CoChat
Sign Up
Cited by 8Electric Vehicles and InfrastructureSmart Grid Energy ManagementOptimal Power Flow Distribution

A Multi-Agent, Laxity-Based Aggregation Strategy for Cost-Effective Electric Vehicle Charging and Local Transformer Overload Prevention

Open access
Christensen, Kristoffer, Jørgensen, Bo Nørregaard, Ma, Zheng Grace·2025·Sustainability
DOI 10.3390/su17093847
Open this paper in CoChat — free accountOpen PDF

Abstract

The rapid electrification of transportation, driven by stringent decarbonization targets and supportive policies, poses significant challenges for distribution system operators (DSOs). When numerous electric vehicles (EVs) charge concurrently, local transformers risk overloading—a problem that current tariff-based strategies do not adequately address. This paper introduces an aggregator-based coordination mechanism that shifts EV charging from congested to underutilized periods using a rule-based scheduling algorithm. Unlike conventional methods that depend on complex real-time pricing signals or optimization-heavy solutions, the aggregator approach uses a simple yet effective “laxity” measure to prioritize charging flexibility. To assess technical and economic viability, a multi-agent simulation was developed to replicate residential user behavior and DSO constraints under the use of a 400 kVA low-voltage transformer. The results indicate that overloads are completely eliminated with minimal inconvenience to users, whose increased charging costs are offset by the aggregator at an annual total of under DKK 6000—significantly lower than the cost of infrastructure reinforcement. This study contributes by (i) quantifying the compensation needed to prevent large-scale overloads, (ii) presenting a replicable, computationally feasible, rule-based aggregator model for DSOs, and (iii) comparing aggregator solutions to costly transformer upgrades, underscoring the aggregator’s role as a viable tool for future distribution systems.

AI snapshot

Written from the paper’s full text.

TL;DR

A rule-based, laxity-prioritized aggregator shifts residential EV charging to prevent local transformer overload while compensating users for added costs. In a 126-household Danish case with 100% EV adoption, it eliminated 587 annual overload hours for about DKK 6020/year in compensation, far below transformer upgrade costs.

Key findings

  • Baseline decentralized real-time pricing caused 587 h/year of transformer overload on a 400 kVA/400 kW transformer, with a maximum peak of 1342 kW and over 95 h above 200% nominal capacity.
  • The aggregator scenario reduced overload hours to 0, raised load factor from 0.089 to 0.299, and lowered the daily average coincidence factor from 0.826 to 0.301.
  • Average charging cost rose slightly from 0.48 to 0.489 DKK/kWh (+1.8%), emissions rose 0.2% to 0.218 kg/kWh, DSO tariff revenue stayed about 179.5k DKK, and EV user dissatisfaction remained 0.
  • Annual aggregator compensation was about DKK 6020, compared with transformer reinforcement estimates of about DKK 115.8k and new-station costs up to DKK 679.8k; a new 400 kVA transformer was about DKK 240k and a 1600 kVA unit about DKK 589k, excluding installation.

Methods

  • Multi-agent simulation in AnyLogic of a Strib, Denmark, radial low-voltage network with 126 households under a 400 kVA transformer, using 2019 hourly baseload data, 2024 DK1 spot prices, Trefor Tariff Model 3.0, and 100% EV adoption (126 EVs).
  • Two scenarios: baseline decentralized real-time pricing versus an aggregator managed by the DSO that centrally reschedules EV charging when forecast load exceeds transformer capacity.
  • Rule-based algorithm ranks EVs by “laxity” (flexibility before departure), zeroes charging during overload periods, shifts load to available cheaper/lower-load slots, and compensates users for added costs; KPIs include overload hours, load factor, coincidence factor, charging cost, CO2-eq emissions, DSO tariff revenue, and dissatisfaction.

