Date of Award

1-1-2022

Degree Type

Dissertation

Degree Name

Doctor of Philosophy in Industrial and Systems Engineering

Specialization

Industrial Engineering

Department

Mechanical, Industrial and Systems Engineering

First Advisor

Gretchen A. Macht

Abstract

Policymakers regularly intensify efforts to increase electric vehicle (EV) adoption to reduce greenhouse gas (GHG) emissions caused by the transportation sector. Despite various incentives, EV adoption has certain drawbacks, including limited ranges, long charging times, and sparse electric vehicle supply equipment (EVSE) infrastructure. Access to EVs is often a privilege for specific communities, causing inequity in adoption and access to public charging infrastructure. To enable a user-centered and equitable deployment of EVSE infrastructure, a detailed understanding of EV charging behavior and its different facets is crucial. This research establishes a holistic approach to understanding user behavior using real-world charging data collected at various public and residential EVSE networks.

Throughout this work, multivariate charging session data is analyzed for EV user charging behavior using various statistical methods, advanced algorithms, and time series analysis. This novel and holistic approach to understanding charging station utilization offers new insights into EVSE network service level requirements and explains the relationship between charging behavior, urbanicity, and locale.

The findings of this work indicate that significant differences in charging behavior between charging networks exist. An urban-rural divide in charging station utilization is identified, which is inconsistent throughout different areas and regions. Users from rural areas charge further away from their homes than urban users, indicating a lack of EVSE in rural areas and supporting the existence of an urban-rural divide in EVSE access. Charging behavior is further found to differ significantly between provinces, especially regarding intensity and etiquette. The results show that longer charging times usually concur with lower charging frequencies, reducing the service level and profitability of EVSE.

Residential and public charging demand follows an inverse daily pattern, represented by a bi-daily switch in peak station utilization. While public charging demand times are concentrated during the day, residential EVSE is mainly used in the evening and throughout the night. The switch in charging demand proposed the “EV Duck Curve” and suggests that residential charging behavior can increase the renewable energy “Duck Curve” issue, which may be further amplified by a future increase in public direct current fast charger deployment.

This work aims to demonstrate the differences in human charging behavior in various contexts by applying robust and multifaceted methods. Current EV charging behavior research lacks a holistic approach that incorporates various types and locations of charging networks when exploring user behavior. This research suggests that findings on charging behavioral patterns in one network are not transferable to another, thus requiring different charging networks to be explored and understood separately. Therefore, policymakers, network providers, and electric utilities should develop detailed EVSE infrastructure plans at the meso-level (e.g., geospatial, regional, or community levels) that serve users across different communities and satisfy disparate charging preferences while ensuring grid stability and EVSE adoption.

Available for download on Friday, May 17, 2024

Share

COinS