Sustainable Energy Grids & Networks最新文献

{"title":"Predictive demand analytics and machine learning in electric power systems for enhancing resilience and efficiency","authors":"Wadim Strielkowski ,&nbsp;Andrey Vlasov ,&nbsp;Kirill Selivanov ,&nbsp;Aleksandr Rasuk ,&nbsp;Luboš Smutka","doi":"10.1016/j.segan.2025.101722","DOIUrl":"10.1016/j.segan.2025.101722","url":null,"abstract":"<div><div>Rapid advancements of the Internet of Things (IoT), Artificial Intelligence (AI), cloud computing, and Big Data have significantly accelerated the adoption of predictive analytics within electric power systems. The integration of predictive analytics offers substantial opportunities for automating control and monitoring processes, thereby enhancing both the resilience and operational efficiency of power grids. This paper introduces a novel predictive analytics framework that uniquely integrates supervised and unsupervised machine learning methods, specifically linear and logistic regression, decision trees, random forests, and clustering algorithms, to simultaneously predict short-term power demand and accurately detect early signs of short circuits and system faults. Utilizing the grid load data from the U.S. Department of Energy's Open Energy Data Initiative (OEDI), our research systematically illustrates the implementation, optimization, and integration of selected machine learning algorithms specifically tailored for power systems. Our empirical results demonstrate substantial efficiency improvements in electric power systems ranging from 14 % to 24 %, with measurable enhancements across reliability indices, economic savings, reductions in environmental impact (lower greenhouse gas emissions), and optimized infrastructure utilization. Furthermore, the paper explicitly addresses regulatory hurdles and industry adoption challenges, outlining how predictive analytics can strategically facilitate technology integration in traditionally conservative power sectors. Finally, the paper provides deeper theoretical synthesis and proposes several specific future research avenues, emphasizing scalability to diverse grid contexts, renewable energy integration, and further exploration of regulatory dynamics. Overall, this study not only highlights the practical benefits of predictive analytics but also significantly contributes to theoretical advancements, strategic planning, and informed policymaking within the energy sector.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"42 ","pages":"Article 101722"},"PeriodicalIF":4.8,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143899317","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Assessing the robustness of machine learning-based voltage calculations for LV networks","authors":"Orlando Pereira ,&nbsp;Vincenzo Bassi ,&nbsp;Tansu Alpcan ,&nbsp;Luis F. Ochoa","doi":"10.1016/j.segan.2025.101716","DOIUrl":"10.1016/j.segan.2025.101716","url":null,"abstract":"<div><div>The integration of distributed energy resources (DERs) into low-voltage (LV) distribution networks requires distribution companies to assess customer voltages for new scenarios involving larger generation (from solar PVs) or increased demand (from electric vehicles [EVs]). Voltage calculations typically depend on power flow analyses, which require detailed three-phase electrical models that are often unavailable for LV networks. As an alternative, machine learning (ML) models can leverage historical smart meter data (active power [P], reactive power [Q] and voltage magnitudes [V]) to capture the underlying physics of the LV network and calculate customer voltages for new scenarios without relying on electrical models. However, their robustness in scenarios beyond their training scope is often overlooked, leading to errors in determining how much DER capacity an LV network can handle. This paper evaluates the robustness of ML-based voltage calculations using Neural Networks (NNs) and Linear Regression (LR), in scenarios that remain within (in-domain) and beyond (out-of-domain) the historical data ranges, such as having more solar PV or EVs. A voltage sensitivity analysis assesses each model’s ability to capture the network’s response to changes in P and Q. The study uses synthetic data from a realistic Australian LV network comprising 31 single-phase customers and 25 % PV penetration. Results indicate that LR models calculate voltages more accurately than NNs, especially in out-of-domain scenarios, although all models exhibit limitations in capturing the network’s sensitivity to P and Q. These findings highlight the need for improving ML models to ensure reliable voltage calculations for applications involving DER integration.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"42 ","pages":"Article 101716"},"PeriodicalIF":4.8,"publicationDate":"2025-04-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143903857","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Goal-oriented heuristic dynamic programming for scheduling of virtual energy hubs with management of intelligent parking lot","authors":"Qingwen Fu ,&nbsp;Yanbo Ding ,&nbsp;Yongfeng Wang ,&nbsp;Liping Yu","doi":"10.1016/j.segan.2025.101718","DOIUrl":"10.1016/j.segan.2025.101718","url":null,"abstract":"<div><div>The increasing complexity of multi-carrier energy systems (MCESs) has introduced significant challenges in the current energy systems necessitates seamless integration and optimization. To address this issue, this work proposes an effective configuration of a virtual energy hub (VEH) to schedule and make multi-energy systems active inside energy markets. In particular, the framework of VEH dynamically schedules the system in such a way that addresses the management of intelligent parking lot (IPL) alongside the MCESs context. The proposed structure of VEH