TRCLC 15-8

Community-Aware Charging Station Network Design for Electrified Vehicles in Urban Areas:  Reducing Congestion, Emissions, Improving Accessibility and Promoting Walking, Bicycling and Use of Public Transportation

PIs: Ratna Babu Chinnam and Alper Murat (Wayne State University)

Summary:

Electric vehicles (EVs) hold many promises including diversification of the transportation energy feedstock, reduction of greenhouse gas and other emissions, and improving public health by improving local air quality. We developed a set of tools to support effective planning of network design for charging stations for EVs in urban areas. Such infrastructure deployment also presents a number of unique opportunities for promoting livability while helping to reduce the negative side-effects of transportation (e.g., congestion and emissions).

Problem:

In this project, we aim at developing and demonstrating (at the proof-of-concept level) a system for the design and deployment of the charging infrastructure in support of the increasing adoption of electrified vehicles to improve livability (reduced congestion, noise, improve walkability) in urban areas, help ease users range anxiety, reduce user costs (e.g., walking), and reduce infrastructure cost. Our vision is to develop analytical data-driven tools and demonstrate that strategically planned and incentivized deployment of charging stations in urban areas would lead to improved livability of urban areas and these benefits will continue to increase with increased adaptation of the electrified vehicles.

Research Results:

To show the efficiency of the two-stage stochastic model, we conducted a variety of experiments using synthetic small and large networks. Characteristics of small and large networks are represented in the following table:

 The preliminary results of finding optimal locations for installing charging stations for small and large networks using Sample Average Approximation method are presented in figures below:

Figure 1: Optimal locations (blue) of charging stations in small network for cases of p = 2 (left) and p = 3 (right) when market share is 1% for BEV and 2% for PHEV. 

Figure 2: Optimal locations (blue) of charging stations in large network for cases of p = 1 (left) and p = 2 (right) when market share is 2% for BEV and 5% for PHEV. 

Average of accessibility to charging stations and average of walking distances are computed for different scenarios of BEV and PHEV market shares in the following tables. Negative values for walking imply that people are willing to walk less if we can install charging stations at the optimal locations. 

 Results:

The two-stage stochastic programming model and the resulting tools are expected to be used by planning agencies in the future. In continuation of the first phase, we expect to extend the study in a second phase. The second phase will enhance the modeling framework in following ways:

a)     Identification of accessibility range for the community for the proposed EV stations network and include a minimum coverage requirement, and assess the impact of uncovered regions within a community.

b)     Inclusion of multi-mode of transportation for a community and incentivize the potential EV charging stations based on their reachability for multi-modes of transportation, and gauge the shift in drivers’ adoption of such EV charging stations.

c)     Provision to quantify the robustness and sensitivity of prescribed network design with respect to the changes in the arrival pattern, walking or adoption behavior of drivers and estimate the influence of pricing scheme on a network design; this will provide useful insights due to the randomness used for the model.

d)     Evaluation of the model with a pilot study for a community by partnering with a regional planning agency such as the SEMCOG.

Presentation

Final Report