Limitations

  • The paper states assumptions of full and passive user participation, consistent historical user behavior, daily EV use, perfect hourly baseload predictions, no physical grid constraints beyond transformer overloads, and compensation sufficient to ensure compliance.
  • The paper also states that partial participation, inaccurate load prediction, random events, data latency, grid voltage, EV battery health, and standards interoperability are not modeled and should be addressed in future work.
  • Evident from the text: the baseload is hourly 2019 data, PV systems, heat pumps, and existing EVs were removed, only one transformer/feeder and one aggregator are studied, and compensation is treated as a flat cost offset rather than a market-based flexibility valuation.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3390/su17093847

Cites

51 of 51 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

8 of 24 cited sources checked · 1 supported · 1 flagged · accuracy not yet measured

  • Premisecitation

    Most aggregator studies focus on large-scale or regional grid services, with less attention paid to localized overloads at individual transformers.

    “First, most aggregator studies focus on large-scale or regional grid services, with less attention paid to localized overloads at individual transformers [9][10][11].”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites [9] Sun, W., Li, Y., Zhang, K., Zheng, X., and Ye, C. (2025). A Two-Stage Scheduling Method for Virtual Power Plants Considering Ultra-Short-Term Forecasting Errors. J. Phys. Conf. Ser., 2936. source
    • cites [10] Faria, P., Spínola, J., and Vale, Z. (2018). Distributed Energy Resources Scheduling and Aggregation in the Context of Demand Response Programs. Energies, 11. source

      The Introduction (p. 2) describes aggregator management at the distribution-network level, and the Conclusions discuss a different gap in prior work: how resources are grouped and remunerated. Neither passage says that most aggregator studies focus on large-scale or regional services rather than individual-transformer overloads. Searches for transformer overload and local congestion found no passage addressing that comparison. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
    • cites [11] Gao, Y., Zhou, X., Ren, J., Wang, X., and Li, D. (2018). Double Layer Dynamic Game Bidding Mechanism Based on Multi-Agent Technology for Virtual Power Plant and Internal Distributed Energy Resource. Energies, 11. source

      The Discussion (page unavailable) says VPP research mainly focuses on overall external operation, supporting the claim’s broad contrast between external and more localized concerns. But the gap it identifies is internal distributed-energy operation and benefits, not localized overloads at individual transformers. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
  • Premisecitation

    Advanced optimization- or machine-learning-based schedulers require extensive computational resources and forecasting precision, making them less accessible to DSOs seeking near-term deployment.

    “Second, advanced optimization-or machine-learning-based schedulers (for example, those discussed in [12,13]) require extensive computational resources and forecasting precision, making them less accessible to DSOs seeking near-term deployment.”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [12] Lahariya, M., Sadeghianpourhamami, N., and Develder, C. (2022). Computationally Efficient Joint Coordination of Multiple Electric Vehicle Charging Points Using Reinforcement Learning. arXiv. source

      The Abstract says computational challenges could limit deployment of an earlier reinforcement-learning approach, and the Introduction says model-based demand-response algorithms face model uncertainty and scalability limits. The paper also presents a more computationally efficient scheduler, so it does not establish that advanced schedulers generally require extensive resources. It does not establish a need for forecasting precision or discuss near-term accessibility to DSOs. Page numbers are unavailable in our copy. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
    • cites [13] Sadeghianpourhamami, 2019, Definition and Evaluation of Model-Free Coordination of Electrical Vehicle Charging with Reinforcement Learning, IEEE Trans. Smart Grid source

      The Generalization to Larger Scales section (p. 9) reports substantial RL training times and says optimization-based approaches, unlike model-free RL, need accurate knowledge of future EV sessions. The Introduction (p. 1) also describes barriers to widespread deployment of model-based approaches. The paper does not say machine-learning schedulers require forecasting precision or assess accessibility to DSOs seeking near-term deployment. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
  • Findingown data

    The aggregator offsets users' increased charging costs at an annual total of under DKK 6000.

    “The results indicate that overloads are completely eliminated with minimal inconvenience to users, whose increased charging costs are offset by the aggregator at an annual total of under DKK 6000significantly lower than the cost of infrastructure reinforcement.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The aggregator is assumed to be controlled by the DSO.
    • The authors state this as an assumption: The aggregator targets EVs with the highest laxity first, assuming these users can be shifted at minimal inconvenience.
  • Findingown data

    In the baseline scenario with all 126 households owning an EV and using decentralized real-time pricing, the transformer experienced 587 hours of overload over the simulated year.

