Energy and AI最新文献

{"title":"An AI-based fuel designer tool for sustainable E-fuel development: Methodology, Validation, and Optimization","authors":"Márton Virt ,&nbsp;Máté Zöldy","doi":"10.1016/j.egyai.2025.100583","DOIUrl":"10.1016/j.egyai.2025.100583","url":null,"abstract":"<div><div>The carbon neutrality of existing internal combustion engines can be significantly enhanced through the use of sustainable e-fuels; thus, their price has to be reduced. Artificial intelligence (AI) offers a promising pathway to streamline and accelerate fuel development by enabling faster and more efficient model creation compared to conventional physicochemical simulations. Despite the apparent advantages, state-of-the-art research typically limits the application of AI to basic predictions within narrow operating ranges. This study introduces a novel AI-based fuel design tool capable of accurately predicting detailed engine performance across a broad range of operating conditions, using comprehensive physicochemical fuel properties as input. The proposed approach provides greater detail and precision than existing state-of-the-art methods. Building on a cost-efficient AI development strategy established in our previous work, the tool was constructed using 17 single-output multilayer perceptron networks. The tool was validated using engine dynamometer measurements with various test fuels, and then it was applied to a fuel optimization task to demonstrate its effectiveness. The results indicate that the tool's predictions closely match actual engine performance. Specifically, 10 out of the 17 models achieved a mean absolute percentage error of &lt;3 %. In the optimization scenario, the optimized fuel had a predicted engine operating score of 40.51 %, while the actual score was 41.3 %, demonstrating the tool’s potential for accurate fuel design. Thus, this novel approach can support the development of low-cost e-fuels, enabling economically viable, carbon-neutral mobility across a wide range of transport applications.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100583"},"PeriodicalIF":9.6,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144781615","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}

{"title":"Transient thermomechanical response of an electrolyte supported planar solid oxide electrolysis cell under dynamic loading conditions","authors":"Tiancheng Cui ,&nbsp;Lige Zhang ,&nbsp;Huijuan Yan ,&nbsp;Yuandalei Cao ,&nbsp;Ruizhu Li ,&nbsp;Zhibin Yang ,&nbsp;Jian-Qiang Wang ,&nbsp;Guoping Xiao","doi":"10.1016/j.egyai.2025.100590","DOIUrl":"10.1016/j.egyai.2025.100590","url":null,"abstract":"<div><div>Solid oxide electrolysis cells (SOECs) offer high efficiency, noble-metal-free construction, and broad power adaptability, making them promising for large-scale green hydrogen production integrated with renewable energy. However, the intermittency of renewable energy induces spatial and temporal gradients in temperature and electrochemical fields of the cell, leading to stress concentrations and potential mechanical failure. Understanding the transient thermomechanical response during dynamic loading of a SOEC is thus crucial for evaluating the electrochemical performance and predicting cell lifetime. While previous studies focused on developing steady-state models to investigate the thermal stress in the cell, we develop a transient multi-physics model to capture the coupled thermo–electro–chemo–mechanical behavior of an electrolyte-supported planar SOEC under dynamic operating conditions. Numerical simulations are performed with varying power control strategies, heating rates and water vapor molar fractions. Results reveal that the electrolyte layer consistently experiences the highest average maximum principal stress and failure risk. Stepped current density control induces a higher heating rate and thermal stress compared to stepped voltage control. Additionally, a slower water molar fraction decline reduces electrochemical reaction and heat absorption rates, leading to a slower temperature rise and reduced thermal stress. To enable rapid and accurate prediction of critical thermomechanical responses, a Random Forest Regression model is trained on simulation data using gas inlet heating rate as input. The model accurately predicts temperature and stress in the electrolyte layer and demonstrates strong generalization on unseen scenarios. The integration of high-fidelity physics-based modeling and machine learning provides a foundation for intelligent SOEC control and real-time optimization, enhancing system reliability and extending operational lifetime under renewable energy operation.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100590"},"PeriodicalIF":9.6,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144829466","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}

