arxiv: 2605.14282 · v1 · submitted 2026-05-14 · 📡 eess.SY · cs.SY

Recognition: no theorem link

Energy Management for Solar-Powered Electric-Bus Charging Station: A Data-Driven Method

Xiaoting Wang , Supun Amarathunga , Pasan Gunawardena , Gregory Kish , Yunwei (Ryan) Li

Authors on Pith no claims yet

Pith reviewed 2026-05-15 02:44 UTC · model grok-4.3

classification 📡 eess.SY cs.SY

keywords energy management systemelectric bus charging stationsolar PVpolynomial chaos expansiondata-driven surrogateuncertainty modelingrenewable integration

0 comments

The pith

A data-driven energy management system for solar-powered electric bus charging stations handles uncertainties in solar output, prices, and bus states using limited data.

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

This paper develops a flexible energy management system for electric bus charging stations that integrate solar generation, energy storage, and vehicle charging. It accounts for uncertainties in solar PV output, electricity prices, and the arrival and departure states of charge of the buses. The approach builds a polynomial chaos expansion surrogate from a limited set of uncertainty samples and applies a nonparametric inference method to enrich the data when historical records are sparse. Case studies on a station serving 20 electric buses show that the method improves decision-making for renewable integration and operational planning. If the surrogate accurately captures the joint distributions, the system can reduce costs and maintain reliability without requiring extensive real-world data collection.

Core claim

The paper claims that a data-driven polynomial chaos expansion surrogate, constructed from limited uncertainty samples and augmented by nonparametric inference, enables an effective energy management system for a solar-powered electric bus charging station that integrates renewable generation, storage, and charging while accounting for uncertainties in solar PV output, electricity prices, and bus arrival/departure states of charge, as verified through case studies on a station with 20 electric buses.

What carries the argument

The data-driven polynomial chaos expansion surrogate model, built from limited samples and enriched via nonparametric inference, that represents the joint distribution of solar PV output, electricity prices, and electric bus states of charge to support optimization decisions.

If this is right

The system can schedule charging and storage use to lower electricity purchase costs while respecting bus availability constraints.
Renewable generation and battery storage integration becomes feasible even with sparse historical data on uncertainties.
Operational decisions remain reliable when solar output, market prices, and bus states vary within the modeled distributions.
The method scales to stations serving dozens of buses without needing large new datasets for each deployment.

Where Pith is reading between the lines

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

The same surrogate technique could extend to other renewable-integrated fleets such as delivery vans or taxis with similar uncertainty profiles.
Real-time implementation might reduce the need for frequent retraining when new weather or price data arrives.
Combining this approach with simple rule-based fallbacks could improve robustness if the surrogate encounters out-of-sample conditions.

Load-bearing premise

The polynomial chaos expansion surrogate built from limited uncertainty samples, together with the nonparametric inference method, accurately represents the joint distribution of solar PV output, electricity prices, and bus arrival and departure states of charge.

What would settle it

Running the proposed EMS on a real solar-powered electric bus charging station with 20 buses and finding that actual daily costs or constraint violations under varying solar output and prices differ substantially from the case study predictions.

Figures

Figures reproduced from arXiv: 2605.14282 by Gregory Kish, Pasan Gunawardena, Supun Amarathunga, Xiaoting Wang, Yunwei (Ryan) Li.

**Figure 2.** Figure 2: EB load charging state for day 1 [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗

**Figure 3.** Figure 3: The histogram and statistics of minimum cost [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗

**Figure 4.** Figure 4: The comparison of histogram of minimum cost [PITH_FULL_IMAGE:figures/full_fig_p005_4.png] view at source ↗

read the original abstract

This paper presents a flexible energy management system (EMS) for an electric bus charging station (EBCS) that integrates renewable generation, energy storage, and electric bus (EB) charging while accounting for uncertainties in solar PV output, electricity prices, and EB arrival/departure state of charge. A data-driven polynomial chaos expansion surrogate is developed from a limited set of uncertainty samples, and a nonparametric inference method is used to enrich the input data when historical data is limited. Case studies on a solar-powered EBCS with 20 EBs demonstrate the effectiveness of the proposed EMS and data-driven method.

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

This paper applies polynomial chaos expansion surrogates plus nonparametric data enrichment to build a practical EMS for solar-powered electric bus charging, with case studies on 20 buses as the main evidence.

