Perpustakaan Digital ITB

Motivated by environmental concerns, electric vehicle (EV) uptake around the world is expected to increase dramatically in the future. Distribution system operators must be able to anticipate the increased load caused by EV integration and mitigate against network asset capacity overloading issues. In this thesis, a mitigation approach suited to distribution networks is proposed that is based on the penetration of smartcharging control devices into consumer households. This allows for the control of loads connected to specific distribution network assets, such as distribution transformers. Multivariate probabilistic modeling used to estimate EV charging demands, daily travel mileage and plug-in times. A Monte Carlo simulation approach is developed to obtain the EV charging load. Mitigation is achieved through fuzzy logic control, for which State of Charge (SOC), the transformer loading, and discount time are inputs. The effectiveness of mitigation in terms of reducing distribution transformer peak load levels and losses, as well as improving voltage stability is demonstrated for a distribution network in Jakarta, Indonesia.