Energy Reports最新文献

Predicting the future climate: Integrating renewable energy and machine learning to address temperature and GHG emissions 预测未来气候：整合可再生能源和机器学习来解决温度和温室气体排放问题

IF 5.1 3区工程技术

Energy Reports Pub Date : 2025-09-20 DOI: 10.1016/j.egyr.2025.09.004

Hiba Darwish , Issa W. AlHmoud , Anish Chand Turlapaty , Balakrishna Gokaraju

{"title":"Predicting the future climate: Integrating renewable energy and machine learning to address temperature and GHG emissions","authors":"Hiba Darwish , Issa W. AlHmoud , Anish Chand Turlapaty , Balakrishna Gokaraju","doi":"10.1016/j.egyr.2025.09.004","DOIUrl":"10.1016/j.egyr.2025.09.004","url":null,"abstract":"<div><div>Over the years, population growth has led to a sharp rise in electricity demand, which in turn has increased the use of fossil fuels, the main source of Greenhouse gas emissions. Renewable energy offers a cleaner alternative, helping reduce these emissions and lessen the impact of climate change. While recent studies primarily focus on the direct correlation between Greenhouse gas emissions and temperature elevation, this study takes a new approach by breaking the analysis into two stages. The first stage investigates the relationship between Greenhouse gas concentrations and the suppression of solar radiation trapped in the atmosphere. Subsequently, the second stage links trapped solar radiation with average temperature patterns. Two mathematical models with 95 % prediction accuracy are developed to estimate the reduction in trapped solar radiation from existing RE projects and predict resulting temperature changes. To improve the prediction accuracy, multiple Machine learning models, namely Decision Tree, Support Vector Machine, and kernel Naive Bayes, were utilized to predict the average temperature based on the two features, Greenhouse gas and the trapped solar radiation. Neural Network Clustering was also employed to capture nonlinear interactions between <span><math><mrow><mi>C</mi><msub><mrow><mi>O</mi></mrow><mrow><mn>2</mn></mrow></msub></mrow></math></span> and solar radiation. This approach overcomes the limitations of traditional methods while aligning with climate physics principles. The paper also addresses the impact of the climate change phenomenon on the evaporation rate in Jordan as a case study, offering insights into its regional environmental effects.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"14 ","pages":"Pages 2399-2419"},"PeriodicalIF":5.1,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145095271","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Short-term wind power prediction based on a new hybrid model 基于新混合模型的短期风电功率预测

IF 5.1 3区工程技术

Energy Reports Pub Date : 2025-09-18 DOI: 10.1016/j.egyr.2025.08.051

Boxuan Lai , Yanfei You , Houlung Cheng

{"title":"Short-term wind power prediction based on a new hybrid model","authors":"Boxuan Lai , Yanfei You , Houlung Cheng","doi":"10.1016/j.egyr.2025.08.051","DOIUrl":"10.1016/j.egyr.2025.08.051","url":null,"abstract":"<div><div>The global energy crisis and the integration of intermittent wind power into electrical grids have necessitated accurate short-term forecasting, which has become crucial for mitigating grid imbalances and compensating for limitations. To address this issue, this study investigates the performance of a novel hybrid prediction model, enhanced through data analysis and neural network refinements used to reduce wind power forecasting errors. This technique employs feature engineering, including statistical and temporal pattern analysis, to extract new input features from raw data and process missing values or outliers. Architectural refinements include self-attention mechanisms within dilated convolution and the decoder-free Transformer design, optimized to capture complex temporal dependencies efficiently. A novel hybrid framework integrating customized models leverages feature engineering to enhance forecasting accuracy. The resulting model, validated on datasets from 14 geographically diverse wind farms, significantly reduces prediction errors. Specifically, feature engineering alone boosted accuracy by at least 5.44% (RMSE), while the final ensemble model, integrating the strengths of individual models, achieved a 7.89% RMSE ranking score improvement in generalization performance compared to the next best single model. These results demonstrate the effectiveness of the proposed technique for reliable short-term wind power forecasting across varied terrains, supporting its use for improved operational planning and grid management.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"14 ","pages":"Pages 2384-2398"},"PeriodicalIF":5.1,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145095270","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Threshold effects of renewable energy and environmental R&D on green growth: A dynamic panel analysis of OECD countries 可再生能源与环境研发对绿色增长的阈值效应：基于OECD国家的动态面板分析

