Herding from Uncoordinated Smart Charging of EVs: A Real-Life Demonstration

Electric vehicle (EV) adoption is accelerating globally, raising concerns about distribution network stability and the effectiveness of smart charging strategies. While time-of-use based smart charging approaches are widely promoted as a solution to manage network impact and demand, this paper presents empirical evidence that uncoordinated smart charging where EVs self-schedule based on the time-of-use tariff signal or algorithm may lead to unintended consequences, specifically a "herding effect" where EVs collectively create new demand peaks. Through empirical analysis of data from a field trial, this paper demonstrates that at 20–30% EV penetration levels, uncoordinated smart charging creates demand peaks that exceed traditional evening peaks. A localised demand diversification service (LDDS) is proposed in this paper, enabling coordinated, adaptive, and context-aware demand management as a mitigation strategy, and its real-life implementation is demonstrated. The empirical evidence on the herding effect and the proposed LDDS provide actionable insights for network operators and flexibility service providers.