read the letter

The core takeaway is that the authors combine a polynomial chaos expansion surrogate built from limited uncertainty samples with nonparametric inference to model joint distributions of solar PV output, electricity prices, and electric bus arrival/departure states of charge, then use that to run an energy management optimization for a charging station. Case studies on a 20-bus system are presented to show the approach works in practice. That is the main deliverable. The work does a solid job of framing a concrete engineering problem and walking through a data-driven workflow that addresses limited historical data, which is common in these installations. The case-study setup gives readers a tangible example of how the surrogate feeds into the EMS decisions, and the focus on joint uncertainties rather than isolated ones is a reasonable choice for this domain. It is a usable extension of existing surrogate techniques to renewable-powered transport charging. The soft spots are mostly about scope and validation depth rather than outright flaws. Both polynomial chaos expansion and nonparametric enrichment are established methods, so the paper is an application rather than a new framework; the advance sits in the specific combination and the EBCS case. The abstract claims the case studies demonstrate effectiveness, but without visible error metrics, surrogate accuracy plots, or statistical significance tests in the provided summary, it is difficult to judge how closely the model matches the true uncertainty distributions or whether the reported gains are robust. If the full paper includes those checks and clear baselines, the results hold up better. This is aimed at engineers working on EV charging infrastructure and renewable integration who need a workable tool for uncertainty handling with sparse data. A reader interested in applied surrogate modeling for energy systems would get value from the workflow and the 20-bus example. It deserves a serious referee because the problem is timely, the method is reproducible enough to evaluate, and the case-study evidence is concrete even if incremental.

Referee Report

0 major / 2 minor

Summary. The manuscript proposes a flexible energy management system (EMS) for a solar-powered electric bus charging station (EBCS) that integrates renewable generation, energy storage, and EB charging. It accounts for uncertainties in solar PV output, electricity prices, and EB arrival/departure states of charge via a data-driven polynomial chaos expansion (PCE) surrogate constructed from limited uncertainty samples, augmented by a nonparametric inference method to enrich data when historical records are scarce. Effectiveness is demonstrated through case studies on a system with 20 EBs.

Significance. If the surrogate accurately captures the joint uncertainty distributions and the resulting EMS yields measurable operational gains, the work offers a practical data-driven framework for stochastic optimization in renewable-powered transport infrastructure. The focus on limited-data scenarios via PCE and nonparametric enrichment is relevant for real-world EBCS deployments where full historical datasets are often unavailable, potentially supporting cost-effective integration of solar resources into public transit charging.

minor comments (2)

The abstract and method overview state that the PCE surrogate and nonparametric enrichment accurately represent the joint distribution of uncertainties, but the manuscript should include explicit validation metrics (e.g., cross-validation error or Kolmogorov-Smirnov statistics) comparing the surrogate to held-out samples to confirm fidelity.
In the case-study section, the claim that the proposed EMS demonstrates effectiveness would be strengthened by reporting quantitative improvements (e.g., percentage reduction in expected cost or peak demand) relative to a deterministic baseline, together with the number of Monte Carlo realizations used for evaluation.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for the constructive review and the recommendation of minor revision. The positive assessment of our data-driven PCE surrogate with nonparametric enrichment for handling uncertainties in solar-powered EBCS energy management is appreciated. As no specific major comments were raised, we will implement minor editorial improvements and clarifications in the revised manuscript to enhance readability and completeness.

Circularity Check

0 steps flagged

No significant circularity in the derivation chain

full rationale

The paper describes a standard data-driven workflow: building a polynomial chaos expansion surrogate from limited uncertainty samples, applying nonparametric inference to enrich data when historical records are sparse, and validating via case studies on a 20-EB solar-powered charging station. No equations, fitting procedures, or self-citations are visible that reduce any claimed prediction or performance metric to an input parameter by construction. The central claim rests on external case-study demonstration rather than self-definitional loops, fitted-input renamings, or load-bearing self-citations. This matches the expected non-circular pattern for surrogate-based optimization papers.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review prevents identification of concrete free parameters, axioms, or invented entities; none are stated explicitly in the provided text.

pith-pipeline@v0.9.0 · 5409 in / 1077 out tokens · 24624 ms · 2026-05-15T02:44:37.354745+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

18 extracted references · 18 canonical work pages

[1]

Optimal control of s olar- powered electric bus networks with improved renewable ener gy on-site consumption and reduced grid dependence,

H. Ren, Z. Ma, C. F. N. Tse, and Y . Sun, “Optimal control of s olar- powered electric bus networks with improved renewable ener gy on-site consumption and reduced grid dependence,” Applied Energy , vol. 323, p. 119643, 2022

work page 2022
[2]

Optimal ev c harging scheduling at electric railway stations under peak load con straints,

G. Pierrou, C. V alero-De la Flor, and G. Hug, “Optimal ev c harging scheduling at electric railway stations under peak load con straints,” Electric Power Systems Research , vol. 235, p. 110612, 2024

work page 2024
[3]

A novel and cost- efﬁcient energy management system for plug-in electric bus charging depot owners,

S. M. Arif, T. T. Lie, B. C. Seet, and S. Ayyadi, “A novel and cost- efﬁcient energy management system for plug-in electric bus charging depot owners,” Electric Power Systems Research , vol. 199, p. 107413, 2021

work page 2021
[4]

Electric bus fast ch arging station resource planning considering load aggregation an d renewable integration,

Y . Cheng, W. Wang, Z. Ding, and Z. He, “Electric bus fast ch arging station resource planning considering load aggregation an d renewable integration,” IET Renewable Power Generation, vol. 13, no. 7, pp. 1132– 1141, 2019

work page 2019
[5]

