Incorporating valley filling and peak shaving in a utility function based management of an electric vehicle aggregator

Shifting load away from the system peak into evening hours when the load is low and the network´s capacity is high is referred to as peak shaving and valley filling. This paper develops an approach to enforce such a charging behavior to a large fleet of individual electric vehicles. The vehicles move in a transportation network which is mapped to a realistic urban electricity distribution system. Each vehicle features a utility function based on which the charging is controlled. The algorithm tunes the utility function parameters order to initiate the desired charging behavior. Individual behavioral constraints, such as a desired state of charge when leaving, are fully considered.