not only facilitates the optimal operation of MCES but also formulates an advanced model to integrate the EV unit into the studied energy plant. This model of VEH is effectively developed to operate within both the thermal and electrical market where it brings more flexibility and efficient energy interactions. To demonstrate the real-world applicability of the proposed structure, this model considers the inherent uncertainties corresponding to the EV behavior, including stochastic changes in their arrival and departure times, and their state of charge (SOC). Furthermore, the electrical energy generated by renewable resources and energy prices have an uncertainty factor which imposes more complexity to the model of VEH. The Goal-Oriented Heuristic Dynamic Programming (Go-HDP) is adopted to solve the scheduling problem to reach the maximum profit from the integrated system. In this framework, the Go-HDP with multi-neural nets (goal, critic, and action nets) is utilized to dynamically respond to the system requirements by interacting as an agent with the environment. By maximizing a reward function that is defined based on the system characteristics, the neural nets of Go-HDP are trained in the iterative process. The comprehensive simulation examinations under typical scenarios of the virtual hub are made to ascertain the feasibility of the proposed framework.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"42 ","pages":"Article 101718"},"PeriodicalIF":4.8,"publicationDate":"2025-04-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143899318","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Time-varying probabilistic models for incipient fault in underground cables","authors":"Zahra Hosseini ,&nbsp;Haidar Samet ,&nbsp;Masoud Jalil ,&nbsp;Teymoor Ghanbari ,&nbsp;Mehdi Allahbakhshi","doi":"10.1016/j.segan.2025.101717","DOIUrl":"10.1016/j.segan.2025.101717","url":null,"abstract":"<div><div>The incipient faults in underground cables are mainly caused by cable insulation failure, defects in splices, and water penetration. Incipient fault modeling is essential to ensure the algorithms' performance and accuracy in detecting incipient faults or generating data under various conditions. This article aims to create and develop a robust yet practical model for incipient faults by considering actual recorded data. Experimental records derived from a laboratory setup are used in the models' identification procedure. Considering that there is no arc model for the incipient fault in underground cables, this article concentrates on driving effective models based on Schwarz equations for incipient fault using actual recorded data. Three modified Schwarz models for modeling the voltage and current of incipient faults in cables are presented. In the proposed models, the idea of time-varying parameters is used to show the time-varying properties of incipient faults. The models' parameters are updated using the least squares method for each cycle of power frequency. The best order of each model is determined using two error indices. Since the model parameters change in every cycle, probability distribution functions (PDFs) were used to show the stochastic behavior of the parameters. As a result, several PDFs are examined for every set of the model's parameters, and the one that best fits the actual data is selected.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"42 ","pages":"Article 101717"},"PeriodicalIF":4.8,"publicationDate":"2025-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143881841","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A coordinated charging/discharging strategy for EVs with flexible mobility in multi-temporary microgrids","authors":"Zhixian Wang ,&nbsp;Davide Falabretti ,&nbsp;Ying Wang ,&nbsp;Kaifeng Zhang","doi":"10.1016/j.segan.2025.101720","DOIUrl":"10.1016/j.segan.2025.101720","url":null,"abstract":"<div><div>The increasing occurrence of extreme weather events can lead to more major power blackouts, which attracts attention to the utilization of temporary microgrids (TMGs) in the restoration stage. However, single TMGs originating after a blackout usually have weak resilience, because of their limited available resources. Therefore, cooperation between microgrids needs urgently to be studied: to this end, this paper proposes a novel charging/discharging strategy for electric vehicles (EVs) to achieve optimal power coordination between TMGs, by taking advantage of EV’s temporal-spatial flexible mobility. First, a new user response willingness model for EVs is established considering move distance and charging compensation, and this model is combined with the EV eligibility assessment based on traveling time to evaluate the EV mobile possibility across TMGs. Then, a suitable resilience factor is proposed to measure the resilience of TMGs considering both the current and future operation conditions. Finally, a novel coordinated EV charging/discharging strategy exploiting the EVs’ mobility is developed based on a mixed integer optimization model, to enhance system resilience and reduce total regulation costs. A case study based on the realistic topology of Milan’s urban area is analyzed, comparing the results obtained with the strategy without EV spatial cooperation. As a result, the proposed strategy can enhance the resilience of TMGs by 20.93 % and reduce the total regulation cost which is paid to regulation resources for ensuring the power balance of each TMG by 8.43 %.