Elect ric bus ﬂeet charging management: A robust optimisation framework addressing battery ageing, time-of-use tariffs, and energy consumpti on uncertainty,

J. A. Manzolli, J. P . F. Trov˜ ao, and C. H. Antunes, “Elect ric bus ﬂeet charging management: A robust optimisation framework addressing battery ageing, time-of-use tariffs, and energy consumpti on uncertainty,” Applied Energy , vol. 381, p. 125137, 2025

work page 2025
[6]

Stochastic energy management of electric bus charging stations with renewable energy integration an d b2g capa- bilities,

P . Zhuang and H. Liang, “Stochastic energy management of electric bus charging stations with renewable energy integration an d b2g capa- bilities,” IEEE Transactions on Sustainable Energy , vol. 12, no. 2, pp. 1206–1216, 2021

work page 2021
[7]

Quasi-mon te carlo based probabilistic small signal stability analysis for po wer systems with plug-in electric vehicle and wind power integration,

W. C. Wong, C. Y . Chung, K. W. Chan, and H. Chen, “Quasi-mon te carlo based probabilistic small signal stability analysis for po wer systems with plug-in electric vehicle and wind power integration,” IEEE Transactions on Power Systems , vol. 28, no. 3, pp. 3335–3343, 2013

work page 2013
[8]

Data-driven l ow-rank tensor approximation for fast grid integration of commerci al ev charging stations considering demand uncertainties,

Y . Jiang, T. H. Ortmeyer, M. Fan, and X. Ai, “Data-driven l ow-rank tensor approximation for fast grid integration of commerci al ev charging stations considering demand uncertainties,” IEEE Transactions on Smart Grid, vol. 14, no. 1, pp. 517–529, 2023

work page 2023
[9]

bcharge: D ata-driven real-time charging scheduling for large-scale electric bu s ﬂeets,

G. Wang, X. Xie, F. Zhang, Y . Liu, and D. Zhang, “bcharge: D ata-driven real-time charging scheduling for large-scale electric bu s ﬂeets,” in 2018 IEEE Real-Time Systems Symposium (RTSS) , 2018, pp. 45–55

work page 2018
[10]

Data-driven estimation o f energy consumption for electric bus under real-world driving cond itions,

Y . Chen, Y . Zhang, and R. Sun, “Data-driven estimation o f energy consumption for electric bus under real-world driving cond itions,” Transportation Research Part D: Transport and Environment , vol. 98, p. 102969, 2021

work page 2021
[11]

A data-driven spa rse polynomial chaos expansion method to assess probabilistic total transfer capability for power systems with renewables,

X. Wang, X. Wang, H. Sheng, and X. Lin, “A data-driven spa rse polynomial chaos expansion method to assess probabilistic total transfer capability for power systems with renewables,” IEEE Transactions on Power Systems, vol. 36, no. 3, pp. 2573–2583, 2021

work page 2021
[12]

A data- driven nonparametric approach for probabilistic load-mar gin assessment considering wind power penetration,

Y . Xu, L. Mili, M. Korkali, K. Karra, Z. Zheng, and X. Chen , “A data- driven nonparametric approach for probabilistic load-mar gin assessment considering wind power penetration,” IEEE Transactions on Power Systems, vol. 35, no. 6, pp. 4756–4768, 2020

work page 2020
[13]

A ﬂe xible energy management system for solar powered electric-bus ch arging stations,

S. Amarathunga, P . Gunawardena, X. Wang, and Y . Li, “A ﬂe xible energy management system for solar powered electric-bus ch arging stations,” in 2025 IEEE Applied Power Electronics Conference and Exposition (APEC) , 2025, pp. 3030–3035

work page 2025
[14]

Chaos polynomial creux et adaptatif pour l a propagation d’incertitudes et l’analyse de sensibilit´ e,

G. Blatman, “Chaos polynomial creux et adaptatif pour l a propagation d’incertitudes et l’analyse de sensibilit´ e,” Ph.D. disse rtation, Universit´ e Blaise Pascal-Clermont-Ferrand II, 2009

work page 2009
[15]

UQLab user manual – polynomia l chaos expansions,

S. Marelli and B. Sudret, “UQLab user manual – polynomia l chaos expansions,” Chair of Risk, Safety and Uncertainty Quantiﬁ cation, ETH Zurich, Tech. Rep. UQLab-V1.1-104, 2018

work page 2018
[16]

B. W. Silverman, Density estimation for statistics and data analysis . Routledge, 2018

work page 2018
[17]

Energy Trading Sys tem (ETS),

Alberta Electric System Operator, “Energy Trading Sys tem (ETS),” http://ets.aeso.ca/, [Online]. Accessed: Aug. 15, 2025

work page 2025
[18]

NSRDB Data Viewer ,

National Solar Radiation Database, “NSRDB Data Viewer ,” https://nsrdb.nrel.gov/data-viewer, [Online]. Accesse d: Aug. 15, 2025

work page 2025