    “The simulation reveals 587 h of overload throughout the year.”
    • Also stated in: § 5.1. Baseline Scenario: Full EV Adoption and Real-Time Pricing · plain
      “A total of 587 overload hours occurs, highlighting the scale of potential transformer stress.”
    • Also stated in: § 6. Discussion · plain
      “In the baseline scenario, with all EVs following decentralized RTP, the transformer was overloaded for hundreds of hours each year.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: User behavior is consistent and follows historical patterns; users do not override aggregator control signals.
    • The authors state this as an assumption: EVs are driving each day, so each EV has a period each day of not being available for aggregation.
    • The authors state this as an assumption: The simulation assumed that the recorded kilowatt-hour values of each hour equate directly to the load in kilowatts.
    • The authors state this as an assumption: EV adoption was assumed to be 100%, meaning all 126 households own an EV.
    • The authors state this as an assumption: Overloads on the transformer occur whenever the load exceeds the nominal 400 kW capacity, assuming a power factor of 1.
    • The authors state this as an assumption: The baseline scenario assumes decentralized charging follows a real-time pricing (RTP) strategy.
  • Findingown data

    The aggregation scenario has average CO2-eq emissions from charging of 0.218 kg/kWh.

    “| 0 | 0.299 | 0.301 | 0.489 | 0.218 | 179.5k | 0 |”
    Quote in the paperNot a substantial claim
  • The quote only partly states this claim. 1
  • Classified as probably not a substantial claim. 12
  • May combine two claims. 16

  • The quote only partly states this claim.

    Interpretationreasoning

    The multi-agent simulation framework, augmented by a business ecosystem modeling approach, captures micro-level charging decisions and macro-level capacity constraints.

    “By embedding each EV user, the aggregator, and the DSO as separate agents, the model captures micro-level charging decisions and macro-level capacity constraints.”
    Quote in the paperSubstantial claim2 of 2 sources checked
    • cites [20] Ma, Z. (2019). Business ecosystem modeling- the hybrid of system modeling and ecological modeling: An application of the smart grid. Energy Inform., 2. source

      The Ecosystem Modeling section (page unavailable) describes agent-based models as addressing disaggregated parts and micro-level interactions, while system dynamics represents aggregated systems. The Conclusion (page unavailable) proposes a business-ecosystem framework that includes simulation, but our copy does not describe a model with EV users, an aggregator, and a DSO as separate agents, or show it capturing charging decisions and capacity constraints. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
    • cites [21] Ma, Z., Christensen, K., and Jørgensen, B.N. (2021). Business ecosystem architecture development: A case study of Electric Vehicle home charging. Energy Inform., 4. source

      The Future works section says a multi-agent simulation of EV home charging “will be developed,” rather than reporting a model that already captures charging decisions or grid constraints. The paper identifies EV users and DSOs as ecosystem roles, but does not describe embedding them and an aggregator as separate simulation agents. (LLM, conf=0.87)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The laxity metric has not been widely explored as a straightforward aggregator tool for distribution system operators.

    “The laxity metric, while not novel in its scheduling theory [16], has not been widely explored as a straightforward aggregator tool for distribution system operators.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [16] Chen, 2021, Smoothed Least-Laxity-First Algorithm for Electric Vehicle Charging: Online Decision and Performance Analysis with Resource Augmentation, IEEE Trans. Smart Grid source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Interpretationcitation

    The aggregator approach suits DSOs operating under constraints that limit the adoption of fully dynamic tariffs.

    “From a regulatory and operational standpoint, the aggregator approach suits DSOs operating under constraints that limit the adoption of fully dynamic tariffs [4,5].”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [4] Agency, D.E. (2025, February 13). Markedsmodel 3.0: Baggrundsnotat om Netvirksomhedernes Anskaffelse af Fleksibilitetsydelser. Available online: https://ens.dk/sites/default/files/media/documents/2024-10/mm_3.0_baggrundsnotat_om_fleksibilitetsydelser_hos_netvirksomheder.pdf.

