Electric Vehicle Charging Infrastructure: Distribution System Impact Assessment

Published: June 2026 Technical Level: Advanced Category: Power Systems Design

Abstract

The accelerating deployment of electric vehicle charging infrastructure at commercial, multifamily, and light industrial facilities creates distribution system impacts that require systematic engineering assessment before installation to avoid transformer overloading, voltage violations, harmonic distortion, and protection coordination failures. EV charging loads differ from conventional commercial loads in several important respects: their power demand is concentrated in time (evening charging peaks when residential loads are already near maximum), their power factor characteristics depend on the charger type (Level 2 EVSE is near unity power factor; DC fast chargers are nonlinear loads with significant harmonic content), and their peak demand magnitude can reach 50 to 150 percent of a commercial building's existing peak demand for large multi-port installations. This paper develops the distribution system impact assessment methodology for EV charging installations: transformer thermal loading analysis including the aging acceleration from EV-driven demand spikes, steady-state voltage impact analysis, IEEE 519-2022 harmonic compliance analysis for DC fast charger installations, and protection coordination verification. The analysis methodology is consistent with FERC Order 2023 fast-track screening criteria and IEEE 1547-2018 interconnection requirements for installations with bidirectional charging capability.

1. Introduction

NEC Article 625 governs the installation of electric vehicle charging systems, and NEC 625.17 requires that the charging system be sized for 100 percent of the continuous load (charging loads are continuous per NEC 625.2). The NEC sizing requirements address the conductor and overcurrent protection sizing for the EVSE circuits themselves, but they do not address the distribution system impacts upstream of the building service entrance. A parking structure installation with 100 Level 2 EVSE ports at 7.2 kW each creates a potential connected load of 720 kW — potentially larger than the building's existing service capacity — and the distribution transformer serving the facility must be evaluated for adequacy before the installation proceeds.

The distribution system impact assessment determines whether the existing infrastructure can accommodate the EV charging load, what upgrades are required, and whether demand management or smart charging controls can defer or reduce the need for infrastructure upgrades. For large EV installations, the assessment may identify that the required infrastructure upgrades have lead times (distribution transformer delivery, utility service upgrade approval) that are longer than the project schedule allows, making early assessment critical to project timing.

2. Transformer Thermal Impact

2.1 Thermal Aging Model

Distribution transformers are rated for continuous operation at their nameplate kVA rating at an ambient temperature of 30°C. Operation above the nameplate rating accelerates the thermal aging of the transformer's cellulose insulation — the paper insulation on the windings — through a reaction rate relationship described by the Arrhenius equation. IEEE Standard C57.91-2011 models the insulation aging acceleration factor as:

$FAA = e^{\left[\frac{E_A}{R}\left(\frac{1}{T_{ref}+273} - \frac{1}{\theta_H+273}\right)\right]}$

Where: $FAA$ is the aging acceleration factor relative to rated temperature operation (dimensionless).

$E_A$ is the activation energy for the aging reaction (J/mol; 111,000 J/mol for the reference insulation system).

$R$ is the universal gas constant, 8.314 J/(mol·K).

$T_{ref}$ is the reference hotspot temperature in °C (110°C for 65°C rise transformers per IEEE C57.91).

$\theta_H$ is the actual hotspot temperature in °C.

At the reference hotspot temperature of 110°C, $FAA = 1.0$ and the transformer ages at the normal rate. At 130°C, $FAA \approx 4.2$ — the transformer ages at more than four times the normal rate. An EV charging load that drives the transformer hotspot to 130°C for 3 hours per evening consumes $3 \times 4.2 = 12.6$ equivalent hours of normal insulation life per day — consuming a full week's worth of normal aging in a single day of elevated operation.

The transformer's hotspot temperature under a given loading profile is calculated using the thermal model of IEEE C57.91, which computes the hotspot as the sum of the ambient temperature, the top-oil temperature rise above ambient (a function of the per-unit loading and the thermal time constant), and the winding hotspot gradient above top-oil. For a 1,000 kVA distribution transformer with a 65°C average winding rise rating, the hotspot temperature at 150 percent of nameplate loading in 30°C ambient is approximately 135 to 145°C, corresponding to aging acceleration factors of 5 to 9 times normal.

2.2 Demand Management Impact

Smart charging systems that implement demand management — spreading the EV charging load across a longer charging window rather than allowing all EVs to charge simultaneously at maximum rate upon arrival — can substantially reduce the peak demand imposed on the distribution transformer. A facility with 50 EV charger ports where vehicles typically arrive between 5 PM and 7 PM and depart after 7 AM has approximately 14 hours of available charging time. If the average vehicle requires 30 kWh of charging energy per session, the average power required per vehicle is 30/14 = 2.1 kW — well below the 7.2 kW Level 2 charger rate. Smart charging that manages the charging rate to meet the energy requirement within the available window while keeping the aggregate demand below a set threshold can serve 50 vehicles with a peak aggregate demand of 50 × 2.1 = 105 kW rather than 50 × 7.2 = 360 kW — a 70 percent reduction in peak demand that dramatically changes the transformer sizing requirement.

