Energy and AI最新文献

{"title":"Distributed nonlinear model predictive control for building energy systems: An ALADIN implementation study","authors":"Steffen Eser,&nbsp;Ben Spoek,&nbsp;Augustinus Schütz,&nbsp;Phillip Stoffel,&nbsp;Dirk Müller","doi":"10.1016/j.egyai.2025.100536","DOIUrl":"10.1016/j.egyai.2025.100536","url":null,"abstract":"<div><div>The implementation of sophisticated control strategies for building energy systems is crucial for improving energy efficiency and occupant comfort. While nonlinear model predictive control offers promising benefits, its application to large-scale building systems remains challenging due to computational complexity and system coupling. This work presents a comprehensive study of Nonlinear Distributed Model Predictive Control (NDMPC) implementation for building energy systems, comparing Alternating Direction Method of Multipliers (ADMM) and Augmented Lagrangian Alternating Direction Inexact Newton (ALADIN) algorithms alongside different modeling approaches. We examine a multi-zone heating system with thermal storage and multiple producers, investigating both Ordinary Differential Equation (ODE)-based and Artificial Neural Network (ANN) based modeling strategies. Through systematic parameter tuning using Bayesian optimization and closed-loop scaling analysis with up to 40 thermal zones, we demonstrate that ALADIN-based NDMPC can achieve performance comparable to centralized model predictive control, showing greater robustness to parameter variations than ADMM. Our results reveal that ANN-based models effectively mitigate distributed integration errors and significantly reduce computation time compared to ODE-based approaches. Detailed computational profiling identifies specific bottlenecks in different NDMPC components. These findings advance the practical implementation of NDMPC in building energy systems, offering concrete strategies for modeling choices, parameter tuning, and system architecture design.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100536"},"PeriodicalIF":9.6,"publicationDate":"2025-06-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144338859","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":"Fast data augmentation for battery degradation prediction","authors":"Weihan Li ,&nbsp;Harshvardhan Samsukha ,&nbsp;Bruis van Vlijmen ,&nbsp;Lisen Yan ,&nbsp;Samuel Greenbank ,&nbsp;Simona Onori ,&nbsp;Venkat Viswanathan","doi":"10.1016/j.egyai.2025.100542","DOIUrl":"10.1016/j.egyai.2025.100542","url":null,"abstract":"<div><div>Degradation prediction for lithium-ion batteries using data-driven methods requires high-quality aging data. However, generating such data, whether in the laboratory or the field, is time- and resource-intensive. Here, we propose a method for the synthetic generation of capacity fade curves based on limited battery tests or operation data without the need for invasive battery characterization, aiming to augment the datasets used by data-driven models for degradation prediction. We validate our method by evaluating the performance of both shallow and deep learning models using diverse datasets from laboratory and field applications. These datasets encompass various chemistries and realistic conditions, including cell-to-cell variations, measurement noise, varying charge-discharge conditions, and capacity recovery. Our results show that it is possible to reduce cell-testing efforts by at least 50 % by substituting synthetic data into an existing dataset. This paper highlights the effectiveness of our synthetic data augmentation method in supplementing existing methodologies in battery health prognostics while dramatically reducing the expenditure of time and resources on battery aging experiments.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100542"},"PeriodicalIF":9.6,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144364418","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":"Continuous variable quantum reinforcement learning for HVAC control and power management in residential building","authors":"Sarvar Hussain Nengroo ,&nbsp;Dongsoo Har ,&nbsp;Hoon Jeong ,&nbsp;Taewook Heo ,&nbsp;Sangkeum Lee","doi":"10.1016/j.egyai.2025.100541","DOIUrl":"10.1016/j.egyai.2025.100541","url":null,"abstract":"<div><div>The use of occupancy information for heating, ventilation, and air conditioning (HVAC) control in smart buildings has become increasingly important for enhancing energy efficiency and occupant comfort. However, residential HVAC control presents significant challenges due to the complex dynamic nature of buildings and the uncertainties associated with heat loads and weather conditions. This study addresses this gap in adaptive and energy efficient HVAC control by introducing a quantum reinforcement learning (QRL) based approach. Unlike conventional reinforcement learning techniques, the QRL leverages quantum computing principles to efficiently handle high dimensional state and action spaces, enabling more precise HVAC control in multi-zone residential buildings. The proposed framework integrates real-time occupancy detection using