      Not checked in the free run.

    • cites [5] Christensen, K., Ma, Z., and Jørgensen, B.N. (2021). Technical, Economic, Social and Regulatory Feasibility Evaluation of Dynamic Distribution Tariff Designs. Energies, 14. source

      The Introduction and §3.4 (page numbers unavailable) explain that DSO tariffs must meet regulatory requirements and that a real-time pricing scheme may fail the transparency requirement. The paper therefore supports the existence of constraints on dynamic tariffs, but it does not say that an aggregator approach suits DSOs facing those constraints; its mention of aggregators concerns forecast parameters. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    In the baseline scenario, over 95 hours exceeded 200% of nominal capacity (800 kW), with a maximum peak of 1342 kW.

    “Over 95 h surpass 200% of the nominal capacity (800 kW), categorized as critical load and with a maximum peak of 1342 kW.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: User behavior is consistent and follows historical patterns; users do not override aggregator control signals.
    • The authors state this as an assumption: EVs are driving each day, so each EV has a period each day of not being available for aggregation.
    • The authors state this as an assumption: The simulation assumed that the recorded kilowatt-hour values of each hour equate directly to the load in kilowatts.
    • The authors state this as an assumption: EV adoption was assumed to be 100%, meaning all 126 households own an EV.
    • The authors state this as an assumption: Overloads on the transformer occur whenever the load exceeds the nominal 400 kW capacity, assuming a power factor of 1.
    • The authors state this as an assumption: The baseline scenario assumes decentralized charging follows a real-time pricing (RTP) strategy.

  • May combine two claims.

    Findingown data

    The minimum compensation cost required to achieve zero overloads for a 400 kVA transformer with full EV adoption is under DKK 6000 annually, significantly lower than the cost of infrastructure reinforcement.

    “The results indicate that overloads are completely eliminated with minimal inconvenience to users, whose increased charging costs are offset by the aggregator at an annual total of under DKK 6000significantly lower than the cost of infrastructure reinforcement.”
    • Also stated in: § 1. Introduction · boosted
      “First, the paper quantitatively establishes the minimum compensation cost required to achieve zero overloads for a 400 kVA transformer with full EV adoption, demonstrating its minimal value relative to the expenditure of upgrading or replacing a transformer station.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The aggregator is assumed to be controlled by the DSO.
    • The authors state this as an assumption: The aggregator targets EVs with the highest laxity first, assuming these users can be shifted at minimal inconvenience.

  • May combine two claims.

    Findingown data

    Overloads are completely eliminated with minimal inconvenience to users.

    “The results indicate that overloads are completely eliminated with minimal inconvenience to users, whose increased charging costs are offset by the aggregator at an annual total of under DKK 6000significantly lower than the cost of infrastructure reinforcement.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The aggregator is assumed to be controlled by the DSO.
    • The authors state this as an assumption: The aggregator targets EVs with the highest laxity first, assuming these users can be shifted at minimal inconvenience.

  • May combine two claims.

    Findingown data

    Most vehicles begin charging in the lowest-priced hour, and some finish charging before peak tariffs.

    “Most vehicles begin charging in the lowest-priced hour, and some finish charging before peak tariffs.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    The baseline scenario has a load factor of 0.089.

    “| 587 | 0.089 | 0.826 | 0.48 | 0.2175 | 179.5k | 0 |”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Under the aggregator strategy, simultaneous charging seldom exceeds 40 EVs, whereas the baseline allows up to 126 EVs to initiate charging in the same period.

    “Under the aggregator strategy, simultaneous charging seldom exceeds 40 EVs, whereas the baseline allows up to 126 EVs to initiate charging in the same period.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Relative to baseline, the aggregation scenario achieved -100% overload, a 235% improvement in load factor, and a 63.6% decrease in coincidence factor.

