Energy and AI最新文献_第5页

Deciphering city-level residential AMI data: An unsupervised data mining framework and case study 解密城市级住宅 AMI 数据：无监督数据挖掘框架和案例研究

IF 9.6

Energy and AI Pub Date : 2025-02-14 DOI: 10.1016/j.egyai.2025.100484

Han Li, Miguel Heleno, Kaiyu Sun, Wanni Zhang, Luis Rodriguez Garcia, Tianzhen Hong

{"title":"Deciphering city-level residential AMI data: An unsupervised data mining framework and case study","authors":"Han Li, Miguel Heleno, Kaiyu Sun, Wanni Zhang, Luis Rodriguez Garcia, Tianzhen Hong","doi":"10.1016/j.egyai.2025.100484","DOIUrl":"10.1016/j.egyai.2025.100484","url":null,"abstract":"<div><div>Buildings account for more than one third of global energy consumption and carbon emissions, making the optimization of their energy use crucial for sustainability. Advanced Metering Infrastructures data offers a rich source of information for understanding and improving building energy performance, yet existing frameworks for leveraging this data are limited. This paper presents a comprehensive data-mining framework for analyzing Advanced Metering Infrastructures data at multiple temporal and spatial scales, beneficial for building owners, operators, and utility companies. Utilizing hourly electricity consumption data for the east region of Portland, Oregon, the study systematically extracts key statistics such as start hour, duration, and peak hour of load periods across daily, weekly, and annual evaluation windows. The framework employs a list of techniques including load-level detection, home vacancy detection, and weather-sensitivity analysis and statistical methods to provide detailed insights into building energy dynamics. As an unsupervised study, it reveals patterns and trends without predefined labels or categories. Key findings highlight the substantial impact of the COVID-19 pandemic on residential energy use, uncover patterns like intraday load variations, weekly consumption trends, and annual weather sensitivity. The insights gained can potentially inform better energy management strategies, support grid operations and planning, guide policy-making for energy efficiency improvements, as well as improve input and assumptions in the building energy modeling. This study opens pathways for future research, including integrating more data sources and collaborating with utility companies to validate hypotheses and further explore building energy use insights.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"20 ","pages":"Article 100484"},"PeriodicalIF":9.6,"publicationDate":"2025-02-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143429795","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Explainable multi-step heating load forecasting: Using SHAP values and temporal attention mechanisms for enhanced interpretability 可解释的多步热负荷预测：利用SHAP值和时间注意机制增强可解释性

IF 9.6

Energy and AI Pub Date : 2025-02-13 DOI: 10.1016/j.egyai.2025.100480

Alexander Neubauer, Stefan Brandt, Martin Kriegel

{"title":"Explainable multi-step heating load forecasting: Using SHAP values and temporal attention mechanisms for enhanced interpretability","authors":"Alexander Neubauer, Stefan Brandt, Martin Kriegel","doi":"10.1016/j.egyai.2025.100480","DOIUrl":"10.1016/j.egyai.2025.100480","url":null,"abstract":"<div><div>The role of heating load forecasts in the energy transition is significant, given the considerable increase in the number of heat pumps and the growing prevalence of fluctuating electricity generation. While machine learning methods offer promising forecasting capabilities, their black-box nature makes them difficult to interpret and explain. The deployment of explainable artificial intelligence methodologies enables the actions of these machine learning models to be made transparent.</div><div>In this study, a multi-step forecast was employed using an Encoder–Decoder model to forecast the hourly heating load for an multifamily residential building and a district heating system over a forecast horizon of 24-h. By using 24 instead of 48 lagged hours, the simulation time was reduced from 92.75s to 45.80s and the forecast accuracy was increased. The feature selection was conducted for four distinct methods. The Tree and Deep SHAP method yielded superior results in feature selection. The application of feature selection according to the Deep SHAP values resulted in a reduction of 3.98% in the training time and a 8.11% reduction in the NRMSE. The utilisation of local Deep SHAP values enables the visualisation of the influence of past input hours and individual features. By mapping temporal attention, it was possible to demonstrate the importance of the most recent time steps in a intrinsic way.</div><div>The combination of explainable methods enables plant operators to gain further insights and trustworthiness from the purely data-driven forecast model, and to identify the importance of individual features and time steps.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"20 ","pages":"Article 100480"},"PeriodicalIF":9.6,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143445462","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Domain-informed CNN architectures for downscaling regional wind forecasts 域通知CNN架构缩小区域风力预报

