Pith. sign in

REVIEW 3 major objections 4 minor 20 references

Impact Analysis of Optimal EV Bi-directional Charging with Spatial-temporal Constraints

T0 review · 3 major / 4 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A fleet-scale EV charge scheduler can meet severe zone power limits with only minimal increase in total charging cost, and V2G arbitrage can turn charging into a net revenue stream.

desk verdict Useful spatial-temporal EV charging case study whose 'minimal impact' cost result is robust for uni-directional charging but whose V2G negative costs are arithmetic artifacts of lossless, price-symmetric storage. read the letter →

arxiv 2507.12877 v1 pith:XX5NJ27T submitted 2025-07-17 math.OC

Xian-Long Lee , Adel N. Toosi , Peter Pudney , Ian McLeod , Muhammad Aamir Cheema , Hao Wang This is my paper · ORCID
classification math.OC MSC 90C11
keywords electricvehiclechargingvehicle-to-gridchargeschedulingmixed-integerlinearprogrammingtime-of-usepricingdistributionnetworkcapacityspatialloadshiftingdemandresponse
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Smart, centrally optimized charging can let a fleet of electric vehicles absorb severe local power capacity constraints at almost no extra aggregate cost, because the scheduler shifts charging across time and across zones. The paper builds a mixed-integer linear program that minimizes total charging cost for EVs moving between zones with different prices and power limits, and tests it on real Melbourne demand and price data with simulated driving plans. Under extreme constraints, total charging cost rises only slightly, from 850 to 858 AUD for uni-directional charging and from -5,774 to -5,164 AUD for V2G in the all-zone case. With bidirectional charging, the fleet can earn net revenue through price arbitrage, though this depends on the paper's assumption of lossless batteries and equal charge and discharge prices.

What carries the argument

The load-bearing object is the mixed-integer linear program that minimizes $\sum_{i,z,t} b_{i,z,t}\lambda(z,t,p_{i,t}) p_{i,t}\Delta t$, subject to per-EV power limits, battery energy dynamics $e_{i,t}=e_{i,t-1}+p_{i,t}\Delta t-d_{i,t}$, battery capacity, final energy targets, and the zone capacity constraint $l_{z,t}+\sum_i p_{i,t} b_{i,z,t}\le c^+_{z,t}$. The zone capacity constraint couples every EV's charging decision to local non-EV demand at each half-hour, and the availability indicator $b_{i,z,t}$ encodes where each EV is parked. This constraint is what makes the problem spatial-temporal and what the paper tightens to test the impact of extreme capacity limits.

What would settle it

Re-run the same model with a round-trip efficiency below 100% (for example, 90%) and with discharge priced below charge; if total V2G cost becomes positive or the cost increase under 0% capacity headroom exceeds the few-AUD range reported in Table III, the central conclusion depends on the lossless-battery assumption. A second check would use real instead of simulated EV driving plans and compare whether the minimal-impact result survives realistic parking availability.

Watch

Extended reading notes

Core claim

The paper's central claim is that when EV charging is scheduled to minimize total fleet cost, even extreme zone-level power capacity constraints (down to zero headroom over existing peak demand) have only a minor effect on overall charging cost. Charging demand shifts to less constrained zones and to cheaper times, so the cost penalty stays small: in the uni-directional case total cost moves from 850 to 858 AUD, and in the V2G case total cost moves from -5,774 to -5,164 AUD while remaining negative. The paper also finds that price profiles shape where and when charging happens, that demand-aligned pricing reduces grid impact, and that V2G charging can produce net revenue through discharging at high-price times and charging at low-price times. Four metrics (peak-demand increase per zone, share of EV energy per zone, total charging cost, and discharged/charged ratio) are used to compare scenarios from both zone and user perspectives.

Load-bearing premise

The paper assumes EV batteries charge and discharge with 100% round-trip efficiency and that discharging receives the same price as charging; if real efficiency losses or price spreads are added, the V2G revenue and the small-cost-impact conclusion could weaken.