    “Relative changes from the baseline are shown in the lower part of Table 6: -100% overload, a 235% improvement in the LF, and a 63.6% decrease in the CF.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: User behavior is consistent and follows historical patterns; users do not override aggregator control signals.
    • The authors state this as an assumption: EVs are driving each day, so each EV has a period each day of not being available for aggregation.
    • The authors state this as an assumption: All EV owners participate in the aggregator scheme; partial or voluntary participation is not modeled.
    • The authors state this as an assumption: Perfect hourly baseload predictions; load predictions are fully accurate.
    • The authors state this as an assumption: No physical grid constraints beyond transformer overloads are modeled (voltage fluctuations, phase imbalances, and power quality issues are excluded).
    • The authors state this as an assumption: Compensation payments are sufficient to incentivize user compliance; users accept schedule shifts as long as their costs are covered.
    • The authors state this as an assumption: The simulation assumed that the recorded kilowatt-hour values of each hour equate directly to the load in kilowatts.
    • The authors state this as an assumption: EV adoption was assumed to be 100%, meaning all 126 households own an EV.
    • The authors state this as an assumption: Overloads on the transformer occur whenever the load exceeds the nominal 400 kW capacity, assuming a power factor of 1.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingnone given

    In the baseline scenario, the CO2-equivalent emissions factor was 0.2175 kg/kWh.

    “The CO2-equivalent emissions factor is 0.2175 kg/kWh”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: User behavior is consistent and follows historical patterns; users do not override aggregator control signals.
    • The authors state this as an assumption: EVs are driving each day, so each EV has a period each day of not being available for aggregation.
    • The authors state this as an assumption: The simulation assumed that the recorded kilowatt-hour values of each hour equate directly to the load in kilowatts.
    • The authors state this as an assumption: EV adoption was assumed to be 100%, meaning all 126 households own an EV.
    • The authors state this as an assumption: Overloads on the transformer occur whenever the load exceeds the nominal 400 kW capacity, assuming a power factor of 1.
    • The authors state this as an assumption: The baseline scenario assumes decentralized charging follows a real-time pricing (RTP) strategy.

  • May combine two claims.

    Premisecitation

    Danish distribution tariffs are currently restricted to time-of-use schemes that often remain static or insufficiently granular, leading to persistent local congestion when consumer responses cluster.

    “Thus, Danish distribution tariffs are currently restricted to time-of-use schemes that often remain static or insufficiently granular, leading to persistent local congestion when consumer responses cluster [3,6].”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [3] Santos, M.F., Machado, J., Novais, P., Cortez, P., and Moreira, P.M. (2025). Multi-agent Based Simulation for Decentralized Electric Vehicle Charging Strategies and Their Impacts. Progress in Artificial Intelligence, Springer Nature Switzerland.

      Not checked in the free run.

    • cites [6] Regulator, D.U. (2025, April 13). Elnetvirksomhedernes Tariffer for Husholdninger. Available online: https://forsyningstilsynet.dk/Media/638536031032121805/Elnetvirksomhedernes%20tariffer%20for%20husholdninger.pdf.

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S2 (checkable), S4 (based).

    Interpretationreasoning

    The multi-agent platform is particularly well suited to evaluating emergent behavior, as local congestion events often arise unexpectedly when many agents follow similar price signals.

    “The multi-agent platform is particularly well suited to evaluating emergent behavior, as local congestion events often arise unexpectedly when many agents follow similar price signals.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    The synergy between business ecosystem modeling and agent-based simulation is uniquely suited to capture the interplay of technical, economic, and social factors in EV home charging aggregation.

    “Hence, the synergy between business ecosystem modeling and agent-based simulation is uniquely suited to capture the interplay of technical, economic, and social factors in EV home charging aggregation.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    The proposed rule-based aggregator model is replicable and computationally feasible for DSOs.

    “presenting a replicable, computationally feasible, rule-based aggregator model for DSOs”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: The aggregator targets EVs with the highest laxity first, assuming these users can be shifted at minimal inconvenience.

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Interpretationnone given

    The rule-based model avoids heavy computational overhead and extends more naturally to real-time operations.

