• EVs

    Current projects:

    • Distributed optimal charging
      Simultaneous charging of EVs at the maximum rate might overload different parts of a distribution network. Assuming that variable-rate charging of EVs is possible, we propose a stable, efficient, robust, and fair distributed control mechanism that adapts the EV charging rate to the status of the grid.

    Past projects:

    • Car pools to reduce range anxiety
      While BEVs are suitable for the majority of drivers’ daily commutes, BEV owners may need to use an ICEV for long trips. We propose placing pools of ICEVs to be used by BEV owners  subscribed to the service. When in need, subscribers would drive to their assigned pool and drive away with an ICEV with high probability. Electrical charging outlets could be kept at the pool so subscribers return to their BEVs fully charged.
    • Mining electric vehicle opinions
      In this project, we are building an opinion mining system that analyzes text from electric vehicle ownership forums and classifies sentences that express opinions.  Such a system can provide a fast, automated and inexpensive way to understand how owners and prospective buyers perceive electric vehicles.
    • Agent-based simulation of the EV ecosystem
      We study an EV ecosystem model that represents a complex system with diff erent entities, relationships,material flow, outcomes, and points of control. In a system such as this, making changes to certain entities often has direct and indirect consequences. As a result, the EV ecosystem model is a potential tool for planning new or existing EV ecosystems, estimating the possible impacts of policies on the EV ecosystem, and forecasting the penetration of EVs under di fferent scenarios. The main contributions of this work include the creation of a complete EV ecosystem model that makes use of existing modeling and simulation methods to represent the dynamics of an EV ecosystem and the simulation of di fferent EV ecosystem scenarios. We conducted case studies on San Francisco and Los Angeles, California, and we found that EV rebates and cost reductions are the most effective means of encouraging EV adoption.
    • Transitioning taxis to Electric Vehicles
      This study investigated if taxi companies can simultaneously save petroleum and money by transitioning to electric vehicles. A process to compute the return on investment (ROI) of transitioning a taxi corporation’s fleet to electric vehicles was proposed. Bayesian data analysis was employed to infer the revenue changes associated with the transition.
    • Balancing network generation and load using EV fleet charging
      The power grid must always balance network load with production. This balance is  achieved through a subset of generators in the network. The output of these generators is continuously changed to match the amount of load in the network. This balance can be equivalently achieved by varying the consumption rate of a number of loads in the grid, as long as there is some level of flexibility in their consumption profile. One of the potential candidates that can be used for this purpose is a fleet of electric vehicles (EVs): When charging a fleet of electric vehicles, we usually want to finish the charging process by a given deadline; however, when and how fast this charging is performed could be flexible. In this project, we study how this potential can be optimally exploited without affecting the main functionality of the fleet.