Editorial extensions

If this is right

  • Distribution planners could impose tight zone capacity limits without large aggregate cost penalties, provided EV charging is centrally optimized.
  • Real-time or demand-aligned price profiles steer charging to low-price, low-demand periods and zones, reducing peak stress compared with retail time-of-use tariffs.
  • Bidirectional charging can make a fleet's total charging cost negative, meaning the fleet earns revenue through arbitrage, under the paper's efficiency and price assumptions.
  • Constraining only the high-demand CBD zone shifts charging demand to the Suburb zone with little cost increase, showing the spatial flexibility of the fleet.
  • The scheduling method gives aggregate cost savings but not necessarily fair per-EV outcomes, a fairness gap the paper itself notes.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural extension is to price battery degradation into the objective; even a small per-kWh discharge cost could erase the reported V2G revenue, since that revenue comes from round-trip arbitrage with lossless batteries.
  • The minimal-cost solution likely overstates real-world flexibility because it assumes perfect foresight of prices and driving plans; under uncertainty, the same constraints could raise costs more.
  • The fairness caveat suggests that allocating the optimized schedule's cost among owners is itself an optimization problem; the aggregate result does not reveal which owners benefit or lose.
  • The single-number-per-zone capacity abstraction could behave differently if applied to a real feeder hierarchy with distinct capacity limits at multiple levels.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper formulates a mixed-integer linear program for centrally optimizing the charging/discharging schedules of an EV fleet across spatial zones and time intervals, subject to per-zone power capacity constraints, EV battery limits, driving energy consumption, and final state-of-charge targets. The objective is total fleet charging cost, with a binary availability parameter restricting each EV to connected zones. The authors evaluate uni-directional and bi-directional (V2G) charging under three price profiles using data inspired by Melbourne zones: real-time, normalized-demand, and retail ToU tariffs, with simulated EV driving plans. Four metrics track zone peak ratios, zone energy shares, total cost, and arbitrage activity. The main reported finding is that tightening zone capacity constraints to extreme levels (down to 0% headroom) produces only a minor increase in total charging cost, while V2G scenarios yield deeply negative costs, interpreted as net revenue.

Significance. If the central result is robust, the paper would provide a practically relevant demonstration that centrally coordinated EV charging can absorb tight distribution-level capacity constraints at small aggregate cost, and that V2G arbitrage can convert price differentials into fleet revenue. The optimization model itself is clearly stated and the paper is transparent about its assumptions, including the use of real price and demand data and simulated driving plans. The formulation is, however, a standard MILP rather than a new algorithmic contribution, and the numerical conclusions are not backed by released code or data. The main value lies in the scenario analysis and the proposed metrics. The V2G conclusions are not robust as presented because the model assumes 100% round-trip battery efficiency and equal charge/discharge prices, which turns price volatility into risk-free arbitrage and directly drives the negative costs in Tables II and III.

major comments (3)
  1. [Section II-B, Eq. (5); Section III-A] The model assumes 100% round-trip battery efficiency, stated after Eq. (5), and that the discharge price matches the charge price, stated in Section III-A. Under these assumptions, every time-of-use price differential becomes a risk-free arbitrage opportunity bounded only by battery capacity and connection time. This is precisely what produces the deeply negative total costs in Table III (e.g., -5,774 AUD at baseline) and what makes the capacity-constraint impact appear minimal in the V2G cases. These assumptions are load-bearing for the bi-directional part of the headline result, not peripheral simplifications. Please add a round-trip efficiency factor and an asymmetric discharge price, or at minimum a sensitivity analysis over realistic efficiency (e.g., 85-95%) and export-price discounts (e.g., 50-80% of import price), and restate the cost-impact conclusions in light of those results.
  2. [Section II-B, Eq. (8)] The final target-energy constraint uses an undefined variable v_{i,t}; the driving energy consumption is defined as d_{i,t} in Eq. (5). As written, Eq. (8) is not fully specified and the model cannot be reproduced without guessing the intended symbol. This should be corrected to d_{i,t}, or v_{i,t} should be defined, and the constraint should be re-verified so that the final state-of-charge condition is unambiguous.
  3. [Section IV-A, Table I] The EV driving plans are simulated using randomized schedules for three user types, and the destination distribution percentages in Table I are assumed rather than estimated from travel-survey or mobility data. Since the quantitative impact metrics and the main conclusion that capacity constraints only minimally affect cost depend on where and when EVs are parked, the paper should either validate the driving-plan assumptions against real travel data or explicitly restrict the conclusions to the synthetic scenarios. A sensitivity analysis over destination distributions and commute-time ranges would clarify how robust the reported cost impact is.
minor comments (4)
  1. [Section III-B] The list of four metrics includes 'proportion of charging energy used for driving,' but the results sections report the discharged/charged ratio instead. Please clarify how these metrics are related and report both consistently.
  2. [Tables II and III] In V2G scenarios the 'total cost' is negative and is effectively net revenue or profit. Consider labeling this column as 'net profit' when V2G is enabled, or at least adding a footnote in the tables so that readers do not interpret negative costs as a computational artifact.
  3. [Section III-A and Figure 2] The zone demand profiles are shown as normalized values, but the capacity constraint c+z,t = (1+eta) lmax_z uses the peak demand lmax_z. It is unclear whether the optimization uses the normalized or the actual megawatt-scale profiles. Please state the actual scale and how the normalization is undone before applying the constraint.
  4. [Section II-B, Eq. (7)] The constraint (7) is written with a time-varying capacity c+z,t, but Section III-A sets a single constant capacity for each zone. The notation suggests time-dependence; please clarify whether c+z,t is constant over t in all experiments or whether time-varying capacities are considered.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the reported costs are optimal objective values of an explicitly stated MILP, and the central claim is a comparative scenario result rather than a fitted prediction.