IF 5.1 3区工程技术

Energy Reports Pub Date : 2025-09-18 DOI: 10.1016/j.egyr.2025.08.041

Khaled Mili , Ismail Bengana , Samir Aouini , Moussa Hezla

{"title":"Threshold effects of renewable energy and environmental R&D on green growth: A dynamic panel analysis of OECD countries","authors":"Khaled Mili , Ismail Bengana , Samir Aouini , Moussa Hezla","doi":"10.1016/j.egyr.2025.08.041","DOIUrl":"10.1016/j.egyr.2025.08.041","url":null,"abstract":"<div><div>This study investigates the roles of renewable energy consumption and environmental R&D investment in promoting sustainable green growth across OECD countries. Using a comprehensive panel dataset spanning 35 OECD nations, we examine how renewable energy adoption and environmental research spending contribute to sustainability outcomes. The research employs the Augmented Mean Group (AMG) methodology to address cross-sectional dependence and heterogeneity in panel data analysis. Results demonstrate that renewable energy consumption has a significant positive impact on green growth (coefficient: 0.0099, p < 0.01), while revealing complex dynamics in the relationship between environmental R&D spending and sustainability outcomes (coefficient: −0.0011, p > 0.05). The study finds that the effectiveness of environmental initiatives depends more on implementation strategies than on research expenditure alone. These findings provide important insights into the differential impacts of renewable energy adoption and R&D investment on sustainable growth in developed economies. The research offers valuable implications for optimizing resource allocation between renewable energy deployment and environmental research initiatives in pursuing sustainability goals.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"14 ","pages":"Pages 2368-2383"},"PeriodicalIF":5.1,"publicationDate":"2025-09-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145095269","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Power quality improvement in Grid‐Connected PV system using fractional order controller and Fractional-order Lipschitz Recurrent Neural Network (FLRNN) 基于分数阶控制器和分数阶Lipschitz递归神经网络（FLRNN）的并网光伏系统电能质量改进

IF 5.1 3区工程技术

Energy Reports Pub Date : 2025-09-16 DOI: 10.1016/j.egyr.2025.08.043

Rayed AlGhamdi , Ghanshyam G. Tejani , Hasim Khan , Naveen Kumar Sharma , Sunil Kumar Sharma , D. Baba Basha

{"title":"Power quality improvement in Grid‐Connected PV system using fractional order controller and Fractional-order Lipschitz Recurrent Neural Network (FLRNN)","authors":"Rayed AlGhamdi , Ghanshyam G. Tejani , Hasim Khan , Naveen Kumar Sharma , Sunil Kumar Sharma , D. Baba Basha","doi":"10.1016/j.egyr.2025.08.043","DOIUrl":"10.1016/j.egyr.2025.08.043","url":null,"abstract":"<div><div>In hybrid renewable energy systems (HRES), particularly with solar, fuel cell, and battery components, common PQ disturbances that occur are voltage sags, swells, and fluctuations. An intelligent FLRNN-FOPID-DSTATCOM control framework is proposed that integrates a Fractional-Order PID (FOPID) controller, whose parameters are optimized using a novel Draft-Mongoose Tailored Earthworm Optimizer (DTEO), and a Fractional-Order Lipschitz Recurrent Neural Network (FLRNN), for PQ improvement under varying load and source conditions. The simulated experimental setup used the MATLAB/Simulink environment, after which this approach underwent a rigorous comparative study with traditional PID, Meta-heuristic PI/PID, and Sliding Mode Controllers (SMC). The quantitative nature of the results proves a substantial reduction of THD down to 0.0043 from undefined baseline THD values, signifying very good harmonic suppression. The system is able to stabilize the PV voltage and current from varying ranges of −200 V to 350 V and −800 A to 300 A, respectively, to steady outputs of 350 V and 300 A. While the battery and DC-link voltages are restored from momentary dips as low as 200 V to steady state voltages of 270–300 V, variability due to voltage sags, swells, and fluctuations are suppressed such that LV is stabilized within ±2.5–3 V and IC ramped effectively to ±500–598 A. Therefore, compared to the existing methods, the proposed controller greatly reduces THD more swiftly with better dynamic response and adaptability to PV variability, saving times required for the neural network training. From the above results, the proposed technique can be considered as a promising high-performance method to enhance power quality in real-time in a grid-connected HRES environment.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"14 ","pages":"Pages 2336-2367"},"PeriodicalIF":5.1,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145095268","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimal coordination of water-energy nexus resources in metropolitan water resource network for approaching carbon neutrality 面向碳中和的城市水资源网络水能关联资源优化协调