    “The rule-based model avoids heavy computational overhead and extends more naturally to real-time operations, making it promising for DSOs needing immediate solutions.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    The load factor and coincidence factor were computed following the International Electrotechnical Commission definitions.

    “The load factor (LF) and coincidence factor (CF) follow the International Electrotechnical Commission definitions [74,75]”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [74] International Electrotechnical Comission (2022, January 12). Load Factor. Available online: https://www.electropedia.org/iev/iev.nsf/display?openform&ievref=691-10-02.

      Not checked in the free run.

    • cites [75] International Electrotechnical Comission (2022, January 12). Coincidence Factor. Available online: https://www.electropedia.org/iev/iev.nsf/display?openform&ievref=691-10-03.

      Not checked in the free run.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    The assessment of each scenario used Key Performance Indicators adapted from prior work.

    “The assessment of each scenario used Key Performance Indicators (KPIs) adapted from [3,63]”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [3] Santos, M.F., Machado, J., Novais, P., Cortez, P., and Moreira, P.M. (2025). Multi-agent Based Simulation for Decentralized Electric Vehicle Charging Strategies and Their Impacts. Progress in Artificial Intelligence, Springer Nature Switzerland.

      Not checked in the free run.

    • cites [63] Jørgensen, B.N., da Silva, L.C.P., and Ma, Z. (2024). Multi-agent Based Simulation for Investigating Electric Vehicle Adoption and Its Impacts on Electricity Distribution Grids and CO2 Emissions. Energy Informatics, Springer Nature.

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    Hence, aggregator-based coordination presents a cost-effective alternative that can defer upgrades while adapting to evolving load patterns from future DERs.

    “Hence, aggregator-based coordination presents a cost-effective alternative that can defer upgrades while adapting to evolving load patterns from future DERs.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: All EV owners participate in the aggregator scheme; partial or voluntary participation is not modeled.
    • The authors state this as an assumption: Compensation payments are sufficient to incentivize user compliance; users accept schedule shifts as long as their costs are covered.

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductcitation

    The simulation recorded overload any time the power surpassed 400 kW and collected statistics on the overload size, aligning with the recommended guidelines in IEC 60076-7.

    “the simulation recorded overload any time the power surpasses 400 kW and collected statistics on the overload size, aligning with the recommended guidelines in IEC 60076-7 [73].”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [73] (2018). Power Transformers–Part 7: Loading Guide for Mineral-Oil-Immersed Power Transformers (Standard No. IEC 60076-7). Available online: https://webstore.iec.ch/en/publication/34351.

      Not checked in the free run.

    • The authors state this as an assumption: Overloads on the transformer occur whenever the load exceeds the nominal 400 kW capacity, assuming a power factor of 1.

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    By handling congestion via a targeted, rule-based method, DSOs can reduce the complexity of user engagement.

    “By handling congestion via a targeted, rule-based method, DSOs can reduce the complexity of user engagement.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    The aggregator strategy has the capability to limit concurrency and reduce peak load.

    “This difference demonstrates the aggregator's capability to limit concurrency and reduce peak load.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationown data

    The total annual compensation payment is on the order of a few thousand Danish kroner-an amount that is minimal compared to the tens or hundreds of thousands often required for expanding physical infrastructure.

    “While some EVs experience marginally higher electricity prices due to load shifts, the total annual compensation payment is on the order of a few thousand Danish kroner-an amount that is minimal compared to the tens or hundreds of thousands often required for expanding physical infrastructure.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: All EV owners participate in the aggregator scheme; partial or voluntary participation is not modeled.
    • The authors state this as an assumption: Compensation payments are sufficient to incentivize user compliance; users accept schedule shifts as long as their costs are covered.

  • Probably not a substantial claim: may not pass S4 (based).

    Interpretationnone given

    The agent-based framework captures both user behavior and grid constraints with high fidelity.

    “The use of a laxity measure enables prioritization without advanced optimization, while the agentbased framework captures both user behavior and grid constraints with high fidelity.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Given an annual aggregator compensation of only DKK 6000, the payback for a DKK 115.8k upgrade would stretch to nearly two decades, and a more expensive new station could take over a century.