{"title":"Optimization framework for energy-efficient and uniform jet impingement cooling for heterogeneous integration packaging","authors":"Hyunho Cho ,&nbsp;Insik Lee ,&nbsp;Seungwoo Kim ,&nbsp;Soosik Bang ,&nbsp;Jaechoon Kim ,&nbsp;Youngsuk Nam","doi":"10.1016/j.egyai.2025.100587","DOIUrl":"10.1016/j.egyai.2025.100587","url":null,"abstract":"<div><div>The imbalance in heat power generated by various types of chips poses an obstacle to the reliability and performance of heterogeneous integration (HI) packaging technology, leading to excessive cooling that reduces the system's energy efficiency. We propose a framework to optimize the impinging nozzle arrangement for energy-efficient uniform jet cooling of HI packages. This framework utilizes a convolutional neural network (CNN)-based surrogate model that learns nozzle arrangements and heating scenarios to predict the temperature non-uniformity of the package. The potential optimal designs predicted by the CNN are used for re-training through an experimentally validated numerical analysis model. Combined with this active learning approach, the proposed hierarchical exploration algorithm accelerates optimization by gradually scaling the design options. The optimization results showed an increase in cooling uniformity by up to 39.5 %, while the cooling COP improved by up to 200 % across the investigated flow rate range (3–8 L/min). The optimized designs were experimentally validated with a maximum error of 4.34 % in average thermal resistance. Our framework achieved up to 45.7 % data savings compared to the random sampling-based approach. Along with a discussion on applying the CNN model to untrained conditions to further enhance optimization efficiency, our work represents a novel approach to broadly address the rapidly evolving diverse heating scenarios of HI, contributing to improved cooling energy efficiency in data centers and enhanced reliability of high-performance processors.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100587"},"PeriodicalIF":9.6,"publicationDate":"2025-08-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144851972","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}

{"title":"Conformal Prediction for electricity price forecasting in the day-ahead and real-time balancing market","authors":"Ciaran O’Connor ,&nbsp;Mohamed Bahloul ,&nbsp;Roberto Rossi ,&nbsp;Steven Prestwich ,&nbsp;Andrea Visentin","doi":"10.1016/j.egyai.2025.100571","DOIUrl":"10.1016/j.egyai.2025.100571","url":null,"abstract":"<div><div>The integration of renewable energy into electricity markets poses significant challenges to price stability and increases the complexity of market operations. Accurate and reliable electricity price forecasting is crucial for effective market participation, where price dynamics can be significantly more challenging to predict. Probabilistic forecasting, through prediction intervals, efficiently quantifies the inherent uncertainties in electricity prices, supporting better decision-making for market participants. This study explores the enhancement of probabilistic price prediction using Conformal Prediction (CP) techniques, specifically Ensemble Batch Prediction Intervals and Sequential Predictive Conformal Inference. These methods provide precise and reliable prediction intervals, outperforming traditional models in validity metrics. We propose an ensemble approach that combines the efficiency of quantile regression models with the robust coverage properties of time series adapted CP techniques. This ensemble delivers both narrow prediction intervals and high coverage, leading to more reliable and accurate forecasts. We further evaluate the practical implications of CP techniques through a simulated trading algorithm applied to a battery storage system. The ensemble approach demonstrates improved financial returns in energy trading in both the Day-Ahead and Balancing Markets, highlighting its practical benefits for market participants.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100571"},"PeriodicalIF":9.6,"publicationDate":"2025-08-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144766607","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}

{"title":"D2: An LLM agent driving end-to-end visual AI modeling in energy platforms","authors":"Yu Li ,&nbsp;Qiaoqiao Zhao ,&nbsp;Min Hou ,&nbsp;Quansheng Bai ,&nbsp;Xiyan Zou ,&nbsp;Changle Xie ,&nbsp;Chang Shu ,&nbsp;Boyang Ma ,&nbsp;Zhijin Li","doi":"10.1016/j.egyai.2025.100582","DOIUrl":"10.1016/j.egyai.2025.100582","url":null,"abstract":"<div><div>This study presents X-AI, a domain-native, agent-driven, and end-to-end modeling platform developed to support digital transformation in the energy sector. X-AI integrates advanced Machine Learning (ML) and Deep Learning (DL) capabilities into a workflow-driven environment that enables energy engineers to construct and deploy predictive models without prior AI expertise. A key innovation is the introduction of Dragon Dawn (D2), an intelligent agent powered by Large Language Models (LLMs) and agent-based reasoning. D2 interprets natural language instructions, retrieves domain-relevant knowledge, orchestrates modeling workflows, and guides multi-step optimization processes, thereby lowering technical barriers and cognitive load for users. To quantitatively evaluate platform usability, a novel metric termed Cognitive-Operation Efficiency Ratio (COER) is proposed, capturing both task efficiency and cognitive effort. Experimental evaluation shows that D2 significantly enhances modeling productivity, with over eightfold improvement in COER. A real-world case study on inflow forecasting in cascade hydropower systems validates the platform’s capabilities. By comparing LSTM and D2-assisted XGBoost models, the study demonstrates how the agent facilitates iterative reasoning, feature enhancement, and hyperparameter tuning. These findings establish X-AI as a practical, scalable AI solution for accelerating intelligent decision-making in the energy domain.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100582"},"PeriodicalIF":9.6,"publicationDate":"2025-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144771902","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}

