabilistic RUL forecasts and construct confidence intervals that quantify prediction uncertainty, thereby refining the overall assessment. Comparative experiments demonstrate that the MIC-BOA-TiDE framework reduces the capacity-forecast MAE by 68.9% versus a back-propagation network and by 46.0% versus a GRU baseline, while its RUL error converges to virtually zero, underscoring its superior accuracy and stability. Additionally, the KDE-based interval prediction results show that at the 95% confidence level, the coverage probability on the CS2_35 dataset reaches 95.27% with an average interval width of 0.0731, confirming the model's effectiveness in quantifying predictive uncertainty.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":"410 ","pages":"Article 127570"},"PeriodicalIF":11.0,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147386225","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2026-05-01Epub Date: 2026-02-13DOI: 10.1016/j.apenergy.2026.127544
Felipe T. Muñoz
This study addresses the challenge of designing cost-efficient and resilient biomass procurement strategies by integrating supplier diversification and contractual flexibility into a mathematical optimization framework. We formulate a mixed-integer linear programming (MILP) model that captures key characteristics of the procurement context, including biomass heterogeneity, moisture variability, storage capacity constraints, logistics, and monthly energy demand. The model reflects the operational and contractual conditions faced in the strategic design of biomass supply chains. A multi-objective extension is introduced to analyze trade-offs between minimizing procurement costs and maximizing the number of contracted biomass profiles, which serves as a proxy for supply diversification.
The model is tested through a real-world case involving a centralized boiler facility within a manufacturing firm in the Biobío region of Chile, where thermal energy supports the industrial process. Computational experiments show that moderate diversification incurs minimal additional costs, whereas aggressive diversification results in sharply increasing marginal costs. Moreover, greater contractual flexibility consistently lowers procurement expenses.
To assess trade-offs among cost, supplier diversification, and contractual flexibility, the -constraint method is used to generate Pareto frontiers under varying flexibility scenarios. This approach provides decision-makers with a structured framework to explore feasible procurement strategies across the cost–diversification–flexibility spectrum.
Our results underscore the importance of integrated modeling approaches in striking a balance between economic efficiency and supply chain resilience. The proposed framework provides practical guidance for procurement managers and supports data-driven contract design in biomass-based energy systems. All model files, data sets, and results are publicly available to ensure transparency and reproducibility.
Additionally, we empirically validate the resilience of the resulting procurement plans through post-hoc disruption case studies, quantifying procurement fulfillment under pure and mixed substitution strategies.
{"title":"Biomass procurement and supplier diversification for energy generation: Optimization models and insights","authors":"Felipe T. Muñoz","doi":"10.1016/j.apenergy.2026.127544","DOIUrl":"10.1016/j.apenergy.2026.127544","url":null,"abstract":"<div><div>This study addresses the challenge of designing cost-efficient and resilient biomass procurement strategies by integrating supplier diversification and contractual flexibility into a mathematical optimization framework. We formulate a mixed-integer linear programming (MILP) model that captures key characteristics of the procurement context, including biomass heterogeneity, moisture variability, storage capacity constraints, logistics, and monthly energy demand. The model reflects the operational and contractual conditions faced in the strategic design of biomass supply chains. A multi-objective extension is introduced to analyze trade-offs between minimizing procurement costs and maximizing the number of contracted biomass profiles, which serves as a proxy for supply diversification.</div><div>The model is tested through a real-world case involving a centralized boiler facility within a manufacturing firm in the Biobío region of Chile, where thermal energy supports the industrial process. Computational experiments show that moderate diversification incurs minimal additional costs, whereas aggressive diversification results in sharply increasing marginal costs. Moreover, greater contractual flexibility consistently lowers procurement expenses.</div><div>To assess trade-offs among cost, supplier diversification, and contractual flexibility, the <span><math><mi>ε</mi></math></span>-constraint method is used to generate Pareto frontiers under varying flexibility scenarios. This approach provides decision-makers with a structured framework to explore feasible procurement strategies across the cost–diversification–flexibility spectrum.</div><div>Our results underscore the importance of integrated modeling approaches in striking a balance between economic efficiency and supply chain resilience. The proposed framework provides practical guidance for procurement managers and supports data-driven contract design in biomass-based energy systems. All model files, data sets, and results are publicly available to ensure transparency and reproducibility.</div><div>Additionally, we empirically validate the resilience of the resulting procurement plans through post-hoc disruption case studies, quantifying procurement fulfillment under pure and mixed substitution strategies.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":"410 ","pages":"Article 127544"},"PeriodicalIF":11.0,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147385992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2026-05-01Epub Date: 2026-02-14DOI: 10.1016/j.apenergy.2026.127518
Anni Hu , Gengyin Li , Tiance Zhang , Ming Zhou , Jianxiao Wang
