Capturing Multivariate Dependencies of EV Charging Events: From Parametric Copulas to Neural Density Estimation

TL;DR

This paper introduces advanced dependence modeling techniques, Vine copulas and neural density estimation, to accurately capture complex relationships in EV charging data, improving synthetic event generation.

Contribution

It is the first to apply Vine copulas and CODINE neural estimation to EV charging, outperforming traditional methods in modeling joint dependencies.

Findings

Vine copulas and CODINE outperform parametric models in capturing EV charging dependencies.

These methods better preserve tail behaviors and correlation structures.

They are highly competitive with state-of-the-art benchmarks like Gaussian Mixture Model Networks.

Abstract

Accurate event-based modeling of electric vehicle (EV) charging is essential for grid reliability and smart-charging design. While traditional statistical methods capture marginal distributions, they often fail to model the complex, non-linear dependencies between charging variables, specifically arrival times, durations, and energy demand. This paper addresses this gap by introducing the first application of Vine copulas and Copula Density Neural Estimation framework (CODINE) to the EV domain. We evaluate these high-capacity dependence models across three diverse real-world datasets. Our results demonstrate that by explicitly focusing on modeling the joint dependence structure, Vine copulas and CODINE outperform established parametric families and remain highly competitive against state-of-the-art benchmarks like conditional Gaussian Mixture Model Networks. We show that these methods offer superior preservation of tail behaviors and correlation structures, providing a robust framework for synthetic charging event generation in varied infrastructure contexts.