IF 9.6

Energy and AI Pub Date : 2025-02-13 DOI: 10.1016/j.egyai.2025.100485

Alexander M. Campbell , Simon C. Warder , B. Bhaskaran , Matthew D. Piggott

{"title":"Domain-informed CNN architectures for downscaling regional wind forecasts","authors":"Alexander M. Campbell , Simon C. Warder , B. Bhaskaran , Matthew D. Piggott","doi":"10.1016/j.egyai.2025.100485","DOIUrl":"10.1016/j.egyai.2025.100485","url":null,"abstract":"<div><div>High-resolution wind speed forecasts are of great importance to the wind energy industry, from short-term energy forecasting and trading to longer-term resource assessment and planning. Generating high-resolution regional wind forecasts currently requires compute-intensive numerical models to downscale from a global forecast. Black-box AI models, once trained, can produce results in a fraction of the time and cost; however, they tend to produce smoothed outputs, are not interpretable and generalise poorly. The domain-informed AI architecture presented in this work seeks to address these problems by incorporating prior static fields directly into the model architecture. Specifically, the proposed approach combines two sequential U-Nets – the first upsamples the input wind fields and expands the number of feature maps, a fusion layer then injects prior static data such as topography, and a second U-Net generates the final output wind field. This approach improves all performance metrics versus a baseline U-Net model and generalises better to out-of-sample scenarios. In addition, this study compares the performance of several loss functions, including standard pixel-wise measures such as mean-squared error, structural similarity and frequency-focused functions, and a function based on Wiener filter theory. All loss functions, with the exception of the Wiener loss, perform comparably and tend to attenuate higher-frequency detail. Although the Wiener loss encourages higher frequencies, it over-estimates amplitudes. A composite Wiener-L1 loss function balances generating high-frequency detail and correctly predicting amplitudes.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"20 ","pages":"Article 100485"},"PeriodicalIF":9.6,"publicationDate":"2025-02-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143474432","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Physics-informed machine learning for enhanced prediction of condensation heat transfer 利用物理信息机器学习加强冷凝传热预测

IF 9.6

Energy and AI Pub Date : 2025-02-08 DOI: 10.1016/j.egyai.2025.100482

Haeun Lee , Cheonkyu Lee , Hyoungsoon Lee

{"title":"Physics-informed machine learning for enhanced prediction of condensation heat transfer","authors":"Haeun Lee , Cheonkyu Lee , Hyoungsoon Lee","doi":"10.1016/j.egyai.2025.100482","DOIUrl":"10.1016/j.egyai.2025.100482","url":null,"abstract":"<div><div>Developing a universal model for predicting condensation heat transfer coefficients remains challenging, particularly for steam–non-condensable gas mixtures, owing to the intricate nonlinear interactions between multiphase flow, heat, and mass transfer phenomena. Data-driven machine learning (ML) shows promise in efficiently and accurately predicting condensation heat transfer coefficients. Research has employed various ML methods—multilayer perceptron neural networks, convolutional-neural-network–based DenseNet, backpropagation neural networks, etc.—to investigate steam condensation with non-condensable gases. However, these exhibit limited extrapolation ability and heavily rely on data quantity owing to their black-box nature. This study proposes a physics-informed ML model that combines physical constraints derived from the modified Nusselt model with conventional data-driven ML techniques. The model's predictive performance is evaluated using a comprehensive database (879 datapoints from 13 studies). A physics-constrained and eight data-driven ML methods are assessed. The results reveal that the physics-constrained approach combined with XGBoost significantly outperforms conventional ML methods on extrapolation datasets (199 datapoints from 3 studies), achieving a mean absolute percentage error of 11.22 %, which is approximately half that of the best-performing fully data-driven model at 21.63 %. The model demonstrates consistent and reliable performance across diverse datasets, making it an effective tool for predicting heat transfer coefficients in steam–non-condensable gas mixtures. By deepening the understanding of the underlying physical processes, the proposed model supports the development of precise and efficient engineering solutions for condensation heat transfer.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"20 ","pages":"Article 100482"},"PeriodicalIF":9.6,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143437004","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Artificial Neural Network (ANN) driven Techno-Economic Predictions for Micro Gas Turbines (MGT) based Energy Applications 人工神经网络（ANN）驱动的微型燃气轮机（MGT）能源应用技术经济预测