IF 5.1 3区工程技术

Energy Reports Pub Date : 2025-09-11 DOI: 10.1016/j.egyr.2025.08.045

Jaeho Lee, Jaewan Suh

{"title":"Optimal coordination of water-energy nexus resources in metropolitan water resource network for approaching carbon neutrality","authors":"Jaeho Lee, Jaewan Suh","doi":"10.1016/j.egyr.2025.08.045","DOIUrl":"10.1016/j.egyr.2025.08.045","url":null,"abstract":"<div><div>The water resource industry has been recognized for its substantial electricity consumption. Moreover, recent movements towards RE100 membership aimed at managing water resources have begun to expedite the technical demands for a novel coordination method. This study proposes the involvement of various types of electricity markets from the perspective of operators within a metropolitan water resource network consisting of flow and treatment capacity of each facility in each MWPF. The proposed method considers not only individual water treatment facilities, but also a comprehensive management approach for the network based on detailed interdisciplinary methods. The treatment capacity and deliverability of small hydropower generators (SHGs) at each node was linearized in piece-wise formation based on experimental and actual specifications provided by the corporation. The analysis is not based on social scientific context but purely con-ducted from the technical perspectives considering the treatment processes that have been neglected in the previous research. Participation in the economic demand response was simulated to avoid compromising the intraday procurement requirements of water resources. The variability mitigation by the renewable energy resources in the network was simulated to be less than 7 % during the entire scheduling period. The economic feasibility of REC markets with multipliers was improved by 58.47 % through appropriate hourly coordination, considering market settlement. The profit from participation in electricity market was simulated to be increased by 165.98 % and 174.32 % in S3 and S4.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"14 ","pages":"Pages 2315-2335"},"PeriodicalIF":5.1,"publicationDate":"2025-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145043899","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Thermo-catalytic pyrolysis and kinetic study of non-edible castor seeds into renewable liquid fuel and value-added chemicals 非食用蓖麻籽热催化热解制备可再生液体燃料及增值化学品的动力学研究

IF 5.1 3区工程技术

Energy Reports Pub Date : 2025-09-10 DOI: 10.1016/j.egyr.2025.08.050

Tanushka Florence Panicker , Richa Gupta , Ranjeet Kumar Mishra , Kaustubha Mohanty

{"title":"Thermo-catalytic pyrolysis and kinetic study of non-edible castor seeds into renewable liquid fuel and value-added chemicals","authors":"Tanushka Florence Panicker , Richa Gupta , Ranjeet Kumar Mishra , Kaustubha Mohanty","doi":"10.1016/j.egyr.2025.08.050","DOIUrl":"10.1016/j.egyr.2025.08.050","url":null,"abstract":"<div><div>Biomass is a promising renewable resource due to its availability, low processing cost, high conversion efficiency, and reduced life-cycle carbon emissions, making it a potential alternative to fossil fuels. However, the pyrolysis-derived pyrolytic oil typically suffers from poor quality, high viscosity, thermal instability, and corrosiveness, limiting its direct use as fuel. These drawbacks can be mitigated through catalytic upgrading. Thus, the present study investigates the kinetic behaviour and thermo-catalytic pyrolysis of castor seeds using a semi-batch reactor at 550 °C, 50 °C min<sup>−1</sup> heating rates and 100 mL min<sup>−1</sup> nitrogen flow rate. Thermogravimetric analysis was conducted at heating rates of 10, 30, and 50 °C min<sup>−1</sup>, applying Ozawa-Flynn-Wall, Friedman (FM), Kissinger-Akahira-Sunose (KAS), Vyazovkin (VZ), Starink (STM), alongside the Criado model. The average apparent activation energies were determined as 174.69, 172.25, 173.53, 66.17, and 156.93 kJ mol<sup>−1</sup> for OFW, KAS, STM, VZ, and FM, respectively. The Criado model confirmed that biomass decomposition follows a complex, multi-step reaction mechanism. Further, pyrolysis using ZSM-5 and ZnO catalysts enhanced pyrolytic oil yields from 49.82 wt% (thermal) to 55.52 wt% and 54.33 wt%, respectively. Catalyst inclusion also improved the carbon content (by 14.12 % for ZSM-5 and 12.19 % for ZnO), higher heating value (by 1.72 and 1.00 MJ kg<sup>−1</sup>), and pH (by 0.89 and 1.50). The resulting biochar exhibited 57.14 % carbon content, 36.23 % oxygen, 19.88 MJ kg<sup>−1</sup> Higher heating value, and a Brunauer-Emmett-Teller surface area of 84.98 m² g<sup>−1</sup>. Overall, this study highlights the potential of castor seeds as a viable feedstock for sustainable fuel production through catalytic biomass valorisation.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"14 ","pages":"Pages 2280-2295"},"PeriodicalIF":5.1,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145026516","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Research on the scenario prediction and classification of emission reduction strategy of carbon emission in china's oil and gas industry 中国油气行业碳排放情景预测与减排策略分类研究

