Desarrollo de un algoritmo de gestión de la demanda energética de un cargador residencial de baterías para vehículos eléctricos en un escenario de tarificación en tiempo real

Abstract

Development of a scheduling algorithm for a residential eV charger under a Real Time Pricing (RTP) scheme as a Demand Side Management (DSM) strategy for reduce load spikes is presented. The proposed algorithm seeks to reduce the kWh cost of charge cycle taking into account user necessities and driving behavior. It is proposed that the scheduling algorithm was embedded on eV charger hardware thus it has to be a low computational complexity algorithm. The proposed scheduling algorithm is composed of three subroutines: at first, tariffs of available time horizon are forecasted through a trend and seasonality simple model in order to do an initial guess of where lower pricing intervals are located; thereafter, with the aim of schedule the charge cycle lowest pricing intervals combination is found using a direct insertion sort algorithm; finally, the scheduled combination is executed with possible adjustments on execution time in function of current real-price. The scheduling algorithm performance is evaluated on a simulation study where user driving behavior and available time horizon length are modeled stochastically and ComEd RTP historical data are used. Lastly, a hardware module as an IoT solution that could be integrated into an eV smart charger is proposed. Such module comprises a user interface implemented on an MSP-EXP430F5529LP Texas instruments microcontroller and a QVGA K350QVG-S1 Kentec touch screen and internet access through a CC3200 Simplelink Texas Instruments Wi-Fi module and a conventional home access point so that device can get current time by STP protocol and access to ComEd API price by HTTP protocol.