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"42 ","pages":"Article 101720"},"PeriodicalIF":4.8,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143885979","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Sustainable local energy communities: The role of peer-to-peer trading, EVs, and RECs on social welfare and emissions","authors":"Elham Mokaramian ,&nbsp;Vito Calderaro ,&nbsp;Vincenzo Galdi ,&nbsp;Giuseppe Graber ,&nbsp;Lucio Ippolito ,&nbsp;Pierluigi Siano","doi":"10.1016/j.segan.2025.101715","DOIUrl":"10.1016/j.segan.2025.101715","url":null,"abstract":"<div><div>Peer-to-Peer (P2P) energy trading is known as a decentralized method of energy management in local energy communities (LECs), which allows consumers and prosumers to directly exchange energy. This model improves resource distribution, balances local demand and supply, and integrates renewable energy sources (RES) and distributed generation (DG), inducing energy independence and sustainability. This study introduces a novel P2P energy trading model for LECs that addresses three main objectives: maximizing welfare, minimizing environmental emissions, and minimizing grid consumption due to high costs and emissions. The proposed model includes RES, electric vehicles (EVs), charging stations, DG, and flexible storage, combined with a multi-objective approach for optimal energy management. In our study, the LEC has been clustered into three zones (residential, commercial, and industrial) each with specific energy needs and resources. These LECs allow for customized energy plans while fostering collaboration across sectors. The model also integrates EV charging stations, hydrogen-based systems (fuel cells and electrolyzers), and distributed electric storage to ensure efficient energy use. Moreover, centralized and decentralized storage and DG systems, enabling seamless energy exchanges both within and across zones, are considered. This cross-zone interaction, facilitated by the P2P trading energy enhances flexibility, optimizes resource use, and promotes energy autonomy. Additionally, the model integrates real-time energy management, allowing prosumers to dynamically manage energy consumption, storage, and trading. The flexibility of P2P exchanges between batteries, DGs, and EVs further improves efficiency, adaptability, and sustainability, making the system more resilient and environmentally friendly. To prove the superiority of the proposed method, three scenarios are considered as an independent operation of LECs, LECs equipped with batteries, and LECs utilizing P2P energy trading with batteries. P2P trading significantly reduces grid consumption by 2.7 % from Scenario 1–2 and 13.77 % from Scenario 2–3. Emissions are also reduced by 2.73 % between Scenario 1 and 2, and a further 13.77 % between Scenario 2 and 3.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"42 ","pages":"Article 101715"},"PeriodicalIF":4.8,"publicationDate":"2025-04-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143881840","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Distribution system state estimation for system identification and network model validation: An experience on a real low voltage network","authors":"Marta Vanin ,&nbsp;Reinhilde D’hulst ,&nbsp;Dirk Van Hertem","doi":"10.1016/j.segan.2025.101710","DOIUrl":"10.1016/j.segan.2025.101710","url":null,"abstract":"<div><div>Distribution network data in utility databases are known to present multiple issues that may lead to problematic results when used in physics-based engines, e.g., leading to constraint violations in (optimal) power flow. This paper discusses the application of state and parameter estimation methods to a real low voltage network, where power and voltage time series from digital meters are used to improve the utility’s network data. Good input data are crucial for the advanced decision support tools that are needed to manage networks with increased shares of low carbon technology.</div><div>Conventional state and parameter estimation methods leverage measurements from a single (or few) time stamp(s) to detect sparse, local data errors or sudden changes in the system (e.g., a line being de-energized). The methods in this paper differ in that their goal is to estimate “historical” states and reconstruct system parameters from scratch for <em>all</em> users and branches. This is possible through the augmentation of conventional state vectors (i.e., voltage phasors) to include asset properties (e.g., phase connectivity), and binding the asset states as time-independent throughout the time series.</div><div>Discussions of real-life experiences are uncommon, but valuable to highlight the differences between working with synthetic or field data. For example, the main contribution of this work rests in exploring the use of state estimation for the statistical validation of data-driven models for real networks, for which the ground-truth is not available (contrary to the case of synthetic data).</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"42 ","pages":"Article 101710"},"PeriodicalIF":4.8,"publicationDate":"2025-04-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143869827","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Expansion planning via decomposition to achieve fully renewable power and freshwater systems","authors":"Mubarak J. Al-Mubarak ,&nbsp;Antonio J. Conejo","doi":"10.1016/j.segan.2025.101713","DOIUrl":"10.1016/j.segan.2025.101713","url":null,"abstract":"<div><div>As the reliance on electricity for producing freshwater continues to grow, the development of an expansion planning model that captures the interdependency between power and freshwater systems becomes increasingly important. This paper proposes a two-stage stochastic expansion planning model that represents the interdependence of these systems and accounts for the uncertainties involved. The first stage represents investments for achieving fully renewable power and freshwater systems, while the subsequent stage represents the operation of both systems. The model accounts for both long-term uncertainties, which pertain to growth in power and freshwater demands, and short-term uncertainties, which pertain to the daily fluctuations in freshwater and power demands as well as in renewable production. Due to the complexity of representing the operation of both systems under numerous operating conditions, expansion planning models often become computationally burdensome. To reduce the computational burden, we propose an effective partitioning technique that relies on Benders’ decomposition, dividing the expansion planning problem into a sufficiently small master problem and numerous subproblems. To enhance convergence, we incorporate the operation constraints pertaining to the worst operating condition into the master problem. Numerical experiments underscore the efficacy of utilizing the proposed technique to solve the expansion planning of large-scale systems.