IF 5.1 3区工程技术

Energy Reports Pub Date : 2025-09-10 DOI: 10.1016/j.egyr.2025.09.001

Haiyan Jin, Lianwei Liu, Xiangyu Sun

{"title":"Research on the scenario prediction and classification of emission reduction strategy of carbon emission in china's oil and gas industry","authors":"Haiyan Jin, Lianwei Liu, Xiangyu Sun","doi":"10.1016/j.egyr.2025.09.001","DOIUrl":"10.1016/j.egyr.2025.09.001","url":null,"abstract":"<div><div>With the introduction of the \"dual carbon\" goal, the pathway for carbon reduction in China's energy system is gradually unfolding. This paper focuses on the oil and gas industry, one of the core sectors contributing to carbon emissions, as the research subject. Initially, a PSO-BP neural network model is employed to estimate the carbon emissions of the oil and gas industry (CEOG), including total carbon emissions (TCEOG) and carbon intensity of the oil and gas industry (CIOG). Subsequently, the integration of potential index and K-Means clustering method is applied to classify the prediction of CEOG, and corresponding emission reduction strategies are proposed based on the classification results. The findings are as follows: (1) A peak in provincial TCEOG by 2030 can only be achieved under a low-carbon scenario. (2) The provinces whose emission reduction potential index ranks the top 5 are the economically developed eastern coastal regions, while the bottom five are the less developed energy-producing regions. Based on a total amount-efficiency model, CEOG is categorized into four groups: HE-HE (high emission-high efficiency), HE-LE (high emission-low efficiency), LE-HE (low emission-high efficiency), and LE-LE (low emission-low efficiency). (3) The comprehensive CE reduction potential and total amount-efficiency results classify the future CEOG scenarios for 30 provinces in China into ten categories for the first time. Based on the actual development status of each province, tailored CE reduction strategies for the oil and gas industry should be formulated for each category, providing a practical, detailed, and quantifiable theoretical basis for the development and adjustment of national, provincial, or regional carbon reduction policies.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"14 ","pages":"Pages 2264-2279"},"PeriodicalIF":5.1,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145026515","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A coordinated AC-DC recovery strategy for suppressing continuous commutation failures in HVDC transmission 一种抑制高压直流输电连续换流故障的交直流协调恢复策略

IF 5.1 3区工程技术

Energy Reports Pub Date : 2025-09-09 DOI: 10.1016/j.egyr.2025.08.044

Jingbo Zhao, Wenbo Li, Dajiang Wang

{"title":"A coordinated AC-DC recovery strategy for suppressing continuous commutation failures in HVDC transmission","authors":"Jingbo Zhao, Wenbo Li, Dajiang Wang","doi":"10.1016/j.egyr.2025.08.044","DOIUrl":"10.1016/j.egyr.2025.08.044","url":null,"abstract":"<div><h3>Summary</h3><div>Addressing the low sensitivity of DC current regulation in traditional linear low-voltage current limit control schemes, this paper proposes a nonlinear AC/DC-VDCOL control strategy based on weighted averaging. This strategy adjusts system operating characteristics by using a weighted average of DC voltage and AC commutation voltage as input. Simulation results demonstrate that the designed control strategy can effectively limit DC current during fault conditions, shorten the recovery process from commutation failures, suppress consecutive commutation failures, maintain power transmission on system lines, and enhance the system's transient recovery characteristics. Based on the improved nonlinear VDCOL control strategy, a coordinated recovery strategy for dynamic amplitude-limited voltage support based on the receiving-end grid's Voltage Source Converter (VSC) is further designed. This strategy employs a reactive power priority control method to enhance the reactive power output capability of the VSC converter station. Simulation results indicate that the coordinated recovery strategy combining the improved nonlinear VDCOL on the DC side with the dynamic amplitude limitation of the VSC on the AC side can effectively suppress consecutive commutation failures, facilitate faster system recovery, provide voltage support to the system, and further enhance the stability of the HVDC system and the receiving-end grid.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"14 ","pages":"Pages 2236-2246"},"PeriodicalIF":5.1,"publicationDate":"2025-09-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145019061","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Energy consumption forecasting of neighborhood living facilities: A panel regression approach 社区生活设施能耗预测：面板回归方法

