Dalarna University's logo and link to the university's website

The sizing of solar district heating (SDH) systems with pit thermal energy storage (PTES) and heat pumps (HP) is conventionally performed using detailed tools (e.g TRNSYS, Modelica etc). While accurate, these simulations are computationally intensive and unsuitable for large parametric studies due to large simulation time. Traditional thumb rules, such as fixed collector-to-storage ratios expressed in m²/m³, are often used as a shortcut but become unreliable when the load profile or integration constraints deviate from standard assumptions. For example, when PTES discharge is limited by the pipe diameter at the integration point, the achievable solar fraction and storage utilization can differ substantially from conventional design guidelines. Such conditions necessitate case-by-case evaluation, for which a fast sensitivity tool can provide valuable guidance. We propose a two-stage methodology. In the first stage, a Python-based model is developed which integrates solar collectors, pit thermal energy storage, and a heat pump, resolving hourly boundary conditions with sub-hourly thermo-hydraulic dynamics. The tool performs random sampling across a million combinations of component’s sizes which enables rapid evaluation of energy balances and techno-economic indicators. In the second stage, only the most promising configuration(s) are re-simulated in TRNSYS, where collector physics, diffuser layouts, hydraulics, and soil–storage coupling are fully represented. This workflow provides a computationally efficient yet physically consistent approach for robust SDH system sizing, offering a faster alternative to direct TRNSYS optimization while maintaining reliability.

Electric vehicles (EVs) and solar photovoltaic (PV) systems pose challenges to existing power grids, which were often not designed to accommodate such large loads or local generation. With growing adoption of these technologies, hosting capacity (HC) studies are essential for understanding their impact and supporting future planning. However, combined HC assessments of EVs and PVs remain limited compared to studies treating the technologies separately, particularly when accounting for spatial uncertainties, temporal correlations, varying implementation levels, and different charging strategies. This study addresses this gap by applying a broadly applicable method for calculating the HC of EVs and PV systems in a Swedish residential grid. A combined time-series and stochastic approach was used across multiple scenarios, supported by geographic information system-based (GIS-based) PV potential analysis. The analysis is based on real-world data and performed as a case study, providing new insights into joint EV-PV integration. Results show that spot price-based charging resulted in the most frequent violations. Evenly distributed charging was the most grid-friendly strategy in winter, while PV generation caused significant violations in summer regardless of charging strategy. Under the even charging strategy, the average number of unique buses with voltage violations during summer was 11–63 times higher than during winter. This is because the strategy uses low power, limiting the impact of EV charging, while high uncontrolled PV production in summer becomes the main driver of voltage problems. The findings highlight the need for strategic charging strategies and additional measures to manage the grid impact of high PV implementation. © 2026 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license. http://creativecommons.org/licenses/by/4.0/

Large-scale electric-vehicle (EV) uptake is challenging the power grid due to a lack of sufficient hosting capacity. Most smart-charging studies still treat EVs as stationary loads in one location and ignore their mobility. This work closes that gap by evaluating an EV-based virtual electricity network (EVEN) that lets vehicles charge at one feeder and discharge at another. We formulate a comprehensive framework that combines inter-network energy-delivery optimization, stochastic time-series hosting capacity analysis, and battery-degradation assessment. The approach is tested on two real-world distribution systems: a voltage-constrained 50-bus rural residential network and a capacity-rich 76-bus industrial network. Simulation results reveal that shifting only 10% of the evening demand from the rural to industrial network cuts the rural undervoltage index by roughly 80% and weekly violation counts by 65%, while adding no more than two minor violations upstream. At this level, average yearly battery-cycling degradation rises modestly from 0.4% to 0.8%. The study thus demonstrates EVEN as a cost-effective, scalable alternative to physical reinforcement and provides the first integrated assessment linking network-level benefits with battery-health impacts. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Electric vehicle (EV) smart charging can support the grid and improve owner economics. Simultaneously, European network tariffs are transitioning from energy-based network tariffs (EBTs) to power-based network tariffs (PBTs). Existing studies rarely distinguish between EBTs and newer tariff designs when evaluating EV smart charging strategies. To bridge this gap, this study compares the technical and economic consequences of shifting from EBTs to PBTs for Swedish prosumers with EVs and rooftop photovoltaics (PV) systems under 5 distinct PBTs introduced for 2025. The optimization framework includes three economic objectives, comprising: total electricity cost, EV battery degradation, and ancillary service revenues. In addition, one technical optimization is used to minimize net load variance. Average results show that switching from EBT to PBT increases electricity costs by 12.5% under uncontrolled EV charging. Nevertheless, using EV smart charging results in net cost reduction percentage (CR) of 2.5–16% for EBT and 7–34% for PBT without the ancillary services (AS) participation, and 6–32.5% for EBT and 10–41% for PBT with the AS participation. The mean CR is nearly double under PBT compared to EBT across all optimized cases, with average CR of 23.6% and 12.2%, respectively. The majority of this reduction is attributable to lower power-based charges by rescheduling EV charging demand. CR is greater under PBT, particularly for smaller fuse ratings, higher production-to-load ratios, and the monthly single power peak rather than the multi-peak averaging count method. Although economic performance differs between tariff structures, technical metrics remain similar. The sensitivity analysis indicates that costs increase more significantly under PBT when the PV system is not implemented or when the EV arrives in the evening. Thus, these findings underscore the critical role of active load management in mitigating net electricity costs under PBT. © 2026 The Authors