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"42 ","pages":"Article 101713"},"PeriodicalIF":4.8,"publicationDate":"2025-04-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143859031","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Day-ahead joint market operation strategy of grid-connected wind farms with flexible allowable generation deviation rates","authors":"Tianhui Meng,&nbsp;Jilai Yu,&nbsp;Yufeng Guo","doi":"10.1016/j.segan.2025.101714","DOIUrl":"10.1016/j.segan.2025.101714","url":null,"abstract":"<div><div>The uncertainty of wind power output affects the efficient operation of the electricity spot market and has become a key factor restricting the participation of wind farms in the market. To this end, this paper proposes a day-ahead joint market operation strategy that considers allowable deviation rates of wind power output. Unlike traditional electricity markets which impose uniform deviation requirements on all wind farms, the main grid side provides a more diverse range of selectable deviation rates. The bidding strategy for wind farms in the joint day-ahead and balancing markets is explored, allowing them to independently select deviation rates and submit schedule curves and offer prices. A joint clearing model for the day-ahead energy-reserve and balancing market is established, incorporating the carbon emission trading costs of thermal power units, with the aim of minimizing the system operating cost. Numerical results indicate that compared with the traditional market participation method, the proposed strategy not only encourages wind farms to improve output accuracy, but also reflects the market economic principle of high quality and high price. Meanwhile, integrating carbon emission trading costs into the model helps to reduce carbon emissions while ensuring the economic operation of the system.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"42 ","pages":"Article 101714"},"PeriodicalIF":4.8,"publicationDate":"2025-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143854935","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Probabilistic Load Forecasting of distribution power systems based on empirical copulas","authors":"Pål Forr Austnes ,&nbsp;Celia García-Pareja ,&nbsp;Fabio Nobile ,&nbsp;Mario Paolone","doi":"10.1016/j.segan.2025.101708","DOIUrl":"10.1016/j.segan.2025.101708","url":null,"abstract":"<div><div>Accurate and reliable electricity load forecasts are becoming increasingly important as the share of intermittent resources in the system increases. <em>Distribution System Operators</em> (DSOs) are called to accurately forecast their production and consumption to place optimal bids in the day-ahead market. Violations of their dispatch-plan requires activation of reserve-power which has a direct cost for the DSO, and also necessitates available reserve-capacity. Forecasts must account for the volatility of weather-parameters that impacts both the production and consumption of electricity. If DSO-loads are small or lower-granularity forecasts are needed, parametric statistical methods may fail to provide reliable performance since they rely on a priori statistical distributions of the variables to forecast. In this paper, we introduce a <em>Probabilistic Load Forecast</em> (PLF) method based on Empirical Copulas (ECs). The model is data-driven, does not need a priori assumption on parametric distribution for variables, nor the dependence structure (copula). It employs a kernel density estimate of the underlying distribution using beta kernels that have bounded support on the unit hypercube. The method naturally supports variables with widely different distributions, such as weather data (including forecasted ones) and historic electricity consumption, and produces a conditional probability distribution for every time step in the forecast, which allows inferring the quantiles of interest. The proposed non-parametric approach differs significantly from previous forecasting methods based on copulas, which typically uses copulas to model hierarchical dependence. Our approach is highly flexible and can produce meaningful forecasts even at very low aggregated levels (e.g. neighborhoods). The bandwidth of the beta kernel density estimators is optimized using <em>Integrated Square Error</em> (ISE) and such optimization can be performed online (i.e. without knowing the realization). We also investigate rule-of-thumb and <em>Quantile Loss</em> (QL) as objectives for the bandwidth-optimization. We present results from an open dataset and showcase the strength of the model with respect to <em>Quantile Regression</em> (QR) using standard probabilistic evaluation metrics.</div></div>","PeriodicalId":56142,"journal":{"name":"Sustainable Energy Grids & Networks","volume":"42 ","pages":"Article 101708"},"PeriodicalIF":4.8,"publicationDate":"2025-04-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143873808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}