IF 9.6

Energy and AI Pub Date : 2025-02-06 DOI: 10.1016/j.egyai.2025.100483

A.H.Samitha Weerakoon, Mohsen Assadi

{"title":"Artificial Neural Network (ANN) driven Techno-Economic Predictions for Micro Gas Turbines (MGT) based Energy Applications","authors":"A.H.Samitha Weerakoon, Mohsen Assadi","doi":"10.1016/j.egyai.2025.100483","DOIUrl":"10.1016/j.egyai.2025.100483","url":null,"abstract":"<div><div>This paper introduces a novel Artificial Neural Network (ANN)-driven methodology for the techno-economic assessment (TEA) of Micro Gas Turbines (MGT) in energy applications, addressing the limitations of traditional TEA approaches which often lack adaptability to dynamic market conditions and technological advancements. The developed ANN model, employing a multi-layer perceptron architecture, leverages advanced machine learning techniques to accurately predict key economic indicators such as Net Present Value (NPV), Internal Rate of Return (IRR), Payback Period (PBP), and Return on Investment (ROI). Analysis of over 450 MGT-related energy project profiles validates the model's efficacy, demonstrating high predictive accuracy with a Mean Squared Error (MSE) of 0.0005 and an R-squared value of 0.993. The model is further validated across key application areas for MGT's, including PV and Solar, Distributed Energy Generation (DEG) and Hydrogen-Natural Gas blended systems for microgrid applications, showcasing its potential to enhance decision-making for energy investments. This approach not only streamlines the economic assessment process, reducing time and effort significantly, but also enhances decision-making for stakeholders by providing rapid, real-time economic analyses. The integration of ANN into MGT TEA sets a new standard for conducting techno-economic evaluations, potentially transforming energy system optimization practices.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"20 ","pages":"Article 100483"},"PeriodicalIF":9.6,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143420490","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Data-driven modeling of open circuit voltage hysteresis for LiFePO4 batteries with conditional generative adversarial network 基于条件生成对抗网络的磷酸铁锂电池开路电压滞后数据驱动建模

IF 9.6

Energy and AI Pub Date : 2025-01-31 DOI: 10.1016/j.egyai.2025.100478

Lisen Yan , Jun Peng , Zeyu Zhu , Heng Li , Zhiwu Huang , Dirk Uwe Sauer , Weihan Li

{"title":"Data-driven modeling of open circuit voltage hysteresis for LiFePO4 batteries with conditional generative adversarial network","authors":"Lisen Yan , Jun Peng , Zeyu Zhu , Heng Li , Zhiwu Huang , Dirk Uwe Sauer , Weihan Li","doi":"10.1016/j.egyai.2025.100478","DOIUrl":"10.1016/j.egyai.2025.100478","url":null,"abstract":"<div><div>The hysteresis effect represents the difference in open circuit voltage (OCV) between the charge and discharge processes of batteries. An accurate estimation of open circuit voltage considering hysteresis is critical for precise modeling of <span><math><msub><mrow><mtext>LiFePO</mtext></mrow><mrow><mtext>4</mtext></mrow></msub></math></span> batteries. However, the intricate influence of state-of-charge (SOC), temperature, and battery aging have posed significant challenges for hysteresis modeling, which have not been comprehensively considered in existing studies. This paper proposes a data-driven approach with adversarial learning to model hysteresis under diverse conditions, addressing the intricate dependencies on SOC, temperature, and battery aging. First, a comprehensive experimental scheme is designed to collect hysteresis dataset under diverse SOC paths, temperatures and aging states. Second, the proposed data-driven model integrates a conditional generative adversarial network with long short-term memory networks to enhance the model’s accuracy and adaptability. The generator and discriminator are designed based on LSTM networks to capture the dependency of hysteresis on historical SOC and conditional information. Third, the conditional matrix, incorporating temperature, health state, and historical paths, is constructed to provide the scenario-specific information for the adversarial network, thereby enhancing the model’s adaptability. Experimental results demonstrate that the proposed model achieves a voltage error of less than 3.8 mV across various conditions, with accuracy improvements of 31.3–48.7% compared to three state-of-the-art models.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"20 ","pages":"Article 100478"},"PeriodicalIF":9.6,"publicationDate":"2025-01-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143094836","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimising quantile-based trading strategies in electricity arbitrage 电力套利中基于分位数的交易策略优化