full rationale

The paper's central claim that overall EV charging costs are only minimally affected under extreme power capacity constraints is a comparative result obtained by solving an explicitly stated mixed-integer linear program (Eqs. 1-8) with real demand/price data and simulated driving plans. The total charging cost is indeed the objective function being minimized, so it is definitionally the minimum cost for each scenario; however, the paper's contribution is not to infer costs from fitted parameters but to compare how the optimal value changes when zone capacity constraints are tightened. No parameter is calibrated to reproduce the headline result, no prediction is derived from a quantity that was fitted to that prediction, and the model does not invoke any self-citation as load-bearing support. The assumptions of 100% round-trip battery efficiency and equal charge/discharge prices are explicit modeling simplifications (stated after Eq. 5 and in Section III-A) that substantially affect the magnitude of the V2G cost results, but they are transparent assumptions about the system being modeled, not circular reasoning. There is no imported uniqueness theorem, no ansatz disguised as citation, and no renaming of a known result as a new one. The derivation chain is therefore self-contained: inputs are stated, the optimization problem is defined, and the reported metrics are computed from the resulting optimal schedules without any reduction to the inputs by construction.

Assumptions & free parameters 6 free parameters · 6 assumptions · 0 invented entities

The central claim rests on a standard LP/MILP model plus hand-set scenario parameters and domain assumptions. The most important assumptions are lossless battery cycling, discharge price equal to charge price, and an unvalidated synthetic driving-plan generator; these are not fitted but are chosen, and they directly shape the quantitative cost and arbitrage results.

free parameters (6)
  • Zone power constraint factor eta = 0, 0.3, 0.6
    Scenario parameter in Section III-A defining c_z,t = (1+eta) lmax_z; not fitted, but directly shapes the constrained-scenario results.
  • Battery initial, target, and capacity = 24 kWh initial, 48 kWh target, 60 kWh capacity
    Hand-set in Section IV-A; drives feasibility and the amount of arbitrage energy available.
  • Charger power limit p+ = 7.4 kW
    Hand-set in Section IV-A; limits how much load can shift between zones and time intervals.
  • Destination distribution percentages = Table I: e.g., CBD residential 10%, work 70%, other 30%
    Hand-assigned in Section IV-A; determines the spatial pattern of charging demand.
  • Travel energy consumption between zones = Figure 4 values: e.g., 1.0, 1.4, 2.4, 2.6 kWh
    Hand-specified based on 'real distances'; affects feasible charging energy and energy transferred between zones.
  • Commute and parking time ranges = e.g., day-worker commute 0.5 to 1.5 hours, work 7 to 9 hours
    Randomly sampled in the driving-plan simulation; no dataset or seed is provided, so results depend on this unobservable choice.
assumptions (6)
  • standard math Gurobi returns optimal solutions to the MILP.
    Section IV-B states Gurobi 11 solves the problem; all reported results assume solver correctness.
  • domain assumption EV charge and discharge transfer energy with no efficiency loss.
    Section II-B after Eq. (5): 'the efficiency factor is not within the scope of this research'. This inflates V2G arbitrage value.
  • domain assumption Discharging price equals charging price in bi-directional scenarios.
    Section III-A states discharging offsets other energy use, so the same price is assumed; this makes arbitrage profitable and produces negative total costs.
  • ad hoc to paper Simulated EV driving plans and destination distributions represent real EV usage.
    Section IV-A describes random pairing and sampled parking and commute times, but no validation data or seed is provided.
  • domain assumption A single aggregate capacity per zone, set as a multiple of peak demand, captures distribution network constraints.
    Section III-A; real feeder constraints and power flows are more complex and spatially heterogeneous than one per-zone cap.
  • domain assumption All EVs participate in a centralized optimizer with perfect foresight of future prices and demand.
    Section II-B; the minimization is over the whole fleet using full-horizon price and demand data, with no informational, privacy, or control-communication constraints.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Impact Analysis of Optimal EV Bi-directional Charging with Spatial-temporal Constraints." pith.science (2026). https://pith.science/paper/XX5NJ27T