3. Harmonic Impact of DC Fast Chargers

3.1 Harmonic Current Spectrum

DC fast chargers (Level 3 EVSE) are nonlinear loads that inject harmonic currents into the distribution system as a byproduct of their AC-DC power conversion. Most DC fast chargers use an active front-end (active rectifier with PWM control) that achieves near-unity power factor at the fundamental frequency and limits low-order harmonics, but still produces harmonic current at and around the switching frequency. The total harmonic distortion of the input current for a modern active front-end DC fast charger is typically 5 to 8 percent THD $_I$ at rated power, which is within the IEEE 519-2022 individual customer harmonic current limits for most service configurations.

However, when multiple DC fast chargers operate simultaneously on the same distribution feeder, the cumulative harmonic current can exceed the feeder's harmonic voltage distortion limit even when individual chargers are compliant, because the harmonic currents from multiple sources add as phasors. IEEE 519-2022 requires that the total voltage harmonic distortion at the point of common coupling between the utility and all customers on the feeder not exceed 5 percent THD $_V$ for systems at or below 69 kV. For a feeder with high source impedance (typical of rural or suburban distribution), the voltage distortion produced by a large DC fast charging installation can approach or exceed this limit.

The harmonic current drawn by DC fast chargers is illustrated in Figure 1, which plots a representative charger current spectrum.

Harmonic current spectrum of a DC fast charger. The horizontal axis is harmonic order and the vertical axis is magnitude as a percentage of the fundamental. Modern active-front-end chargers concentrate residual distortion in the lower.

Figure 1. Harmonic current spectrum of a DC fast charger. The horizontal axis is harmonic order and the vertical axis is magnitude as a percentage of the fundamental. Modern active-front-end chargers concentrate residual distortion in the lower characteristic orders, with the 5th and 7th dominating. The engineer should observe that while an individual charger may meet device-level limits, the aggregate total demand distortion at the point of common coupling from many chargers is the quantity that must be checked against the IEEE 519 limit for the site.

4. Protection Coordination

EV charging infrastructure creates two protection coordination considerations that differ from conventional commercial loads. First, Level 2 EVSE circuits are continuous loads per NEC 625.2, and NEC 625.40 requires that the branch circuit supplying an EVSE be rated at not less than 125 percent of the EVSE's rated current. This 125 percent rule changes the coordination between the branch circuit overcurrent device and the feeder overcurrent device compared to a standard 100 percent continuous load circuit. Second, bidirectional EVSE (V2G or V2H capable chargers) can inject current back into the distribution system during vehicle-to-grid discharge events, and the protection relay coordination must account for the reverse current direction as well as the forward load current.

Related Work

The analysis in this paper connects to several companion studies in this library. Readers concerned with the upstream and downstream engineering will find Electric Vehicle Charging Infrastructure develops a closely related aspect of the same problem, while Electric Vehicle Charging Load Management extends the treatment into an adjacent domain. For the broader methodological context, Electric Vehicle Charging Infrastructure Deployment provides complementary depth.

Conclusion

EV charging infrastructure imposes distribution-system impacts that require systematic assessment before installation because EV loads differ from conventional commercial loads in their temporal concentration, their power-factor and harmonic characteristics, and their coincidence with existing peaks, and the analysis developed in this paper shows that transformer thermal aging, harmonic distortion from DC fast chargers, and protection coordination must all be evaluated together. The central engineering conclusion is that the binding constraint is frequently transformer thermal capacity under the coincident evening peak rather than the nominal kVA rating, and that an automatic load-management system can relieve this constraint by capping the managed EV load below the sum of charger nameplate ratings. The harmonic analysis establishes that DC fast chargers are nonlinear loads whose current spectrum must be checked against IEEE 519 limits at the point of common coupling. For the practicing engineer, the operative takeaway is that EV charging impact assessment is an integrated study of thermal, harmonic, and protection effects, and that demand management is usually the most cost-effective mitigation for the thermal constraint that otherwise drives a service upgrade.

References

[1] NFPA 70, National Electrical Code, Article 625, 2023 edition, NFPA, 2023.

[2] IEEE Standard C57.91-2011, Guide for Loading Mineral-Oil-Immersed Transformers and Step-Voltage Regulators, IEEE, 2012.

[3] IEEE Standard 519-2022, Recommended Practice and Requirements for Harmonic Control in Electric Power Systems, IEEE, 2022.

[4] IEEE Standard 1547-2018, Standard for Interconnection and Interoperability of Distributed Energy Resources, IEEE, 2018.

[5] FERC Order 2023, Improvements to Generator Interconnection Procedures and Agreements, FERC, 2023.

[6] EPRI, Electric Vehicle and Plug-in Hybrid Electric Vehicle Impacts on Distribution Systems, EPRI Technical Report 1021785, 2010 (updated 2023).

[7] J. C. Das, Power System Harmonics and Passive Filter Designs, IEEE Press/Wiley, 2015.

[8] SAE International, SAE J1772: Electric Vehicle and Plug-in Hybrid Electric Vehicle Conductive Charge Coupler, SAE, 2017.