deep learning with operational data, including power consumption patterns, air conditioner control data, and external temperature variations. To evaluate the effectiveness of the proposed approach, simulations were conducted using real world data from 26 residential households over a three month period. The results demonstrate that the QRL based HVAC control significantly reduces energy consumption and electricity costs while maintaining thermal comfort. Compared to the deep deterministic policy gradient method, the QRL approach achieved a 63% reduction in power consumption and a 64.4% decrease in electricity costs. Similarly, it outperformed the proximal policy optimization algorithm, leading to an average reduction of 62.5% in electricity costs and 62.4% in power consumption.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100541"},"PeriodicalIF":9.6,"publicationDate":"2025-06-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144480981","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":"Efficient sampling of polycyclic aromatic compounds for free energy predictions through active learning","authors":"Mohammed I. Radaideh ,&nbsp;Matt Raymond ,&nbsp;Paolo Elvati ,&nbsp;Jacob C. Saldinger ,&nbsp;Majdi I. Radaideh ,&nbsp;Angela Violi","doi":"10.1016/j.egyai.2025.100528","DOIUrl":"10.1016/j.egyai.2025.100528","url":null,"abstract":"<div><div>The physical growth of Polycyclic Aromatic Compounds (PACs) to soot particles plays a significant role in understanding the chemistry of soot formation. Insights into the process can be gained from PACs’ free energy of dimerization landscape. However, because the infeasibly large space of possible PAC dimers cannot be exhaustively simulated, researchers must train machine learning models on a subset of data to impute the rest. To this end, we propose and assess an active learning approach to discovering the optimal PACs for training a machine learning model to predict PACs’ association and dissociation free energies. The comparison between active learning and random sampling showed that active learning has faster loss convergence, requiring fewer training samples to reach the same level of accuracy. The trained model accurately modeled unseen PACs and exhibited robustness against changes in the sampling space used to train the model. More broadly, this work shows how active learning can optimize the design and improve the understanding of more expensive models in specific domains.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100528"},"PeriodicalIF":9.6,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144306846","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":"Explaining deep neural network models for electricity price forecasting with XAI","authors":"Antoine Pesenti,&nbsp;Aidan O’Sullivan","doi":"10.1016/j.egyai.2025.100532","DOIUrl":"10.1016/j.egyai.2025.100532","url":null,"abstract":"<div><div>Electricity markets are highly complex, involving lots of interactions and complex dependencies that make it hard to understand the inner workings of the market and what is driving prices. Econometric methods have been developed for this, white-box models, however, they are not as powerful as deep neural network models (DNN). In this paper, we use a DNN to forecast the price and then use XAI methods to understand the factors driving the price dynamics in the market. The objective is to increase our understanding of how different electricity markets work. To do that, we apply explainable methods such as SHAP and Gradient, combined with visual techniques like heatmaps (saliency maps) to analyse the behaviour and contributions of various features across five electricity markets. We introduce the novel concepts of SSHAP values and SSHAP lines to enhance the complex representation of high-dimensional tabular models.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100532"},"PeriodicalIF":9.6,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144313302","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":"Improving HVAC control with transfer learning: Using padding techniques for cross-building pre-training and fine-tuning","authors":"Kevlyn Kadamala,&nbsp;Des Chambers,&nbsp;Enda Barrett","doi":"10.1016/j.egyai.2025.100531","DOIUrl":"10.1016/j.egyai.2025.100531","url":null,"abstract":"<div><div>Recent advancements have shown that control strategies using Deep Reinforcement Learning (DRL) can significantly improve the management of HVAC control and energy systems in buildings, leading to significant energy savings and better comfort. Unlike conventional rule-based controllers, they demand considerable time and data to develop effective policies. Transfer learning using pre-trained models can help address this issue. In this work, we use imitation learning (IL) as a method of pre-training and reinforcement learning (RL) for fine-tuning. However, HVAC systems can vary depending on the location, building size, structure, construction materials and weather conditions. The diversity in HVAC control systems across different buildings complicates the use of IL and RL. Neural network weights trained on the source building cannot be directly transferred to the target building because of differences in input features and the number of control equipment. To overcome this problem, we propose a novel padding method to ensure that both the source and target buildings share the same state space dimensionality. Thus, the trained neural network weights are transferable, and only the output layer must be adjusted to fit the dimensionality of the target action space. Additionally, we evaluate the performance of an existing padding technique for comparison. Our experiments show that the novel padding technique outperforms zero padding by 1.37% and training from scratch by 4.59% on average.