arxiv: 2507.12703 · v4 · submitted 2025-07-17 · 📡 eess.SY · cs.SY

Joint Price and Power MPC for Peak Power Reduction at Workplace EV Charging Stations

Thibaud Cambronne , Samuel Bobick , Wente Zeng , Scott Moura This is my paper

Pith reviewed 2026-05-19 05:10 UTC · model grok-4.3

classification 📡 eess.SY cs.SY

keywords electric vehicle chargingdemand chargemodel predictive controlprice optimizationpeak power reductionworkplace chargingMonte Carlo simulation

0 comments p. Extension

The pith

A joint price-and-power model predictive control framework reduces peak demand and operator costs at workplace EV charging stations.

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

The paper develops a strategy that offers EV users a menu of prices designed to encourage selection of controllable charging service. It then applies model predictive control to schedule power delivery while accounting for the chosen service types. The goal is to lower the station's peak power draw and thereby cut the utility demand charge that forms a major part of operating costs. Monte Carlo simulations indicate the method outperforms a benchmark optimization approach and produces substantial cost reductions. This matters for commercial charging operators because demand charges can dominate electricity bills when many vehicles arrive during the same window.

Core claim

The paper proposes a model predictive control approach that jointly optimizes a menu of price options to incentivize users to select controllable charging service and the resulting power allocation, thereby reducing both demand charge and overall operator costs at workplace EV charging stations, with Monte Carlo simulations showing outperformance relative to a state-of-the-art benchmark.

What carries the argument

The joint price-and-power model predictive control optimization that designs the price menu and power trajectories together over a receding horizon.

If this is right

Station operator costs fall through lower demand charges when the price menu successfully shifts users to controllable service.
The MPC algorithm outperforms a state-of-the-art benchmark optimization strategy in Monte Carlo trials.
Peak power consumption at the charging station is lowered by coordinating price incentives with power scheduling.
Overall electricity costs decrease when both demand charge and energy charge components are addressed simultaneously.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The same joint-optimization structure could be tested on public fast-charging sites where user elasticity may differ from workplace patterns.
Integrating real-time user feedback into the price-menu generation step would allow the controller to adapt when observed behavior deviates from the elasticity model.
Adding vehicle-to-grid capabilities as an additional controllable service option in the price menu might further increase peak-reduction potential.

Load-bearing premise

Users will respond to the offered price menu by selecting controllable charging service at rates that match the modeled demand elasticity.

What would settle it

Field measurements at an operating workplace station that record actual user price selections and the resulting peak power draw, then compare those values to the simulation predictions under the same arrival patterns.

Figures

Figures reproduced from arXiv: 2507.12703 by Samuel Bobick, Scott Moura, Thibaud Cambronne, Wente Zeng.

**Figure 2.** Figure 2: Example of control actions for benchmark (left) and MPC (right) [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗

**Figure 3.** Figure 3: Example of pricing strategy for benchmark (top) and MPC (bottom) [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

read the original abstract

Demand charge, a utility fee based on an electricity customer's peak power consumption, often constitutes a significant portion of costs for commercial electric vehicle (EV) charging station operators. This paper explores control methods to reduce peak power consumption at workplace EV charging stations in a joint price and power optimization framework. We optimize a menu of price options to incentivize users to select controllable charging service. Using this framework, we propose a model predictive control approach to reduce both demand charge and overall operator costs. Through a Monte Carlo simulation, we find that our algorithm outperforms a state-of-the-art benchmark optimization strategy and can significantly reduce station operator costs.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Joint price-power MPC for workplace EV stations shows simulated cost savings over benchmarks but hinges on an uncalibrated user elasticity model.

read the letter

This paper's core idea is a model predictive control scheme that jointly optimizes charging power limits and a menu of prices at workplace EV stations to cut peak demand and demand charges. The Monte Carlo runs indicate lower operator costs than a benchmark strategy. That is the main takeaway for someone scanning the abstract and results section. The joint formulation is the actual extension here. Prior MPC work on EV charging already handles power scheduling, but folding in price incentives to steer users toward controllable sessions creates a single optimization loop that directly targets the demand charge problem. The simulation evidence supports outperformance under the modeled conditions, which is a reasonable applied step for this setting. The soft spot is the user response model. The claimed savings depend on how users select between immediate and controllable charging based on the offered prices, using some demand elasticity function. Nothing in the provided material shows calibration against real workplace EV data or sensitivity tests for different elasticity values. If actual behavior deviates, the peak reduction disappears. That assumption is load-bearing and should be probed in review, though it is a standard limitation rather than a fatal one for an initial study. This is for control and energy systems researchers focused on EV infrastructure and demand response. Readers working on similar optimization problems for commercial charging would find the joint price-power setup useful as an example. The work is concrete enough on the application and method to deserve a serious referee, even with the modeling questions. I would recommend sending it to peer review.