IF 9.6

Energy and AI Pub Date : 2025-01-27 DOI: 10.1016/j.egyai.2025.100476

Ciaran O’Connor , Joseph Collins , Steven Prestwich , Andrea Visentin

{"title":"Optimising quantile-based trading strategies in electricity arbitrage","authors":"Ciaran O’Connor , Joseph Collins , Steven Prestwich , Andrea Visentin","doi":"10.1016/j.egyai.2025.100476","DOIUrl":"10.1016/j.egyai.2025.100476","url":null,"abstract":"<div><div>Efficiently integrating renewable resources into electricity markets is vital for addressing the challenges of matching real-time supply and demand while mitigating revenue losses caused by curtailments. To address this challenge effectively, the incorporation of storage devices can enhance the reliability and efficiency of the grid, improving market liquidity and reducing price volatility. In short-term electricity markets, participants face numerous options, each presenting unique challenges and opportunities, with trading strategies fundamental towards maximising profits. This study explores the optimisation of day-ahead and balancing market trading in the Irish electricity market from 2019 to 2022, leveraging quantile-based forecasts. Employing three trading approaches with practical constraints, our research evaluates trading strategies, increases trading frequency, and employs flexible timestamp orders. Our findings underscore the profit potential of simultaneous participation in both day-ahead and balancing markets, especially with larger battery storage systems; despite increased costs and narrower profit margins associated with higher-volume trading, with the implementation of dynamic dual-market strategies playing a significant role in maximising profits and addressing market challenges. Finally, we evaluate the economic viability of four commercial battery storage systems through scenario analysis, showing that larger batteries achieve shorter returns on investment.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"20 ","pages":"Article 100476"},"PeriodicalIF":9.6,"publicationDate":"2025-01-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143094837","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Physics-Informed Neural Network for modeling and predicting temperature fluctuations in proton exchange membrane electrolysis 用于模拟和预测质子交换膜电解温度波动的物理信息神经网络

IF 9.6

Energy and AI Pub Date : 2025-01-22 DOI: 10.1016/j.egyai.2025.100474

Islam Zerrougui , Zhongliang Li , Daniel Hissel

{"title":"Physics-Informed Neural Network for modeling and predicting temperature fluctuations in proton exchange membrane electrolysis","authors":"Islam Zerrougui , Zhongliang Li , Daniel Hissel","doi":"10.1016/j.egyai.2025.100474","DOIUrl":"10.1016/j.egyai.2025.100474","url":null,"abstract":"<div><div>Proton Exchange Membrane (PEM) electrolysis stands as a cornerstone technology in the clean energy sector, driving the production of hydrogen and oxygen from water. A critical aspect of ensuring the efficiency and safety of this process lies in the precise monitoring and control of temperature at the electrolysis outlet. However, accurately characterizing temperature changes within the PEM electrolysis system can be challenging due to the fluctuation of renewable energies. This study introduces an approach integrating data with fundamental physics principles known as Physics-Informed Neural Networks (PINNs). This method solves differential equations and estimates the unknown parameters governing the temperature dynamics within the PEM electrolysis system. We consider two distinct scenarios: a zero-dimensional model and a one-dimensional model. The results demonstrate the PINN’s proficiency in accurately identifying the parameters and solving for temperature fluctuations within the system with different input conditions. Furthermore, we compare the PINN with the Long Short-Term Memory (LSTM) method to predict the outlet temperature of the electrolysis. The PINN outperformed the LSTM method, highlighting its reliability and precision, achieving a Mean Squared Error (MSE) of 0.1596 compared to 1.2132 for LSTM models. The proposed method shows a high performance in dealing with sensor noises and avoids overfitting problems. This synergy of physics knowledge and data-driven learning opens new pathways towards real-time digital twins, enhanced predictive control, and improved reliability for PEM electrolysis and other complex, data-scarce energy systems.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"20 ","pages":"Article 100474"},"PeriodicalIF":9.6,"publicationDate":"2025-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143094839","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Methodology for predicting material performance by context-based modeling: A case study on solid amine CO2 adsorbents 基于情景的建模预测材料性能的方法：固体胺CO2吸附剂的案例研究

