Impact analysis of electric vehicles on distribution systems considering uncertainties

This paper proposes a stochastic modeling and simulation technique for analyzing the impacts of electric vehicles (EV) charging on distribution network. Different from the deterministic models used in the previous studies, the models for feeder daily load profile, EV start charging time, and battery state of charge (SOC) during charging are derived based on actual measurements or survey data, and represented as stochastic parameters using Roulette wheel selection concept. Voltage and congestion impact indicators are defined and comparison of deterministic and stochastic analytical approaches in providing information required in distribution system planning for accommodating EV charging needs is conducted. Comparative results show the capability of stochastic models in reflecting system loss and security impacts due to EV integrations. Information about security risks such as over-current and under-voltage can be considered for optimal network reinforcement planning. A mitigation scheme with a controlled charging algorithm that can be used to relieve operation problems is also presented.