{"title":"Augmented robustness in home demand prediction: Integrating statistical loss function with enhanced cross-validation in machine learning hyperparameter optimisation","authors":"Banafshe Parizad,&nbsp;Ali Jamali,&nbsp;Hamid Khayyam","doi":"10.1016/j.egyai.2025.100584","DOIUrl":"10.1016/j.egyai.2025.100584","url":null,"abstract":"<div><div>Sustainable forecasting of home energy demand (SFHED) is crucial for promoting energy efficiency, minimizing environmental impact, and optimizing resource allocation. Machine learning (ML) supports SFHED by identifying patterns and forecasting demand. However, conventional hyperparameter tuning methods often rely solely on minimizing average prediction errors, typically through fixed k-fold cross-validation, which overlooks error variability and limits model robustness. To address this limitation, we propose the Optimized Robust Hyperparameter Tuning for Machine Learning with Enhanced Multi-fold Cross-Validation (ORHT-ML-EMCV) framework. This method integrates statistical analysis of k-fold validation errors by incorporating their mean and variance into the optimization objective, enhancing robustness and generalizability. A weighting factor is introduced to balance accuracy and robustness, and its impact is evaluated across a range of values. A novel Enhanced Multi-Fold Cross-Validation (EMCV) technique is employed to automatically evaluate model performance across varying fold configurations without requiring a predefined k value, thereby reducing sensitivity to data splits. Using three evolutionary algorithms Genetic Algorithm (GA), Particle Swarm Optimization (PSO), and Differential Evolution (DE) we optimize two ensemble models: XGBoost and LightGBM. The optimization process minimizes both mean error and variance, with robustness assessed through cumulative distribution function (CDF) analyses. Experiments on three real-world residential datasets show the proposed method reduces worst-case Root Mean Square Error (RMSE) by up to 19.8% and narrows confidence intervals by up to 25%. Cross-household validations confirm strong generalization, achieving coefficient of determination (R²) of 0.946 and 0.972 on unseen homes. The framework offers a statistically grounded and efficient solution for robust energy forecasting.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100584"},"PeriodicalIF":9.6,"publicationDate":"2025-07-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144826602","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}

{"title":"Understanding the role of autoencoders for stiff dynamical systems using information theory","authors":"Vijayamanikandan Vijayarangan ,&nbsp;Harshavardhana A. Uranakara ,&nbsp;Francisco E. Hernández–Pérez ,&nbsp;Hong G. Im","doi":"10.1016/j.egyai.2025.100567","DOIUrl":"10.1016/j.egyai.2025.100567","url":null,"abstract":"<div><div>Using information theory, this study provides insights into how the construction of latent space of autoencoder (AE) using deep neural network (DNN) training finds a smooth (non-stiff) low-dimensional manifold in the stiff dynamical system. Our recent study (Vijayarangan et al. 2023) reported that an AE combined with neural ODE (NODE) as a surrogate reduced order model (ROM) for the integration of stiff chemically reacting systems led to a significant reduction in the temporal stiffness, and the behavior was attributed to the identification of a slow invariant manifold by the nonlinear projection using the AE. The present work offers a fundamental understanding of the mechanism of formation of a non-stiff latent space and stiffness reduction by employing concepts from information theory and better mixing. The learning mechanisms of both the encoder and the decoder are explained by plotting the evolution of mutual information and identifying two different phases. Subsequently, the density distribution is plotted for the physical and latent variables, which shows the transformation of the <em>rare event</em> in the physical space to a <em>highly likely</em> (more probable) event in the latent space provided by the nonlinear autoencoder. Finally, the nonlinear transformation leading to density redistribution is explained using concepts from information theory and probability.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100567"},"PeriodicalIF":9.6,"publicationDate":"2025-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144781616","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}

{"title":"Data-driven modeling of polymer electrolyte fuel cells: Towards predictive analytics with explainable artificial intelligence","authors":"Ali Malek ,&nbsp;Max Dreger ,&nbsp;Nima Shaigan ,&nbsp;Chaojie Song ,&nbsp;Kourosh Malek ,&nbsp;Jasna Jankovic ,&nbsp;Michael Eikerling","doi":"10.1016/j.egyai.2025.100577","DOIUrl":"10.1016/j.egyai.2025.100577","url":null,"abstract":"<div><div>Polymer electrolyte fuel cells will be an essential technology of the emerging hydrogen economy. However, optimizing their cost and performance necessitates understanding of how different parameters affect their operation. This optimization problem involves numerous interrelated design and operational parameters. However, developing the required understanding through experimental studies alone would be inefficient. Physical modelling is a much-needed complement to experiment but is constrained by simplifying assumptions that diminish the models' predictive capabilities. As a supplement to experiment and physical modelling, we employ a data-based assessment that leverages machine learning techniques to support and enhance decision-making. We first evaluate the predictive accuracy of various machine learning models, including artificial neural networks, to predict the polarization behavior of polymer electrolyte fuel cells, harnessing an extensive experimental dataset. We then apply explainable artificial intelligence techniques, including Gini feature importance and Shapley additive explanations value analyses, to understand how these models incorporate data into the prediction process. Probabilistic analyses can help identify relationships between predictions and feature values. We demonstrate that insights derived from Shapley additive explanations value analysis are consistent with literature data on the thermodynamics and kinetics of relevant electrochemical reaction and transport processes. Our study highlights the potential of interpretable and explainable tools to offer a holistic analysis of the impacts of various interrelated operational and design parameters on the performance of the fuel cell. In the future, such explainable tools could help identify gaps in experimental data and pinpoint research priorities.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100577"},"PeriodicalIF":9.6,"publicationDate":"2025-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144781608","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}