IF 9.6

Energy and AI Pub Date : 2025-01-21 DOI: 10.1016/j.egyai.2025.100477

Shuangjun Li , Zhixin Huang , Yuanming Li , Shuai Deng , Xiangkun Elvis Cao

{"title":"Methodology for predicting material performance by context-based modeling: A case study on solid amine CO2 adsorbents","authors":"Shuangjun Li , Zhixin Huang , Yuanming Li , Shuai Deng , Xiangkun Elvis Cao","doi":"10.1016/j.egyai.2025.100477","DOIUrl":"10.1016/j.egyai.2025.100477","url":null,"abstract":"<div><div>Traditional materials informatics leverages big data and machine learning (ML) to forecast material performance based on structural features but often overlooks valuable textual information. In this work, we proposed a novel methodology for predicting material performance through context-based modeling using large language models (LLMs). This method integrates both numerical and textual information, enhancing predictive accuracy and scalability. In the case study, the approach is applied to predict the performance of solid amine CO<sub>2</sub> adsorbents under direct air capture (DAC) conditions. ChatGPT 4o model was used to employ in-context learning to predict CO<sub>2</sub> adsorption uptake based on input features, including material properties and experimental conditions. The results show that context-based modeling can reduce prediction error in comparison to traditional ML models in the prediction task. We adopted Sapley Additive exPlanations (SHAP) to further elucidate the importance of various input features. This work highlights the potential of LLMs in materials science, offering a cost-effective, efficient solution for complex predictive tasks.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"20 ","pages":"Article 100477"},"PeriodicalIF":9.6,"publicationDate":"2025-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143094838","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Comparison of Kolmogorov–Arnold Networks and Multi-Layer Perceptron for modelling and optimisation analysis of energy systems 用于能源系统建模和优化分析的Kolmogorov-Arnold网络和多层感知器的比较

IF 9.6

Energy and AI Pub Date : 2025-01-16 DOI: 10.1016/j.egyai.2025.100473

Talha Ansar , Waqar Muhammad Ashraf

{"title":"Comparison of Kolmogorov–Arnold Networks and Multi-Layer Perceptron for modelling and optimisation analysis of energy systems","authors":"Talha Ansar , Waqar Muhammad Ashraf","doi":"10.1016/j.egyai.2025.100473","DOIUrl":"10.1016/j.egyai.2025.100473","url":null,"abstract":"<div><div>Considering the improved interpretable performance of Kolmogorov–Arnold Networks (KAN) algorithm compared to multi-layer perceptron (MLP) algorithm, a fundamental research question arises on how modifying the loss function of KAN affects its modelling performance for energy systems, particularly industrial-scale thermal power plants. In this regard, first, we modify the loss function of both KAN and MLP algorithms and embed Pearson Correlation Coefficient (PCC). Second, the algorithmic configurations built on PCC, i.e., KAN_PCC and MLP_PCC as well as original architecture of KAN and MLP are deployed for modelling and optimisation analyses for two case studies of energy systems: (i) energy efficiency cooling and energy efficiency heating of buildings, and (ii) power generation operation of 660 MW capacity thermal power plant. The analysis reveals superior modelling performance of KAN and KAN_PCC algorithms than those of MLP and MLP_PCC for the two case studies. KAN models are embedded in the optimisation framework of nonlinear programming and feasible optimal solutions are estimated, maximising thermal efficiency up to 42.17 ± 0.88 % and minimising turbine heat rate to 7487 ± 129 kJ/kWh corresponding to power generation of 500 ± 14 MW for the thermal power plant. It is anticipated that the scientific, research and industrial community may benefit from the fundamental insights presented in this paper for the ML algorithm selection and carrying out model-based optimisation analysis for the performance enhancement of energy systems.</div></div>","PeriodicalId":34138,"journal":{"name":"Energy and AI","volume":"20 ","pages":"Article 100473"},"PeriodicalIF":9.6,"publicationDate":"2025-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143094840","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0