The global transition toward low-carbon transportation is driving rapid growth in electric vehicle (EV) adoption. As millions of EVs intergraded to power grids, their aggregated charging and discharging behaviors present both a challenge and an opportunity. However, turning this potential into reliable, dispatchable services remains highly challenging. EV charging patterns are inherently uncertain and diverse. At the same time, distribution networks impose strict physical constraints that must be respected at all times. These factors make it difficult for electric vehicle aggregators (EVAs) to confidently commit to specific regulation capacities, limiting their ability to participate in electricity markets. To bridge this gap, a probability-guaranteed feasible region (PGFR) framework is proposed in this paper to provide a reliable and quantifiable representation of the EVA's admissible power-exchange range under different confidence levels. The proposed framework employs inverse-function analysis to convert probabilistic uncertainties into tractable deterministic constraints. It then incorporates charging complementarity through McCormick envelope relaxation, enabling an explicit representation of the coupling relationships among EV charging and discharging behaviors. Finally, an outer progressive approximation method is adopted to efficiently handle the high dimensionality and temporal dependence inherent in EVA operation. The PGFR provides a balanced view of operational flexibility, avoiding overly conservative or overly optimistic feasible region descriptions that may otherwise cause economic losses for EVAs or security risks for the power grid. Case studies on a modified IEEE 33-bus distribution system and a 141-bus Venezuelan distribution network verify that the proposed approach provides a reliable and practical tool for EVAs to fully utilize their flexibility potential in supporting future power systems with high renewable energy penetration and significant uncertainty.
{"title":"Enhancing grid balancing services from electric vehicle aggregators under uncertainty: A probability-guaranteed feasible region approach","authors":"Anni Hu , Gengyin Li , Tiance Zhang , Ming Zhou , Jianxiao Wang","doi":"10.1016/j.apenergy.2026.127518","DOIUrl":"10.1016/j.apenergy.2026.127518","url":null,"abstract":"<div><div>The global transition toward low-carbon transportation is driving rapid growth in electric vehicle (EV) adoption. As millions of EVs intergraded to power grids, their aggregated charging and discharging behaviors present both a challenge and an opportunity. However, turning this potential into reliable, dispatchable services remains highly challenging. EV charging patterns are inherently uncertain and diverse. At the same time, distribution networks impose strict physical constraints that must be respected at all times. These factors make it difficult for electric vehicle aggregators (EVAs) to confidently commit to specific regulation capacities, limiting their ability to participate in electricity markets. To bridge this gap, a probability-guaranteed feasible region (PGFR) framework is proposed in this paper to provide a reliable and quantifiable representation of the EVA's admissible power-exchange range under different confidence levels. The proposed framework employs inverse-function analysis to convert probabilistic uncertainties into tractable deterministic constraints. It then incorporates charging complementarity through McCormick envelope relaxation, enabling an explicit representation of the coupling relationships among EV charging and discharging behaviors. Finally, an outer progressive approximation method is adopted to efficiently handle the high dimensionality and temporal dependence inherent in EVA operation. The PGFR provides a balanced view of operational flexibility, avoiding overly conservative or overly optimistic feasible region descriptions that may otherwise cause economic losses for EVAs or security risks for the power grid. Case studies on a modified IEEE 33-bus distribution system and a 141-bus Venezuelan distribution network verify that the proposed approach provides a reliable and practical tool for EVAs to fully utilize their flexibility potential in supporting future power systems with high renewable energy penetration and significant uncertainty.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":"410 ","pages":"Article 127518"},"PeriodicalIF":11.0,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147385974","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2026-05-01Epub Date: 2026-02-17DOI: 10.1016/j.apenergy.2026.127486
Oscar Delbeke , Giacomo Bastianel , Kaan Yurtseven , Hakan Ergun , Jens D. Moschner , Johan Driesen
Floating photovoltaics (PV) are rapidly scaling up solar power beyond on-land PV. Whilst offshore floating PV (OFPV) is still in pilot phase, its combination with offshore wind could enable an efficient common use of costly transmission infrastructure. This work presents a detailed, quantitative case study assessing the integration of offshore floating PV with offshore wind. Through stochastic generation expansion planning, the optimal distribution of OFPV within a representative Dutch offshore wind farm is determined. In the power collection network, OFPV is best connected to the substation, or to the wind turbines electrically nearest to it. To evaluate the economic performance of the hybrid solar-wind system, its electrical integration with the Central Western European grid is simulated. The study reveals that a considerable amount of OFPV can be integrated in a modern offshore wind farm without hindering the transmission of wind power, with the export cables being the main bottleneck in power transfer, followed by the substation transformers and the array cables. However, this is accompanied by a significant amount of OFPV curtailment. As the capacity factors of offshore wind turbines increase, the remaining transmission gap in their connections, which OFPV can utilise without any transmission expansion, narrows. Finally, cost targets are derived for which the integrated offshore solar system would break even in the analysed case, revealing challenging economic prospects. The work identifies opportunities for hybrid offshore solar-wind farms and highlights key technical and economic challenges to be addressed.