The integration of electric vehicles (EVs) into power systems offers a promising pathway to enhance energy resilience during grid outages. Most of the existing studies treat EV outage support as a short-term, technical problem, with limited attention to economics, business models (e.g., ownership and stakeholders), or long-term outages. To bridge the gap, this study presents a comprehensive framework to evaluate the techno-economic feasibility of using EV mobility as a decentralized electricity delivery solution during both short-term and long-term outages. A rule-based control strategy is developed to operate EVs under realistic operational constraints, with the primary objective of maximizing energy delivery and minimizing unmet demand during outages. The framework incorporates real-world data, including household and critical load profiles, EV battery capacities, ownership types, outage durations, travel distances, and seasonal variations. Scenario-based simulations are performed to compare various technical and economical key performance indicators. Results show that EVs with personal ownership type and large capacity (100 kWh) can effectively meet residential energy needs during short-term outages. In long-term outages, expanding the EV fleet effectively reduce unmet demand, and higher solar generation in summer and spring significantly improves energy availability compared to winter, where unmet load is more than twice as high under identical conditions. The findings offer practical insights for stakeholders and policymakers in developing EV-based resilience strategies for future energy systems. Furthermore, the study highlights the importance of designing appropriate business models to support the integration of mobile energy services into resilience planning. © 2025 The Authors.

District heating (DH) systems in Sweden are facing growing challenges due to rising biomass prices and uncertainty surrounding future supply availability. Integrating solar thermal collectors with pit thermal energy storage (PTES) offers a promising pathway for reducing dependence on biomass and improving system sustainability. However, the optimal integration and techno-economic performance of such systems under real-world limitations (such as spatial and hydraulic constraints) remain largely underexplored. This study develops a two-stage modeling framework to evaluate solar + PTES integration while accounting for practical constraints such as sitting feasibility and discharge power limitations. A Python-based simplified model first generates a sensitivity matrix mapping heat delivery performance across a wide range of collector areas and storage volumes, which guides the selection of PTES sizes for subsequent detailed modelling in TRNSYS. Three system configurations are analysed: a real-world case with waste heat and pipe limitations, a generalized case without waste heat but with pipe constraints, and an ideal case without any integration barriers. Monte Carlo simulations and sensitivity analyses are performed to quantify economic uncertainties and identify key influencing factors. Results show that discharge power limitations and pipe constraints significantly penalise the system performance, leading to higher levelized costs of heat (LCOH) compared with the ideal benchmark. The findings highlight the necessity of simultaneous consideration of spatial and techno-economic constraints to achieve cost-effective solar fractions and optimal storage design in real-world DH systems.

The integration of electric vehicles (EVs) into the power infrastructure can be a viable solution to enhance resilience during power outages. This paper proposes a framework to evaluate the techno-economic feasibility of using EV mobility to support household electricity needs during short-term outages under real-world operational limitations. For instance, EVs get charged in functioning power grid and delivers electricity to areas with outages. A rule-based control strategy is developed for energy delivery operation for EVs in order to maximize energy supply as part of a framework. The study assesses performance across key reliability and economic indicators. Simulation results across various scenarios show that personal EVs with sufficient battery capacity can effectively meet household demand in short-term outage scenarios. On the other hand, rented EVs and prolonged outages reduce system efficiency and economic viability. The study can offer valuable insights to stakeholders and policymakers for planning EV-based resilience solutions for future energy systems. © 2025, Scanditale AB. All rights reserved.

The rapid growth of electric vehicles (EVs) is placing new demands on residential distribution networks. Smart charging and Vehicle-to-Grid (V2G) technologies offer potential solutions for mitigating the large peak load and enhancing the grid hosting capacity (HC), yet their effectiveness across varying EV penetration levels remains underexplored. Therefore, this study evaluates how EV penetration levels affect the effectiveness of these two technologies. Furthermore, to help improve the grid performances, this study also pioneeringly proposes two hypothetical pricing settings—diverse electricity buying prices and reduced electricity selling prices—and evaluates their performances by comparing with existing price settings. Using real-world network and EV charging data from Sweden, we assess peak load and HC under different scenarios of power flow direction, charging controls, and electricity prices strategies across eight EV penetration levels. Results reveal that coordinated charging, especially with V2G, more effectively reduces peak loads compared to individual controls. However, V2G, if not regulated well, can increase peak loads at high EV penetration levels. Diversified electricity buying prices help lower aggregated peak loads but are less effective in enhancing local HC due to peak load shifting rather than peak load reduction. Additionally, high electricity selling prices benefit at low EV penetration but become less effective as penetration grows. The findings suggest that electricity pricing strategies and charging controls should adapt dynamically to the level of EV penetration. These insights provide critical guidance to policymakers, distribution system operators, and aggregators in designing adaptive pricing and control strategies to integrate EVs without overburdening the grid.

Accurate electric vehicle (EV) load forecasting is crucial for efficient grid operations and demand-side management, yet it is challenging in data-scarce scenarios. Transfer learning (TL) offers a solution by transferring knowledge from data-rich to data-limited scenarios. However, when the knowledge domain exhibits highly diverse behaviors, applying TL alone could introduce large biases, reducing accuracy and limiting its effectiveness. To address this problem, this study proposes a hybrid deep learning-based framework that integrates TL and K-means clustering. The proposed approach consists of two phases. In the source domain phase, a deep-learning-based model is trained using the full dataset and then fine-tuned using clustered user behaviors. In the target domain phase with limited data, TL is applied to transfer knowledge from the source-domain finetuned cluster models. For validation, the developed prediction method has been tested using real-world datasets and compared with two other cases: one with applying TL from the source-domain base model trained from full dataset, and one without applying TL. Results show the hybrid method improves forecasting accuracy, reducing the normalized root mean squared error by 3.99 % and 8.22 %, respectively. This study establishes a structured approach for targeted knowledge transfer, enhancing prediction accuracy in data-scarce settings. The framework is scalable and adaptable to other energy forecasting applications, supporting sustainable and resilient energy management.