Referee Report

1 major / 2 minor

Summary. The paper proposes a joint price-and-power model predictive control (MPC) framework for workplace EV charging stations. A price menu is optimized to incentivize users to select controllable (flexible) charging rather than immediate charging; the MPC then allocates power subject to the resulting demand to reduce peak power and the associated demand charge. Monte Carlo simulations are reported to show that the joint MPC outperforms a state-of-the-art benchmark and yields substantial operator-cost reductions.

Significance. If the Monte Carlo results prove robust, the work offers a practical, incentive-based approach to peak shaving at commercial EV chargers that could lower demand charges and improve grid utilization. The explicit coupling of pricing decisions with real-time power control via MPC is a constructive contribution to the literature on EV charging control.

major comments (1)

[§4.2 and §5] §4.2 (User Choice Model) and §5 (Monte Carlo Experiments): The headline cost-reduction claim rests on a specific demand-elasticity / utility function that maps the offered price menu to the probability that a user selects controllable charging. No calibration against observed workplace EV data is presented, and no sensitivity study is reported that varies the elasticity parameter or choice probabilities. Because the simulated peak reduction and operator savings are direct consequences of this unvalidated mapping, the quantitative results cannot be considered reliable without such analysis.

minor comments (2)

[Abstract] Abstract: The Monte Carlo results are summarized without stating the number of trials, the arrival/demand distributions, or any error bars or statistical tests; adding these details would strengthen the abstract.
[§2–3] Notation: The distinction between the price-menu decision variables and the subsequent power-allocation variables is not always explicit in the early sections; a short table of decision variables would improve readability.

Simulated Author's Rebuttal

1 responses · 1 unresolved

We thank the referee for their constructive review and for recognizing the potential practical value of the joint price-power MPC framework. We address the major comment on the user choice model below and will revise the manuscript to incorporate additional analysis.

read point-by-point responses

Referee: [§4.2 and §5] §4.2 (User Choice Model) and §5 (Monte Carlo Experiments): The headline cost-reduction claim rests on a specific demand-elasticity / utility function that maps the offered price menu to the probability that a user selects controllable charging. No calibration against observed workplace EV data is presented, and no sensitivity study is reported that varies the elasticity parameter or choice probabilities. Because the simulated peak reduction and operator savings are direct consequences of this unvalidated mapping, the quantitative results cannot be considered reliable without such analysis.

Authors: We thank the referee for this observation. The user choice model in §4.2 employs a standard multinomial logit formulation with elasticity parameters drawn from values reported in the EV charging literature for workplace settings. While we do not possess proprietary user-level workplace EV data that would permit direct empirical calibration, the Monte Carlo study in §5 is intended to illustrate the behavior of the overall control framework under this modeling choice. To strengthen the quantitative claims, the revised manuscript will include a dedicated sensitivity study that systematically varies the elasticity parameter and the baseline choice probabilities across a broad and plausible range. We will report the resulting distributions of peak-power reduction and operator cost savings to demonstrate that the performance advantages remain consistent. revision: yes

standing simulated objections not resolved

Direct calibration of the demand-elasticity parameters against non-public, operator-specific workplace EV user data.

Circularity Check

0 steps flagged

No circularity: MPC optimization and Monte Carlo evaluation remain independent of inputs

full rationale

The paper formulates a joint price-power MPC that optimizes a price menu and power schedule to minimize demand charges plus operator costs, then evaluates the closed-loop performance via Monte Carlo simulation under an assumed demand-elasticity model for user choice. The reported cost reductions and outperformance versus benchmark are direct numerical outcomes of running the optimizer on the simulated trajectories; they are not obtained by fitting a parameter to the target metric and then relabeling it as a prediction. No self-citation chain, uniqueness theorem, or ansatz is invoked to force the central result. The derivation chain (MPC formulation → simulated trajectories → cost metric) is therefore self-contained and does not reduce to its own inputs by construction.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review prevents full enumeration; the approach implicitly relies on unstated user choice models and simulation assumptions whose independence from the result cannot be verified.

pith-pipeline@v0.9.0 · 5635 in / 1054 out tokens · 26946 ms · 2026-05-19T05:10:58.567410+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Cost/FunctionalEquation.lean washburn_uniqueness_aczel unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

min z,p E[f(z,p,m)] subject to demand-charge constraints (5)–(7), energy constraints (16)–(18)

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

20 extracted references · 20 canonical work pages · 1 internal anchor

[1]

Charging infrastructure access and operation to reduce the grid impacts of deep electric vehicle adoption,

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
[2]

Addressing challenges to electric vehicle charging in multifamily residential buildings,

D. Peterson, “Addressing challenges to electric vehicle charging in multifamily residential buildings,” Ph.D. dissertation, 2011

work page 2011
[3]

Pricing workplace charging: financial viability and fueling costs,

B. Williams and J. DeShazo, “Pricing workplace charging: financial viability and fueling costs,”Transportation Research Record, vol. 2454, no. 1, pp. 68–75, 2014

work page 2014
[4]

Identifying potential markets for behind-the-meter battery energy storage: A survey of us demand charges,