{"title":"Hybrid offshore solar-wind farms: the potential of integrating floating photovoltaics with offshore wind","authors":"Oscar Delbeke , Giacomo Bastianel , Kaan Yurtseven , Hakan Ergun , Jens D. Moschner , Johan Driesen","doi":"10.1016/j.apenergy.2026.127486","DOIUrl":"10.1016/j.apenergy.2026.127486","url":null,"abstract":"<div><div>Floating photovoltaics (PV) are rapidly scaling up solar power beyond on-land PV. Whilst offshore floating PV (OFPV) is still in pilot phase, its combination with offshore wind could enable an efficient common use of costly transmission infrastructure. This work presents a detailed, quantitative case study assessing the integration of offshore floating PV with offshore wind. Through stochastic generation expansion planning, the optimal distribution of OFPV within a representative Dutch offshore wind farm is determined. In the power collection network, OFPV is best connected to the substation, or to the wind turbines electrically nearest to it. To evaluate the economic performance of the hybrid solar-wind system, its electrical integration with the Central Western European grid is simulated. The study reveals that a considerable amount of OFPV can be integrated in a modern offshore wind farm without hindering the transmission of wind power, with the export cables being the main bottleneck in power transfer, followed by the substation transformers and the array cables. However, this is accompanied by a significant amount of OFPV curtailment. As the capacity factors of offshore wind turbines increase, the remaining transmission gap in their connections, which OFPV can utilise without any transmission expansion, narrows. Finally, cost targets are derived for which the integrated offshore solar system would break even in the analysed case, revealing challenging economic prospects. The work identifies opportunities for hybrid offshore solar-wind farms and highlights key technical and economic challenges to be addressed.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":"410 ","pages":"Article 127486"},"PeriodicalIF":11.0,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147386029","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2026-05-01Epub Date: 2026-02-17DOI: 10.1016/j.apenergy.2026.127507
Yun-Jia Deng , Jiang Huang , Sheng-Hua Xiong , Zhen-Song Chen , Muhammet Deveci
Accurate estimation of the State of Charge (SOC) is essential for enhancing the efficiency and reliability of Battery Management Systems (BMS) in Internet of Things (IoT) applications. This study introduces the Pattern-Aware Transformer Model (PATM), an interpretable framework for SOC prediction in Float-Nominal (FN), Constant-Current (CC), and Energy Release (ER) scenarios. PATM extends the standard Transformer architecture by incorporating a pattern embedding mechanism that explicitly encodes operating conditions and directs adaptive attention allocation. A feature engineering pipeline that combines mutual information (MI) ranking and principal component analysis (PCA) reduces dimensionality while preserving physically relevant variables. On real-world data, PATM achieves an RMSE of 2.08 10−3 and an of 0.9998, outperforming the Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) baselines. Compared with single-scenario CC modeling, multi-scenario learning reduces RMSE, MAE, MSE, and MAPE by 54.9%, 80.1%, 79.6%, and 75.9%, respectively. Ablation studies further demonstrate that removing the embedding module increases RMSE by 2.4%, MAE by 17.8%, and MSE by 4.9%, while leaving nearly unchanged. This indicates that the embedding mechanism enhances cross-scenario robustness and error stability. SHapley Additive exPlanations (SHAP) analysis and attention visualizations reveal the model’s dependence on physically relevant factors, including temperature gradients, voltage fluctuations, and internal resistance.
