Uncertainty quantification in load profiles with rising EV and PV adoption: the case of residential, industrial, and office buildings

Aiko Fias, Md Umar Hashmi, and Geert Deconinck · 2025

Read Uncertainty quantification in load profiles with rising EV and PV adoption: the case of residential, industrial, and office buildings by Aiko Fias, Md Umar Hashmi, and Geert Deconinck (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-01914.txt.

Introduction

Rooftop solar panels and electric vehicles are reshaping how buildings draw power, hour by hour, in ways grid operators struggle to predict. Aiko Fias, Md Umar Hashmi, and Geert Deconinck compare methods in 2025 for quantifying that uncertainty in net electricity load profiles.

They study three building types, homes, industrial sites, and offices, under growing adoption of solar generation and vehicle charging. The question is which metrics honestly capture the spread of possible loads at the local distribution level. Choosing the right measure matters, because planners size equipment and set prices around these uncertain forecasts.

Catalog record

CREATORAiko Fias, Md Umar Hashmi, and Geert Deconinck
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE73 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2512-01914

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Uncertainty quantification in load profiles with rising EV and PV adoption: the case of residential, industrial, and office buildings Aiko Fias, Md Umar Hashmi, and Geert Deconinck Abstract The integration of photovoltaic (PV) generation and electric vehicle (EV) charging introduces significant uncertainty in electricity consumption patterns, particularly at the distribution level. This paper presents a comparative study for selecting metrics for uncertainty quantification (UQ) for net load profiles of residential, industrial, and office buildings under increased DER penetration. A variety of statistical metrics is evaluated for their usefulness in quantifying uncertainty, including, but not limited to, standard deviation, entropy, ramps, and distance metrics. The proposed metrics are classified into baseline-free, with baseline and error-based. These UQ metrics are evaluated for increased penetration of EV and PV. The results highlight suitable metrics to quantify uncertainty per consumer type and demonstrate how net load uncertainty is affected by EV and PV adoption. Additionally, it is observed that joint consideration of EV and PV can reduce overall uncertainty due to compensatory effects of EV charging and PV generation due to temporal alignment during the day. Uncertainty reduction is observed across all datasets and is most pronounced for the office building dataset. I Introduction The energy sector is undergoing a rapid transformation driven by the urgent need to decarbonize. In this transition, distributed energy resources (DERs) such as photovoltaic (PV) systems and electric vehicles (EVs) play a crucial role [ 12 ] . Their use supports climate goals. However, these technologies also introduce new issues. Unlike traditional centralized power generation, DERs are unpredictable and controlled by users, making them inherently uncertain, introducing challenges for grid operators in terms of operation and planning [ 28 ] . Further, the high simultaneity factor 1 1 1 Simultaneity factor of DERs measures the probability/extent to which multiple DERs (e.g. PV, EV, batteries etc.) produce or consume at the same time. of DERs showcased in [ 13 ] for EVs and in [ 21 ] for PV generation necessitates the impact assessment of large-scale integration of DERs. At the distribution level, high DER penetration alters traditional load and generation patterns. For instance, PV systems inject variable power based on solar irradiance conditions, while EVs can create sharp, localized demand surges during charging periods. As a result, traditional deterministic grid planning methods are no longer sufficient. Distribution networks (DNs) must be designed and operated under uncertainty, accounting for probabilistic variations in both load and generation. To address this, uncertainty quantification (UQ) methods have gained attention [ 29 ] . These methods aim to characterize and represent the uncertainty inherent in stochastic variables, such as net load, in order to support system operation and planning under uncertainty. To model the uncertainty, it is crucial to understand how to quantify it. This motivates the present work, where we aim to identify UQ metrics for growing PV and EV penetration. UQ methods are typically classified as aleatory and epistemic [ 8 ] . In this work, only aleatory UQ is considered, as we do not assume additional parameter measurement other than consumer load profile time-series data for UQ. I-A Literature Review UQ in load profiles is often achieved through statistical metrics such as mean and standard deviation. These metrics can be used for example to fit parametric distributions to historical data. For example, [ 23 ] models load currents using the beta distribution. To fit this distribution to the historical data, only the mean, standard deviation, minimum and maximum are needed [ 24 ] . Another study using the beta distribution is [ 30 ] . [ 32 ] evaluates the Weibull and log-normal probability distributions for household electricity consumption. It used the mean and standard deviation of the data and the fitted distribution to evaluate variations between the two. In [ 7 ] , the uncertainty in electricity demand is represented by a Gaussian distribution. The Gaussian distribution is fully characterized by the mean and standard deviation. The research on uncertainty in load profiles is not solely focused on using statistical metrics to fit probability distributions. [ 27 ] , for example, conducted a statistical analysis of load curves for residential, commercial, and industrial consumers. The study recommends using representative load curves, which are defined statistically by the mean and standard deviation. [ 26 ] and [ 33 ] also used mean and standard deviation to represent uncertainty in daily load profiles. Additionally, this last study uses entropy as a measure to quantify uncertainty. Authors in [ 1 ] use 705 building load profiles to training for predicting standard deviation curves, as a que for uncertainty quantification of 2250 unseen curves from 30 buildings. Table I summarizes a few studies and highlights the features of the data used for UQ. TABLE I : Used metrics in literature to quantify uncertainty Paper Mean Std Higher Monthly Min/ Entropy ID Dev moments consumption Max [ 23 ] ✓ ✓ ✓ [ 30 ] ✓ ✓ ✓ [ 32 ] ✓ ✓ [ 7 ] ✓ ✓ [ 27 ] ✓ ✓ ✓ [ 33 ] ✓ ✓ ✓ [ 38 ] ✓ [ 26 ] ✓ ✓ ✓ [ 18 ] ✓ ✓ [ 4 ] ✓ ✓ [ 9 ] ✓ ✓ ✓ [ 37 ] ✓ ✓ A lot of research considering DERs, takes either EV or PV into account in their analysis. Studies such as [ 7 ] , [ 15 ] , and [ 9 ] analyzed the impact of PV uncertainties on the network. Other studies like [ 18 ] , [ 25 ] , and [ 6 ] , have addressed the uncertainties associated with EV in their analysis. However, many studies start to focus on how these uncertainties affect distribution grids when both PV and EV are present simultaneously. For example, [ 31 ] looked at both PV and EV uncertainties to analyze their impact on the low voltage network. The study found that the demand by EV is compensated by the power injections of PV, in effect causes a lower burden on the DN. Despite the variety of approaches in existing literature, several important gaps remain. Most studies rely on basic statistical metrics, primarily the mean and standard deviation, to quantify the uncertainty, which may not fully capture the complexity of load variability under high DER penetration. Furthermore, still a lot of works focus on either EV or PV separately, neglecting the combined and potentially interacting effects of multiple DERs on net load uncertainty. I-B Contributions This paper addresses the above research gaps through a comprehensive uncertainty quantification study focused on net load profiles impacted by both PV and EV penetration. The contributions of this work are the following: • A variety of uncertainty quantification metrics , classified as baseline-free, baseline-dependent and error-based metrics, are evaluated based on their usefulness in quantifying uncertainty in behind-the-meter load profiles with increased EV or PV penetration for residential consumers, industrial consumers and office building load. This leads to recommendations on suitable metrics for each consumer type. • The uncertainty sensitivity of the different types of load profiles towards per-unit increase in EV and/or PV penetration is analyzed. The results indicate which DER contributes the most to specific aspects

Read the complete text