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100531"},"PeriodicalIF":9.6,"publicationDate":"2025-06-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144306847","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 comprehensive study of high-efficiency optimization method on designing segmented annular thermoelectric module","authors":"Shuhao Wang ,&nbsp;Yajing Sun ,&nbsp;Hui Chen","doi":"10.1016/j.egyai.2025.100534","DOIUrl":"10.1016/j.egyai.2025.100534","url":null,"abstract":"<div><div>To address the limitations of traditional numerical simulation methods in determining the optimal structure parameters of thermoelectric module, such as complex modeling procedures, low computational efficiency, and poor adaptability to multi-objective design, this paper introduces an efficient structural optimization approach of segmented annular thermoelectric module that combines the uniformly equivalent element integral method and multi-parameter and multi-objective optimization algorithm under both constant temperature and heat flux boundary conditions. The optimization results show that the optimal resistance ratio is independent of the boundary conditions, and the optimal thermoelectric leg ratios remain approximately 1.2 across all studied cases in this study. Notably, the optimal segment ratios are highly sensitive to the temperatures at the two ends of the optimized segmented annular thermoelectric module under all conditions and can be directly calculated using the proposed fitting formulas. In addition, an optimal total thermoelectric leg angle exists for the segmented annular thermoelectric module to achieve the maximum temperature difference within the operating temperature range of the thermoelectric materials. The output power and efficiency of the optimized segmented annular thermoelectric module can be predicted using the parameter-based fitting formulas, with relative errors below 3 % when compared to the direct optimization results. The proposed method in this paper offers significant advantages in terms of modeling simplicity, computational efficiency, and highly compatible with machine learning frameworks, thereby enabling artificial intelligence-assisted design and optimization pipelines for segmented annular thermoelectric modules.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100534"},"PeriodicalIF":9.6,"publicationDate":"2025-06-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144291442","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 temporal fusion transformer augmented GeoAI framework for estimating hourly land surface solar irradiation","authors":"Xuan Liao ,&nbsp;Man Sing Wong ,&nbsp;Rui Zhu ,&nbsp;Zhe Wang","doi":"10.1016/j.egyai.2025.100529","DOIUrl":"10.1016/j.egyai.2025.100529","url":null,"abstract":"<div><div>Accurate estimation of land surface solar irradiation is critical for effective solar energy utilization and planning of solar photovoltaic planning. Although traditional machine learning methods have been demonstrated to estimate solar irradiation effectively, they face challenges in modeling over large regions, as well as lacking of ability to model spatial diversity and temporal dynamics of solar irradiation, and providing limited interpretability. To address these limitations, this study proposed a geospatial artificial intelligence framework augmented by Temporal Fusion Transformer for hourly estimation of land surface solar irradiation. As a case study in Australia, the results demonstrate superior performance with the coefficient of the determination, the mean absolute error, and Root Mean Square Error as high as 0.90, 0.25(kWh/m²), and 0.63(kWh/m²), showing improvements of 21.62–66.67 %, 78.37–85.98 %, and 62.81–73.25 %, respectively, compared to the benchmarks of other methods, including Support Vector Regression, Random Forest, Gradient Boosting Machine, AdaBoost, Long Short-Term Memory, Temporal Convolutional Network, ConvLSTM, Transformer, and Graph Neural Network. Furthermore, interpretability results of the model indicate that among the temporal variables, observed solar irradiation and clear sky solar irradiation significantly contribute to the model’s performance. The results show this framework enhanced accuracy and interpretability for solar irradiation estimation over large areas, providing valuable insights for future studies and supporting decision-making for developing the renewable energy industry.