Control and clustering of electric vehicle chargers for the provision of grid services

The transportation and power sectors are experiencing a paradigm shift. On the one hand, the transition away from fossil fuels in the transportation sector is paving the way for the emergence of electric mobility. On the other hand, the shift towards a sustainable power system necessitates novel approaches to power system operation and planning. Therefore, a synergy between electric mobility and renewable energy sources (RESs) can contribute significantly to the progress of both industries. In this context, the charging infrastructure serves as the link between the transportation and power sectors, encompassing both electrical and communication aspects. The prospects and hurdles of electrifying transportation hinge on the positioning, variety, utilization, and functionalities of this charging infrastructure.
To date, slow charging is by far the most widely utilized type of charging infrastructure for public and private charging sessions. Here, slow charging cor- relates with long vehicle parking times, allowing for better accommodating the charging energy demand in combination with the restrictions of the power grid.
This thesis investigates the potential for controlling electric vehicles (EVs) and AC chargers for the provision of grid services to benefit both the power and transport sectors. Research is focused on the AC charging infrastructure comprising electric vehicle supply equipment (EVSE) and the vehicle on-board charger (OBC). The coordination of such technologies is of paramount importance for the delivery of grid services in support of a RES dominated power system. The thesis is divided into three parts.
The first part discusses the transport-power sector coupling conundrum in a systematic way. The primary objective is to establish a connection between ancillary services and EV flexibility to aid system operators (SOs) and flexibility providers in understanding the role and optimal location of EV charging clusters in the power system. To attain this objective, a comprehensive review of ancillary services is imperative, taking into account the operational challenges of the power system. Among the diverse range of ancillary services, those that can be provided by the charging infrastructure are highlighted and classified into 12 geoelectric charging clusters. The second objective focuses on the EV flexibility supply chain and identifies seven actors regardless of geographical considerations. Here, it is important to highlight the functionalities required for EVSEs. Therefore, a smart EVSE is defined as an electric device that provides protection, communication at least scheduling, and at most modulation, phase curtailment (3 to 1-phase switch), and phase switching for the EV charging process.

The second part continues with EVSEs and OBC control capabilities by transitioning from theoretical to practical ground. Since the OBC is responsible for the conversion from AC-to-DC current (charging the battery), the first research area addresses the efficiency and reactive power curves of the smart charging operation. Here, a methodology based on CANBUS/on board diagnostics port (OBDII) readings to evaluate the characteristics of all commercial OBC is introduced and successfully validated with 38 different light-duty EVs models from the past 11 years. The results show that smart charging (by modulation) can increase charging losses from 1 to 10 %. The projections show an efficiency between 88-95% by 2030 and a saturation between 90-96% by 2035. Additionally, some models consume larger amounts of reactive power at lower currents or vice versa. Furthermore, the second objective focuses on developing and validating a method to measure the entire control loop speed (measurement-control action-EVSE-OBC) when delivering grid services. The findings indicate that OBC remains the bottleneck in providing faster grid services. Nonetheless, some automakers offer the possibility to achieve a control action of less than one second.

The third part centers around the utilization of flexibility through smart residential charging applications in Denmark and Norway. Both cases rely on real data. The Danish cases focus on simulations of behind-the-meter (BTM) applications with a novel autonomous distributed control architecture for EVSEs. The aforementioned approach seeks to enhance the overall charging experience for EV owners while aligning it with the support for the grid. The findings indicate that strategies incorporating price and emission signals not only achieve their intended objectives but also yield reductions in both costs and carbon dioxide (CO2) emissions. Furthermore, the Norwegian case improves our understanding of residential charging coincidence factor (CF) by examining the correlations with i) temperature and seasonality, ii) time of day, iii) day and time of week. This study also conducts a comparative analysis between natural (normal) and smart charging behavior. Here, considering market synchronization, the study explores the influence of smart charging on power system operations. Such large-scale infrastructure could pose the risk of adverse instant power delivery effect, as simultaneous charging could potentially strain the grid and require substantial grid investments. Therefore, the solution to the implications of synchronization lies in improving optimization algorithms to better share the available power capacity of the grid. These algorithms can be improved by considering the efficiency curves mentioned above and improving the control loop speed.