@misc{pith2026250712877,
  author       = {Pith},
  title        = {Pith review of: Impact Analysis of Optimal EV Bi-directional Charging with Spatial-temporal Constraints},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/XX5NJ27T}},
  note         = {Machine review of arXiv:2507.12877}
}
read the original abstract

The growth in Electric Vehicle (EV) market share is expected to increase power demand on distribution networks. Uncoordinated residential EV charging, based on driving routines, creates peak demand at various zone substations depending on location and time. Leveraging smart charge scheduling and Vehicle-to-Grid (V2G) technologies offers opportunities to adjust charge schedules, allowing for load shifting and grid support, which can reduce both charging costs and grid stress. In this work, we develop a charge scheduling optimization method that can be used to assess the impact of spatial power capacity constraints and real-time price profiles. We formulate a mixed-integer linear programming problem to minimize overall charging costs, taking into account factors such as time-varying EV locations, EV charging requirements, and local power demands across different zones. Our analysis uses real data for pricing signals and local power demands, combined with simulated data for EV driving plans. Four metrics are introduced to assess impacts from the perspectives of both EV users and zones. Results indicate that overall EV charging costs are only minimally affected under extreme power capacity constraints.

Figures

Figures reproduced from arXiv: 2507.12877 by the authors.

Figure 1
Figure 1. illustrates EVs traveling throughout a metropoli￾tan area based on their daily driving routines. Residential EVs that drive to different locations, such as workplaces, supermarkets, and homes, are placed across multiple “zones” within the metropolitan area. A “zone” represents an abstract area, which can be defined by geographic location or spatial grouping based on specific interests. These zones may contain Zone Z… view at source ↗
Figure 2
Figure 2. Local energy demand for the three example zones [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Example price profiles B. Analysis Metrics We report four metrics to evaluate the results of each scenario: • increase in peak demand in each zone • proportion of EV energy use in each zone • total charging cost across all EVs • proportion of charging energy used for driving. The increase in peak demand in zone z is denoted µz, and is given by the ratio of the maximum demand with EVs to the maximum original demand i… view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: The simulated energy consumption (kWh) for traveling within and [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]
Figure 6
Figure 6. Figure 6: Distribution results for different price profiles [PITH_FULL_IMAGE:figures/full_fig_p004_6.png]
Figure 7
Figure 7. Figure 7: Distribution results with zone power constraints (All-zone). (d): Uni [PITH_FULL_IMAGE:figures/full_fig_p005_7.png]
Figure 8
Figure 8. Figure 8: Distribution results with zone power constraints. (g),(h): Cost [PITH_FULL_IMAGE:figures/full_fig_p005_8.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

20 extracted references · 11 canonical work pages

  1. [1]

    IEA , ``Global EV outlook 2024,'' https://www.iea.org/reports/global-ev-outlook-2024, 2024

    work page 2024

  2. [2]

    Muratori, ``Impact of uncoordinated plug-in electric vehicle charging on residential power demand,'' Nature Energy, vol

    M. Muratori, ``Impact of uncoordinated plug-in electric vehicle charging on residential power demand,'' Nature Energy, vol. 3, no. 3, pp. 193--201, 2018

    work page 2018

  3. [3]

    J. Y. Yong, W. S. Tan, M. Khorasany, and R. Razzaghi, ``Electric vehicles destination charging: An overview of charging tariffs, business models and coordination strategies,'' Renewable and Sustainable Energy Reviews, vol. 184, p. 113534, 2023

    work page 2023

  4. [4]

    C. B. Jones, W. Vining, M. Lave, T. Haines, C. Neuman, J. Bennett, and D. R. Scoffield, ``Impact of electric vehicle customer response to time-of-use rates on distribution power grids,'' Energy Reports, vol. 8, pp. 8225--8235, 2022

    work page 2022

  5. [5]