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100529"},"PeriodicalIF":9.6,"publicationDate":"2025-06-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144298169","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":"Online prediction of combustion temperature field in a furnace from operating parameters and multi-point temperature data","authors":"Shipeng Ren ,&nbsp;Yuan An ,&nbsp;Yang Pu ,&nbsp;Yikang Liu ,&nbsp;Chun Lou ,&nbsp;Mooktzeng Lim","doi":"10.1016/j.egyai.2025.100530","DOIUrl":"10.1016/j.egyai.2025.100530","url":null,"abstract":"<div><div>The paper proposed a prediction method of combustion temperature field in a coal-fired boiler of a 350 MW unit through deep learning. The method utilizes operating parameters and multi-point temperature data as inputs for online predicting temperature field. Firstly, to establish the mapping relationship between temperature field and operating parameters as well as multi-point temperature data, a data set was constructed. In the data set, the temperature fields were obtained through the inversion of thermal radiation imaging model, while the operating parameters were collected from the distributed control system of the unit. Then, a transpose convolutional neural network (TCNN) model was developed to obtain the mapping relationship based on the data set. In the simulation study, multi-point temperature data were obtained through the forward calculation of the thermal radiation imaging model. The impact of the quantity and location of multi-point temperature data on generalization ability of the TCNN model was analyzed. In the experimental study, multi-point temperature data were measured by image probes. A comparative analysis was conducted to evaluate generalization ability of the TCNN model with and without the addition of multi-point temperature data, benchmarking against existing methods. With the addition of multi-point temperature data, the mean absolute percentage errors of predicted temperature fields are all less than 1.6 % at four stable loads, while the maximum relative error of average value of predicted temperature field decreases from 7.24 % to 2.77 % during variable load process. The proposed prediction method has promising potential for online combustion monitoring in the furnace.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100530"},"PeriodicalIF":9.6,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144195531","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 machine learning case study in nuclear fusion: Assessment of the absolute deuterium-tritium fusion power of ITER with gamma-ray spectroscopy","authors":"C. Landsmeer ,&nbsp;G. Marcer ,&nbsp;A. Dal Molin ,&nbsp;M. Rebai ,&nbsp;D. Rigamonti ,&nbsp;B. Coriton ,&nbsp;G. Gorini ,&nbsp;M. Guerini Rocco ,&nbsp;A. Kovalev ,&nbsp;A. Muraro ,&nbsp;M. Nocente ,&nbsp;E. Perelli Cippo ,&nbsp;A. Polevoi ,&nbsp;O. Putignano ,&nbsp;F. Scioscioli ,&nbsp;G. Croci ,&nbsp;M. Tardocchi","doi":"10.1016/j.egyai.2025.100526","DOIUrl":"10.1016/j.egyai.2025.100526","url":null,"abstract":"<div><div>Nuclear fusion holds great potential as a carbon-neutral means of electricity production. However, technical aspects of its implementation remain challenging. The real-time measurement of the fusion power released during Deuterium-Tritium (DT) fusion is one such aspect. The use of tools from artificial intelligence may help to solve this issue.</div><div>Recently, during experiments performed at the Joint European Torus, a novel method was developed to measure the fusion power in magnetic confinement fusion devices. Said method exploits the fact that gamma-rays released by the DT fusion reaction can be registered with a gamma-ray spectrometer. Expanding on this work, a machine learning algorithm was developed to estimate DT fusion power at ITER by use of the Radial Gamma-Ray Spectrometer (RGRS) measurements, as well as the magnetic equilibrium as an additional source of information.</div><div>The algorithm was trained and tested on a set of 75 simulations of ITER DT plasma scenarios. By testing the algorithm by repeated 5-fold cross-validation, the average deviation of the estimated fusion power from the reference was found to be 0.32%, while the relative error had a standard deviation of 0.97%. When statistical fluctuations were included in the analysis, the lowest measurable fusion power resulted to be around 30<!--> <!-->MW, making the RGRS suitable for the fusion power measurement requirements at ITER.</div><div>This project demonstrated that a machine learning approach leads to promising results when coupled with prior knowledge and the integration of various kinds of sensor and simulation data. This and related algorithms may eventually contribute to the development of fusion power as a reliable, carbon-neutral source of energy.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"21 ","pages":"Article 100526"},"PeriodicalIF":9.6,"publicationDate":"2025-05-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144195380","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}