{"title":"Pattern-aware transformer for SOC estimation in IoT-based battery management systems: Toward energy-efficient and interpretable modeling","authors":"Yun-Jia Deng , Jiang Huang , Sheng-Hua Xiong , Zhen-Song Chen , Muhammet Deveci","doi":"10.1016/j.apenergy.2026.127507","DOIUrl":"10.1016/j.apenergy.2026.127507","url":null,"abstract":"<div><div>Accurate estimation of the State of Charge (SOC) is essential for enhancing the efficiency and reliability of Battery Management Systems (BMS) in Internet of Things (IoT) applications. This study introduces the Pattern-Aware Transformer Model (PATM), an interpretable framework for SOC prediction in Float-Nominal (FN), Constant-Current (CC), and Energy Release (ER) scenarios. PATM extends the standard Transformer architecture by incorporating a pattern embedding mechanism that explicitly encodes operating conditions and directs adaptive attention allocation. A feature engineering pipeline that combines mutual information (MI) ranking and principal component analysis (PCA) reduces dimensionality while preserving physically relevant variables. On real-world data, PATM achieves an RMSE of 2.08 <span><math><mo>×</mo></math></span> 10<sup>−3</sup> and an <span><math><msup><mi>R</mi><mn>2</mn></msup></math></span> of 0.9998, outperforming the Long Short-Term Memory (LSTM) and Gated Recurrent Unit (GRU) baselines. Compared with single-scenario CC modeling, multi-scenario learning reduces RMSE, MAE, MSE, and MAPE by 54.9%, 80.1%, 79.6%, and 75.9%, respectively. Ablation studies further demonstrate that removing the embedding module increases RMSE by 2.4%, MAE by 17.8%, and MSE by 4.9%, while leaving <span><math><msup><mi>R</mi><mn>2</mn></msup></math></span> nearly unchanged. This indicates that the embedding mechanism enhances cross-scenario robustness and error stability. SHapley Additive exPlanations (SHAP) analysis and attention visualizations reveal the model’s dependence on physically relevant factors, including temperature gradients, voltage fluctuations, and internal resistance.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":"410 ","pages":"Article 127507"},"PeriodicalIF":11.0,"publicationDate":"2026-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147386062","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2026-04-15Epub Date: 2026-02-04DOI: 10.1016/j.apenergy.2026.127466
Yili Tang , Quan Li , Xiaochao Zuo , Huaming Yang
Composite phase change materials (CPCMs) are progressively replacing conventional phase change materials in latent heat storage technology due to their superior stability. In recent years, finite element analysis (FEA) has advanced significantly in the applied studies of CPCMs. As a numerical simulation technique, FEA facilitates the development of approximate models to address the complex geometric structures and multi-physics coupling challenges within CPCMs, establishing itself as an ideal tool for predicting and optimizing the performance of CPCMs. However, to date, the lack of comprehensive reviews evaluating the importance of FEA in the design and application of CPCMs remains. This review addresses this gap through an examination of current research and practices. It begins by discussing the relevant heat transfer modeling and numerical research progress of FEA in the thermal property enhancement mechanisms of various CPCMs. Then it summarizes the application of FEA to phase change components in different energy storage applications. Finally, the challenges and opportunities for the future development of FEA in the thermal property enhancement analysis of CPCMs are outlined. This review helps researchers better utilize FEA to assess the enhancement of thermal properties in CPCMs, thereby identifying possible future research directions.