J. A. McLaren, P. J. Gagnon, and S. Mullendore, “Identifying potential markets for behind-the-meter battery energy storage: A survey of us demand charges,” National Renewable Energy Lab.(NREL), Golden, CO (United States), Tech. Rep., 2017

work page 2017
[5]

Pricing ev charging service with demand charge,

Z. J. Lee, J. Z. Pang, and S. H. Low, “Pricing ev charging service with demand charge,”Electric Power Systems Research, vol. 189, p. 106694, 2020

work page 2020
[6]

Controlled workplace charging of electric vehicles: The impact of rate schedules on transformer aging,

S. Powell, E. C. Kara, R. Sevlian, G. V . Cezar, S. Kiliccote, and R. Rajagopal, “Controlled workplace charging of electric vehicles: The impact of rate schedules on transformer aging,”Applied Energy, vol. 276, p. 115352, 10 2020

work page 2020
[7]

Estimating the Benefits of Electric Vehicle Smart Charging at Non-Residential Locations: A Data-Driven Approach

E. C. Kara, J. S. MacDonald, D. Black, M. Berges, G. Hug, and S. Kiliccote, “Estimating the benefits of electric vehicle smart charging at non-residential locations: A data-driven approach,”CoRR, vol. abs/1503.01052, 2015. [Online]. Available: http://arxiv.org/abs/ 1503.01052

work page internal anchor Pith review Pith/arXiv arXiv 2015
[8]

Learning assisted demand charge mitigation for workplace electric vehicle charging,

K. Chaudhari, P. M. Connor, J. Comden, and J. King, “Learning assisted demand charge mitigation for workplace electric vehicle charging,”IEEE Access, vol. 10, pp. 48 283–48 291, 2022

work page 2022
[9]

Ev charging scheduling under demand charge: A block model predictive control approach,

L. Yang, X. Geng, X. Guan, and L. Tong, “Ev charging scheduling under demand charge: A block model predictive control approach,” IEEE Transactions on Automation Science and Engineering, vol. 21, no. 2, pp. 2125–2138, 2024

work page 2024
[10]

Inducing human behavior to maximize operation performance at pev charging station,

T. Zeng, S. Bae, B. Travacca, and S. Moura, “Inducing human behavior to maximize operation performance at pev charging station,”IEEE Transactions on Smart Grid, vol. 12, no. 4, pp. 3353–3363, 2021

work page 2021
[11]

Learning and optimizing charging behavior at pev charging stations: Randomized pricing experiments, and joint power and price optimization,

H. Obeid, A. T. Ozturk, W. Zeng, and S. J. Moura, “Learning and optimizing charging behavior at pev charging stations: Randomized pricing experiments, and joint power and price optimization,”Applied Energy, vol. 351, p. 121862, 2023

work page 2023
[12]

Joint price and power optimization experiment for workplace charging stations,

A. T. ¨Ozt¨urk, H. Obeid, T. Zeng, W. Zeng, and S. J. Moura, “Joint price and power optimization experiment for workplace charging stations,” Sustainable Cities and Society, p. 106784, 2025. [Online]. Available: https://www.sciencedirect.com/science/article/pii/S2210670725006584

work page 2025
[13]

Regularization and variable selection via the elastic net,

H. Zou and T. Hastie, “Regularization and variable selection via the elastic net,”Journal of the Royal Statistical Society Series B: Statistical Methodology, vol. 67, no. 2, pp. 301–320, 2005

work page 2005
[14]

Xgboost: A scalable tree boosting system,

T. Chen and C. Guestrin, “Xgboost: A scalable tree boosting system,” inProceedings of the 22nd acm sigkdd international conference on knowledge discovery and data mining, 2016, pp. 785–794

work page 2016
[15]

Electric schedule business electric vehicle,

“Electric schedule business electric vehicle,” 2023. [Online]. Avail- able: https://www.pge.com/assets/rates/tariffs/CommlElecVehicle 230601-230630.xlsx

work page 2023
[16]

Quantifying the impact of building load forecasts on optimizing energy storage systems,

L. Li, Y . Ju, and Z. Wang, “Quantifying the impact of building load forecasts on optimizing energy storage systems,”Energy and Buildings, vol. 307, p. 113913, 2024

work page 2024
[17]

Dynamic load management — chargepoint,

ChargePoint Inc., “Dynamic load management — chargepoint,” https: //www.chargepoint.com/en-gb/businesses/dynamic-load-management, 2025, accessed: 2025-09-17

work page 2025
[18]

Evgo optima,

EVgo Services LLC, “Evgo optima,” https://www.evgo.com/optima/, 2025, accessed: 2025-09-17

work page 2025
[19]

Adaptive charging network for electric vehicles

G. Lee, T. Lee, Z. Low, S. H. Low, and C. Ortega, “Adaptive charging network for electric vehicles.” IEEE, 12 2016, pp. 891–895

work page 2016
[20]

Attention is all you need,

A. Vaswani, “Attention is all you need,”Advances in Neural Informa- tion Processing Systems, 2017

work page 2017