    “Given an annual aggregator compensation of only DKK 6000, the payback for a DKK 115.8k upgrade would stretch to nearly two decades, and a more expensive new station could take over a century.”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: All EV owners participate in the aggregator scheme; partial or voluntary participation is not modeled.
    • The authors state this as an assumption: Compensation payments are sufficient to incentivize user compliance; users accept schedule shifts as long as their costs are covered.

  • May combine two claims.

    Interpretationown data

    The results support a shift in focus from consumer cost savings toward infrastructure deferral, highlighting a DSO-oriented use case for aggregator-based flexibility that is both technically feasible and economically rational.

    “Together, these results support a shift in focus from consumer cost savings toward infrastructure deferralhighlighting a DSO-oriented use case for aggregator-based flexibility that is both technically feasible and economically rational.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: The Strib, Denmark distribution network and the modeled EV population are representative of low-voltage distribution grids where the aggregator would be deployed.

  • May combine two claims.

    Premisecitation

    Vendor prices for a new 400 kVA transformer are around DKK 240k, while a 1600 kVA transformer costs about DKK 589k.

    “Vendor prices for a new 400 kVA transformer are around DKK 240k [82], while a 1600 kVA transformer costs about DKK 589k [83].”
    Quote in the paperSubstantial claim0 of 2 sources checked
    • cites [82] Bels.dk (2025, March 11). Elsewedy Electr Transformer 400kva 10,5/0,42kv dyn11. Available online: https://bels.dk/elsewedy-electr-transformer-400kva-10-5-0-42kv-dyn11-5705157303039.

      Not checked in the free run.

    • cites [83] Bels.dk (2025, March 11). Elsewedy Electr Transformer 1600kva 10,5/0,42kv dyn11. Available online: https://bels.dk/elsewedy-electr-transformer-1600kva-10-5-0-42kv-dyn11-5705157303343.

      Not checked in the free run.

27 of 27

  • The paper relies on this without support.

    The overall cost-effectiveness of aggregator solutions compared to capital-intensive upgrades remains insufficiently quantified.

    “Third, the overall cost-effectiveness of aggregator solutions compared to capital-intensive upgrades remains insufficiently quantified.”

  • The paper relies on this without support.

    Upgrading transformers is technically straightforward but expensive and may be underutilized if load growth projections do not fully materialize.

    “Upgrading transformers is technically straightforward but expensive and may be underutilized if load growth projections do not fully materialize.”

  • The paper relies on this without support.

    Current tariff-based strategies do not adequately address local transformer overload.

    “a problem that current tariff-based strategies do not adequately address.”

  • The authors state this as an assumption.

    The aggregator is assumed to be controlled by the DSO.

    “The aggregator is assumed to be controlled by the DSO, which grants it the authority to modify individual charging plans when needed.”

  • The authors state this as an assumption.

    The aggregator targets EVs with the highest laxity first, assuming these users can be shifted at minimal inconvenience.

    “The aggregator targets EVs with the highest laxity first, assuming these users can be shifted at minimal inconvenience.”

  • The authors state this as an assumption.

    User behavior is consistent and follows historical patterns; users do not override aggregator control signals.

    “User behavior is consistent and follows historical patterns.Charging schedules are based on empirical load profiles and assume users do not override aggregator control signals.”

  • The authors state this as an assumption.

    EVs are driving each day, so each EV has a period each day of not being available for aggregation.

    “EVs are driving each day.EVs will have a period each day of not being available for aggregation.”

  • The authors state this as an assumption.

    All EV owners participate in the aggregator scheme; partial or voluntary participation is not modeled.

    “All EV owners participate in the aggregator scheme.Partial or voluntary participation is not modeled, allowing a clearer assessment of theoretical upper-bound performance.”

  • The authors state this as an assumption.

    Perfect hourly baseload predictions; load predictions are fully accurate.

    “Perfect hourly baseload predictions.Small load fluctuations within an hour are neglected and averaged for the hour due to available data.Furthermore, the load predictions are used to reschedule the charging schedules and are simulated as fully accurate.”