{"title":"A multi-factor collaborative electricity load forecasting method based on feature importance and multi-scale feature extraction","authors":"Qiao Yan ,&nbsp;Wenpeng Cao ,&nbsp;Yi Yan ,&nbsp;Chengdong Li ,&nbsp;Chongyi Tian ,&nbsp;Wen Kong","doi":"10.1016/j.egyai.2025.100579","DOIUrl":"10.1016/j.egyai.2025.100579","url":null,"abstract":"<div><div>In power systems, environmental fluctuations and electricity price volatility introduce uncertainties in user energy consumption behaviors, posing significant challenges to reliable energy planning. Existing studies often overlook the coupled relationships between the importance and correlations of multiple complex variables, lack consideration of the weighting and distribution of multi-dimensional features across multi-scale spaces, and fall short in multi-scale extraction and fusion of complex spatiotemporal characteristics. To address these issues, this paper proposes a multi-factor collaborative load forecasting method based on feature importance and multi-scale feature extraction. First, a novel evaluation model integrating feature importance and correlation is developed, and a comprehensive feature importance assessment method is proposed. Then, a multi-dimensional weighting extraction framework is designed, from which a multi-dimensional weight matrix and its multi-layer input structure are constructed. Finally, a multi-scale fusion model driven by a multi-channel convolutional neural network is developed. The backbone network is a multi-channel convolutional structure, consisting of a multi-level feature extraction module in the front, a multi-scale sampling mechanism in the middle, and a multi-scale feature fusion architecture in the rear. Based on the proposed comprehensive feature importance assessment method, a multi-factor collaborative load forecasting model is established, achieving accurate load prediction. Experimental results demonstrate that, compared with various state-of-the-art forecasting models, the proposed method reduces Mean Absolute Error (MAE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE) by up to 28.30 %, 24.14 %, and 30.35 %, respectively.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100579"},"PeriodicalIF":9.6,"publicationDate":"2025-07-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144851973","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}

{"title":"Integrating artificial intelligence into energy management: A case study on energy consumption data analysis and forecasting in a German manufacturing company","authors":"Marius Wigger,&nbsp;Peter Burggräf,&nbsp;Fabian Steinberg,&nbsp;Alexander Becher,&nbsp;Benjamin Heinbach","doi":"10.1016/j.egyai.2025.100576","DOIUrl":"10.1016/j.egyai.2025.100576","url":null,"abstract":"<div><div>The pressing need to enhance energy efficiency, as outlined by the United Nations within its Sustainable Development Goals, underscores the importance of reducing energy consumption in manufacturing companies. Energy management systems are essential in achieving this goal by systematically identifying inefficiencies and uncovering potential savings through the analysis of energy consumption data. With the growing utilisation of Artificial Intelligence (AI), there is significant potential to leverage advanced data analytics to predict future energy consumption and detect anomalies. However, current research focus on the theoretical development and refinement of AI models, a practical integration of AI into energy management systems within manufacturing companies remains limited, particularly in small and medium-sized enterprises (SMEs). This case study introduces a proof-of-concept implemented in a German manufacturing company to demonstrate how AI models can be integrated into energy management systems using existing data resources. Utilising the Cross-Industry Standard Process for Data Mining (CRISP-DM) framework, an AI-based evaluation algorithm was developed to detect changes in energy consumption patterns and predict future consumption. The findings reveal that AI models such as Long short-term memory enable the prediction of future energy consumption with remarkable accuracy as well as the identification of deviations that traditional systems might overlook. This study emphasises the transformative capacity of AI-driven energy management systems in enhancing operational efficiency and facilitating compliance with ISO 50001 standards. It provides a practical approach for broader adoption of intelligent data analytics in energy management, particularly for SMEs, aiming to pave the way towards a more sustainable industrial sector.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100576"},"PeriodicalIF":9.6,"publicationDate":"2025-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144757546","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}