{"title":"Finite element analysis of composites for latent heat storage technology: a comprehensive review","authors":"Yili Tang , Quan Li , Xiaochao Zuo , Huaming Yang","doi":"10.1016/j.apenergy.2026.127466","DOIUrl":"10.1016/j.apenergy.2026.127466","url":null,"abstract":"<div><div>Composite phase change materials (CPCMs) are progressively replacing conventional phase change materials in latent heat storage technology due to their superior stability. In recent years, finite element analysis (FEA) has advanced significantly in the applied studies of CPCMs. As a numerical simulation technique, FEA facilitates the development of approximate models to address the complex geometric structures and multi-physics coupling challenges within CPCMs, establishing itself as an ideal tool for predicting and optimizing the performance of CPCMs. However, to date, the lack of comprehensive reviews evaluating the importance of FEA in the design and application of CPCMs remains. This review addresses this gap through an examination of current research and practices. It begins by discussing the relevant heat transfer modeling and numerical research progress of FEA in the thermal property enhancement mechanisms of various CPCMs. Then it summarizes the application of FEA to phase change components in different energy storage applications. Finally, the challenges and opportunities for the future development of FEA in the thermal property enhancement analysis of CPCMs are outlined. This review helps researchers better utilize FEA to assess the enhancement of thermal properties in CPCMs, thereby identifying possible future research directions.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":"409 ","pages":"Article 127466"},"PeriodicalIF":11.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146170938","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2026-04-15Epub Date: 2026-02-06DOI: 10.1016/j.apenergy.2026.127489
Qiang Gao, Yongping Huang, Chengbin Zhang
As electric vehicles advance, maintaining the lithium-ion batteries’ performance and safety has become more crucial. To confront this challenge, a liquid immersion battery cooling system employing flow guides with fish-shaped holes based on an innovative pulse control technique is developed. The battery modules’ electro-thermal performance and overall heat transfer performance are investigated through both experimental and numerical approaches. The effects of different flow guides and pulse control methods on the temperature variation, voltage equalization and pumping cost under different operating conditions are analyzed systematically. Compared to other flow guide designs, the battery module using flow guides with fish-shaped holes exhibits better cooling performance and electro-thermal equalization behavior, with maximum reductions in maximum temperature, maximum temperature difference, pumping cost and voltage deviation of 4.9%, 8.9%, 48.8% and 10.3%, respectively. Additionally, the liquid immersion battery cooling system employing the multi-inlet coordinated staggered pulse control method has an advantage in temperature regulation and voltage equalization over the traditional synchronous pulse control method, particularly at a 50% output ratio. Under an equivalent average flow rate, the multi-inlet coordinated staggered pulse control method not only improves thermal stability and cooling performance but also enhances the battery pack’s equalization performance and overall heat transfer performance, especially at a 25% output ratio. Moreover, $1 per m3 invested generates approximately 15.7 W for the proposed battery thermal management system, while for the Tesla Model S, it generates about 15 W. Compared with the Tesla Model S, the proposed BTMS has better compactness and cost-effectiveness.
{"title":"Improving the electro-thermal performance of battery module using a pulse liquid immersion cooling strategy","authors":"Qiang Gao, Yongping Huang, Chengbin Zhang","doi":"10.1016/j.apenergy.2026.127489","DOIUrl":"10.1016/j.apenergy.2026.127489","url":null,"abstract":"<div><div>As electric vehicles advance, maintaining the lithium-ion batteries’ performance and safety has become more crucial. To confront this challenge, a liquid immersion battery cooling system employing flow guides with fish-shaped holes based on an innovative pulse control technique is developed. The battery modules’ electro-thermal performance and overall heat transfer performance are investigated through both experimental and numerical approaches. The effects of different flow guides and pulse control methods on the temperature variation, voltage equalization and pumping cost under different operating conditions are analyzed systematically. Compared to other flow guide designs, the battery module using flow guides with fish-shaped holes exhibits better cooling performance and electro-thermal equalization behavior, with maximum reductions in maximum temperature, maximum temperature difference, pumping cost and voltage deviation of 4.9%, 8.9%, 48.8% and 10.3%, respectively. Additionally, the liquid immersion battery cooling system employing the multi-inlet coordinated staggered pulse control method has an advantage in temperature regulation and voltage equalization over the traditional synchronous pulse control method, particularly at a 50% output ratio. Under an equivalent average flow rate, the multi-inlet coordinated staggered pulse control method not only improves thermal stability and cooling performance but also enhances the battery pack’s equalization performance and overall heat transfer performance, especially at a 25% output ratio. Moreover, $1 per m<sup>3</sup> invested generates approximately 15.7 W for the proposed battery thermal management system, while for the Tesla Model S, it generates about 15 W. Compared with the Tesla Model S, the proposed BTMS has better compactness and cost-effectiveness.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":"409 ","pages":"Article 127489"},"PeriodicalIF":11.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146171491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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 : 2026-04-15Epub Date: 2026-01-28DOI: 10.1016/j.apenergy.2026.127432
Kunxiang Liu , Bo Liu , Yu Wang , Haijiang Wang , Jun Yang , Chen Zhao
In the global energy transition, hydrogen fuel cells have drawn a lot of attention as a clean energy source. Developing new energy vehicles that rely on hydrogen fuel cells as their primary power source is crucial to reaching net-zero carbon emissions. As the central component and key to the overall operation of new energy vehicles, fuel cell energy management (FCEM) is crucial, particularly for enhancing durability and fuel economy. However, the literature screening process in existing bibliometric studies is often opaque and lacks publicly available criteria, leading to irreproducible findings. To address this, we propose a transparent and reproducible bibliometric framework that integrates an enhanced Word2Vec model for systematic literature screening. Our AI-driven screening method, based on calculating the similarity of titles, abstracts, and keywords, is validated by achieving 91.4751% alignment with the Web of Science (WOS) relevance ranking, offering a quantifiable and automated alternative to opaque screening processes. Using this framework, we systematically analyze the characteristics of FCEMS-related scholarship in terms of publication journals, country geographic distribution, institutional collaborations, author collaborations, and keyword co-occurrence frequencies. The analysis reveals a pattern of policy-associated growth: post-2015, China contributes to 45% of the global FCEM literature, likely benefiting from the national hydrogen energy strategy. Furthermore, we detail FCEMS strategies including rule-based, optimization-based, and learning-based approaches, summarize their research progress in applications such as vehicles, aircraft, and ships, and analyze future research trends from multiple perspectives. This work represents the first integration of bibliometrics with natural language processing (NLP) for algorithmic literature screening, and its inaugural application in the FCEMS domain.
