CIRED workshop on E-mobility and power distribution systems Porto, 2-3 June 2022 Paper n° 1255 CIRED 2022 Workshop 1/5 POTENTIAL BENEFITS OF SCHEDULING ELECTRIC VEHICLE SESSIONS OVER LIMITING CHARGING POWER Marc CAÑIGUERAL Joaquim MELÉNDEZ University of Girona – Spain University of Girona – Spain marc.canigueral@udg.edu joaquim.melendez@udg.edu ABSTRACT This work addresses the management of charging infrastructure to cope with the problem of congestion in the electric grid by exploiting electric vehicle (EV) flexibility. Benefits of scheduling charging sessions based on user profiles is compared with two traditional methods for limiting charging power: using static a dynamic control signal. The user profile concept is a classification strategy that consists in assigning a label to every EV user, referring to a connection pattern in the daily use of the charging infrastructure. The study analyses pros and cons of these three approaches and highlight the advantages of demand-response programs based on the user profiles. Thus, while limiting the power of charging sessions with a static signal causes a rebound effect, using a dynamic signal requires impacting a high amount of sessions. Scheduling (i.e. postponing) the charging sessions associated to profiles with low-variance provides a higher efficiency of the demand-response program since the same objective (i.e. peak demand reduction to 350 kW) is achieved with a lower number of participating sessions (40% fewer than limiting charging power with dynamic signal). INTRODUCTION The electrification of final energy demand that consumes fossil fuels as energy source, such as transportation or heating systems, can lead the distribution grid to critical congestions during peak hours, mostly in low-voltage distribution grids and at substation level [1]. At the same time, the challenge that supposes the introduction of electric vehicle (EV) also provides an opportunity to distribute the storage resources and make use of a strong flexibility potential. The flexibility that an electric vehicle can provide to the electric system is of a special interest for the DSO in order to support congestion management at specific geographic locations [2]. The flexibility of a single vehicle is small, but the aggregated impact of a fleet or the management of large charging infrastructures (e.g. public charging stations) can be significant to participate in flexibility markets through aggregators [3]. Thus, EV flexibility aggregator could be a new market player together with other actors already active in the EV sector, such as parking or fleet owners, Charging Point Operators (CPO) or e-mobility service providers (EMSP) [4]. At the same time, the method to control the EV charging process can be also diverse, for example the charging point (communication Aggregator - CPO), the EV itself (communication Aggregator - Car manufacturer) or the Home Energy Management System (communication Aggregator - HEMS). The following sections describe and analyze three different approaches to manage the EV load for congestion management from the aggregator point of view: (1) static signals to limit charging power, (2) dynamic signals to limit charging power and (3) dynamic signals to schedule (i.e. postpone) charging sessions. Section 2 describes the three flexibility strategies and learnings from real pilot projects aiming to highlight the contributions of this paper. Section 3 presents the simulations done for these three smart charging methods. Finally, Section 4 concludes with main results from these simulations. SMART CHARGING FOR GRID CONGESTION MANAGEMENT During last years, several pilot projects in Europe have evaluated the local impact of EV charging in the electric network and the methods to avoid grid congestions. One of the most extended methods is the limitation of the power that the charging points can supply with a static control signal or power profile. An example of a successful pilot is the Flexpower project in Amsterdam [5], where the phase current of 39 public charging stations (25A nominal) was limited to 20A from 07:00–08:30 and to 6A from 17:00–20:00, but increased to 35A during the rest of the day. In the second phase of the same project, Flexpower 2 [6], the phase current was limited from 18:00 to 21:00 to a maximum of 8A, and the rest of the day to 35A, except during the cloudy days (no PV production), when the current was limited to 25A from 6:30 to 18:00. Another pilot that implemented a static power profile limitation was carried out by the Dutch DSO Enexis [7], using domestic charging points in this case, reducing the charging current from 17:00 to 22:00 to a maximum of 6A. However, both Dutch pilots experienced some drawbacks on their implementations. The most direct side-effect of constraining the EV load between a peak period is a rebound effect. The energy is shifted from the constrained to the unconstrained time period. Even though a rebound