IF 5.1 3区工程技术

Energy Reports Pub Date : 2025-09-08 DOI: 10.1016/j.egyr.2025.08.048

Jaemoon Kim , Jong Ho Hong , Jitae Kim

{"title":"Energy consumption forecasting of neighborhood living facilities: A panel regression approach","authors":"Jaemoon Kim , Jong Ho Hong , Jitae Kim","doi":"10.1016/j.egyr.2025.08.048","DOIUrl":"10.1016/j.egyr.2025.08.048","url":null,"abstract":"<div><div>Energy consumption forecasting plays a crucial role in establishing a plan for Green Remodeling (GR) to achieve carbon neutrality in the building sector. This study aims to forecast greenhouse gas emissions from daycare centers and medical facilities among neighborhood living facilities, which are the primary targets of GR. We use panel regression models to alleviate endogeneity concerns with annual panel data of 66 buildings over a 2-year period. We evaluate the performance of each model by comparing the predicted annual energy consumption with the actual values. The empirical analysis results show that forecasting using panel Generalized Least Squares (GLS), while taking into account the heteroscedasticity observed in our data, resulted in a lower Root Mean Square Error (RMSE = 17,687) compared to other regression models. Furthermore, the GLS model showed comparable performance to AI methods, accurately predicting energy consumption within a ± 30 % error margin in 57.1 % of test cases. Therefore, when predicting building energy consumption, it is considered that analysis through an appropriate regression model not only allows for the inference of causal relationships but also aids in efficient prediction by saving time and costs. This study can be used to assess the effectiveness of GR in achieving the greenhouse gas reduction goal and can contribute to developing an efficient carbon-neutral strategy through GR.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"14 ","pages":"Pages 2191-2203"},"PeriodicalIF":5.1,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145009962","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Addressing photovoltaic (PV) forecasting challenges: Satellite-driven data models for predicting actual PV generation using hybrid (LSTM-GRU) model 解决光伏（PV）预测挑战：使用混合（LSTM-GRU）模型预测实际光伏发电的卫星驱动数据模型

IF 5.1 3区工程技术

Energy Reports Pub Date : 2025-09-04 DOI: 10.1016/j.egyr.2025.08.034

Teklebrhan Negash , Nahom Weldemikael , Merhawi Ghebregziabiher , Yemane Tedla , Seres István , Farkas István

{"title":"Addressing photovoltaic (PV) forecasting challenges: Satellite-driven data models for predicting actual PV generation using hybrid (LSTM-GRU) model","authors":"Teklebrhan Negash , Nahom Weldemikael , Merhawi Ghebregziabiher , Yemane Tedla , Seres István , Farkas István","doi":"10.1016/j.egyr.2025.08.034","DOIUrl":"10.1016/j.egyr.2025.08.034","url":null,"abstract":"<div><div>This study proposes a robust approach for predicting actual PV generation in data-scarce regions using satellite-derived inputs, addressing key limitations in current forecasting models. Its novelty lies in applying a modified z-score transformation to bridge the distribution gap between satellite-derived and measured PV generation by introducing a clear and transparent empirical relationship between the two data sets. The effectiveness of the proposed approach is rigorously validated across a diverse set of well-established models (XGBoost, SARIMAX, CNN-LSTM, LSTM-GRU, and informer) through three distinct scenarios using 17 years of PVGIS satellite data and one year of measured PV generation data from Areza, Eritrea. The Informer model consistently outperformed others, underscoring its suitability for complex forecasting tasks, while traditional models showed low performance. The first scenario, which uses satellite-derived data for both training and testing, serves as a baseline to verify model performance and reliability under consistent conditions. In scenarios 2 and 3, actual PV generation was forecasted using models trained on satellite-derived data without and with modified z-score transformation, respectively. The transformed data (scenario 3) yielded promising accuracy, achieving an enhancement by up to 43 % in R<sup>2</sup> compared to the untransformed case (scenario-2). Furthermore, results showed that the prediction error difference between the first and third scenarios was only 0.69 %, indicating a nearly negligible disparity. Notably, data transformation improves forecasting accuracy across all models, demonstrating the approach’s robustness and effectiveness in data-scarce regions. The findings provide practical guidance for researchers, system operators, and policymakers aiming to scale PV integration in data-scarce regions.</div></div>","PeriodicalId":11798,"journal":{"name":"Energy Reports","volume":"14 ","pages":"Pages 2141-2156"},"PeriodicalIF":5.1,"publicationDate":"2025-09-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144988317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0