Abstract:
|
Understanding when and where battery electric vehicles (BEVs) are charged is important for the planning and design of charging network. Based on the data of a stated choice experiment among current BEV owners, we applied dynamic discrete choice models (DDCMs) to understand the charging choices of BEV drivers on long-distance trips. DDCMs explicitly account for the stochastic nature of the charging decisions: decisions at the earlier stations influence the utility of the future choices; the expectation of the future options influences those earlier decisions; and choices are made under the uncertainty of future energy consumption. This is a marked departure from prior work on charging behavior modeling, which has largely treated the sequential choices as independent. The results show that battery state of charge and whether the vehicle can reach the next station with the remaining range are the primary factors contributing to the charging decisions. Based on the model results, we also show the monetary value of increasing charging power, moving the charging stations closer to highway exits, and having amenities such as restrooms, restaurants, and WiFi near the charging stations.
|