    M. A. Sohail, R. Khan, S. H. Mukhtar, and A. Usman, ``Impact analysis of time-of-use pricing enabled electric vehicle charging to the uncoordinated charging on a distribution network,'' in 2023 IEEE Power & Energy Society General Meeting (PESGM). 1em plus 0.5em minus 0.4em IEEE, 2023, pp. 1--5

    work page 2023

  6. [6]

    Powell, G

    S. Powell, G. V. Cezar, L. Min, I. M. Azevedo, and R. Rajagopal, ``Charging infrastructure access and operation to reduce the grid impacts of deep electric vehicle adoption,'' Nature Energy, vol. 7, no. 10, pp. 932--945, 2022

    work page 2022

  7. [7]

    Flataker, O

    A. Flataker, O. K. Malmin, O. A. Hjelkrem, R. Rana, M. Korp s, and B. N. Tors ter, ``Impact of home-and destination charging on the geographical and temporal distribution of electric vehicle charging load,'' in 2022 18th International Conference on the European Energy Market (EEM). 1em plus 0.5em minus 0.4em IEEE, 2022, pp. 1--6

    work page 2022

  8. [8]

    Rahman, I

    S. Rahman, I. A. Khan, and M. H. Amini, ``A review on impact analysis of electric vehicle charging on power distribution systems,'' in 2020 2nd International Conference on Smart Power & Internet Energy Systems (SPIES). 1em plus 0.5em minus 0.4em IEEE, 2020, pp. 420--425

    work page 2020

Show all 20 references
  1. [9]

    Rahman, I

    S. Rahman, I. A. Khan, A. A. Khan, A. Mallik, and M. F. Nadeem, ``Comprehensive review & impact analysis of integrating projected electric vehicle charging load to the existing low voltage distribution system,'' Renewable and Sustainable Energy Reviews, vol. 153, p. 111756, 2022

    2022

  2. [10]

    Powercor, ``Distribution annual planning report and data,'' Data available at https://www.powercor.com.au/network-planning-and-projects/network-data/

  3. [11]

    Eason, B

    G. Eason, B. Noble, and I. N. Sneddon, ``On certain integrals of Lipschitz-Hankel type involving products of Bessel functions,'' Phil. Trans. Roy. Soc. London, vol. A247, pp. 529--551, April 1955

    1955

  4. [12]

    Clerk Maxwell, A Treatise on Electricity and Magnetism, 3rd ed., vol

    J. Clerk Maxwell, A Treatise on Electricity and Magnetism, 3rd ed., vol. 2. Oxford: Clarendon, 1892, pp.68--73

  5. [13]

    I. S. Jacobs and C. P. Bean, ``Fine particles, thin films and exchange anisotropy,'' in Magnetism, vol. III, G. T. Rado and H. Suhl, Eds. New York: Academic, 1963, pp. 271--350

    1963

  6. [14]

    Elissa, ``Title of paper if known,'' unpublished

    K. Elissa, ``Title of paper if known,'' unpublished

  7. [15]

    Nicole, ``Title of paper with only first word capitalized,'' J

    R. Nicole, ``Title of paper with only first word capitalized,'' J. Name Stand. Abbrev., in press

  8. [16]

    Yorozu, M

    Y. Yorozu, M. Hirano, K. Oka, and Y. Tagawa, ``Electron spectroscopy studies on magneto-optical media and plastic substrate interface,'' IEEE Transl. J. Magn. Japan, vol. 2, pp. 740--741, August 1987 [Digests 9th Annual Conf. Magnetics Japan, p. 301, 1982]

    1987

  9. [17]

    Young, The Technical Writer's Handbook

    M. Young, The Technical Writer's Handbook. Mill Valley, CA: University Science, 1989

    1989

  10. [18]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

  11. [19]

    " 'urlprefix :=

    11em plus .33em minus .07em 4000 4000 100 4000 4000 500 `\.=1000 = #1 \@IEEEnotcompsoconly \@IEEEcompsoconly #1 * [1] 0pt [0pt][0pt] #1 * [1] 0pt [0pt][0pt] #1 * \| ** #1 \@IEEEauthorblockNstyle \@IEEEcompsocnotconfonly \@IEEEauthorblockAstyle \@IEEEcompsocnotconfonly \@IEEEco...

  12. [20]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.