在全球能源转型中，氢燃料电池作为一种清洁能源备受关注。开发以氢燃料电池为主要动力源的新能源汽车对于实现净零碳排放至关重要。燃料电池能量管理（FCEM）作为新能源汽车整体运行的核心部件和关键，对于提高耐久性和燃油经济性至关重要。然而，现有文献计量学研究中的文献筛选过程往往是不透明的，缺乏可公开获得的标准，导致不可重复的发现。为了解决这个问题，我们提出了一个透明和可重复的文献计量框架，该框架集成了一个增强的Word2Vec模型，用于系统的文献筛选。我们的人工智能驱动的筛选方法基于计算标题、摘要和关键词的相似度，与Web of Science （WOS）相关排名的一致性达到91.4751%，为不透明的筛选过程提供了可量化和自动化的替代方案。在此框架下，我们从发表期刊、国家地理分布、机构合作、作者合作和关键词共现频率等方面系统分析了fcems相关学术研究的特征。分析揭示了一种与政策相关的增长模式：2015年后，中国贡献了全球45%的氢能源文献，可能受益于国家氢能战略。在此基础上，详细介绍了基于规则的、基于优化的和基于学习的FCEMS策略，总结了它们在车辆、飞机和船舶等领域的研究进展，并从多个角度分析了未来的研究趋势。这项工作代表了文献计量学与自然语言处理（NLP）在算法文献筛选中的首次整合，以及它在FCEMS领域的首次应用。
{"title":"Fuel cell energy management strategies (FCEMS): a Word2Vec-driven bibliometric framework for trend mapping and algorithmic advancements","authors":"Kunxiang Liu , Bo Liu , Yu Wang , Haijiang Wang , Jun Yang , Chen Zhao","doi":"10.1016/j.apenergy.2026.127432","DOIUrl":"10.1016/j.apenergy.2026.127432","url":null,"abstract":"<div><div>In the global energy transition, hydrogen fuel cells have drawn a lot of attention as a clean energy source. Developing new energy vehicles that rely on hydrogen fuel cells as their primary power source is crucial to reaching net-zero carbon emissions. As the central component and key to the overall operation of new energy vehicles, fuel cell energy management (FCEM) is crucial, particularly for enhancing durability and fuel economy. However, the literature screening process in existing bibliometric studies is often opaque and lacks publicly available criteria, leading to irreproducible findings. To address this, we propose a transparent and reproducible bibliometric framework that integrates an enhanced Word2Vec model for systematic literature screening. Our AI-driven screening method, based on calculating the similarity of titles, abstracts, and keywords, is validated by achieving 91.4751% alignment with the Web of Science (WOS) relevance ranking, offering a quantifiable and automated alternative to opaque screening processes. Using this framework, we systematically analyze the characteristics of FCEMS-related scholarship in terms of publication journals, country geographic distribution, institutional collaborations, author collaborations, and keyword co-occurrence frequencies. The analysis reveals a pattern of policy-associated growth: post-2015, China contributes to 45% of the global FCEM literature, likely benefiting from the national hydrogen energy strategy. Furthermore, we detail FCEMS strategies including rule-based, optimization-based, and learning-based approaches, summarize their research progress in applications such as vehicles, aircraft, and ships, and analyze future research trends from multiple perspectives. This work represents the first integration of bibliometrics with natural language processing (NLP) for algorithmic literature screening, and its inaugural application in the FCEMS domain.</div></div>","PeriodicalId":246,"journal":{"name":"Applied Energy","volume":"409 ","pages":"Article 127432"},"PeriodicalIF":11.0,"publicationDate":"2026-04-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146049141","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"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