  • The authors state this as an assumption.

    No physical grid constraints beyond transformer overloads are modeled (voltage fluctuations, phase imbalances, and power quality issues are excluded).

    “No physical grid constraints beyond transformer overloads are modeled.Voltage fluctuations, phase imbalances, and power quality issues are excluded from the current scope.”

  • The authors state this as an assumption.

    Compensation payments are sufficient to incentivize user compliance; users accept schedule shifts as long as their costs are covered.

    “Compensation payments are sufficient to incentivize user compliance.Users are assumed to accept schedule shifts as long as their costs are covered.”

  • The authors state this as an assumption.

    The simulation assumed that the recorded kilowatt-hour values of each hour equate directly to the load in kilowatts.

    “Because the dataset is limited to the hourly resolution, the simulation assumed that the recorded kilowatt-hour values of each hour equate directly to the load in kilowatts.”

  • The authors state this as an assumption.

    EV adoption was assumed to be 100%, meaning all 126 households own an EV.

    “EV adoption was assumed to be 100%, meaning all 126 households own an EV.”

  • The authors state this as an assumption.

    Overloads on the transformer occur whenever the load exceeds the nominal 400 kW capacity, assuming a power factor of 1.

    “Overloads on the transformer occur whenever the load exceeds the nominal 400 kW capacity (assuming a power factor of 1).”

  • The authors state this as an assumption.

    The baseline scenario assumes decentralized charging follows a real-time pricing (RTP) strategy.

    “This RTP strategy serves as the default assumption for decentralized charging in the baseline scenario.”

  • The paper relies on this without support.

    The Strib, Denmark distribution network and the modeled EV population are representative of low-voltage distribution grids where the aggregator would be deployed.

    “The distribution network in Strib, Denmark, used for the case study.”

  • The paper relies on this without support.

    The compensation paid to EV users is sufficient to ensure their participation in the aggregator scheme.

    “The compensation paid by the aggregator to each EV user.”
17 of 17

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Annual compensation is described as under DKK 6000 but reported as DKK 6020

    Minor

    The abstract says the annual compensation total is under DKK 6000, whereas the results report about DKK 6020.

1 of 1

Cites

Flagged first, then by citations

Research works printed but not linked: 1

  1. (2024). IEEE Standard for Smart Energy Profile Application Protocol (Standard No. IEEE 2030.5-2023). Available online: https://standards.ieee.org/ieee/2030.5/11216/.

Cited by

Sorted by citations

Related

Authors
Christensen, K., Jørgensen, B. N., & Ma, Z. G.
Affiliations
  • University of Southern Denmark (DK)
  • SDU Center for Energy Informatics, Maersk Mc-Kinney Moeller Institute, The Faculty of Engineering, University of Southern Denmark, 5230 Odense, Denmark
ORCID
Venue
Sustainability
Publisher
MDPI AG
Published
2025-04-24
Volume
17
Issue
9
Pages
3847
ISSN
2071-1050 (ISSN-L)
DOI
10.3390/su17093847
OpenAlex
W4409767019
Cited by
8
References
49
Citations by year
2026: 5 · 2025: 3
FWCI
1.81
Type
journal article
Language
en
Open access
Licence
Locations
  • Sustainability (publishedVersion, cc-by)
  • University of Southern Denmark Research Portal (University of Southern Denmark) (acceptedVersion, cc-by)
  • arXiv (Cornell University) (submittedVersion, cc-by)
  • RePEc: Research Papers in Economics (submittedVersion)
  • arXiv (Cornell University) (cc-by)
Funders
  • Danish Industry Foundation (134232-510227)
  • EUDP (134232-510227)
Topics
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Electric Vehicles and Infrastructure
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Smart Grid Energy Management
  • Physical Sciences › Engineering › Electrical and Electronic Engineering › Optimal Power Flow Distribution
Keywords
electric vehicle charging, rule-based scheduling, multi-agent simulation, distribution system operators, residential EV charging, low-voltage transformers, EV charging coordination

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-28.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-28.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review