We got the power: Predicting available capacity for vehicle-to-grid services using a deep recurrent neural network

Vehicle-to-grid (V2G) services utilise a population of electric vehicle batteries to provide the aggregated capacity required participate in power and energy markets. Such participation relies on prediction available support reliable delivery agreed reserves at future time. In this work real historical trip data from fleet vehicles belonging University Nottingham was used simulation developed show how battery state-of-charge would vary if these trips were taken that charged simulated charging station locations. A time series forecasting neural network predict for next 24-h period given input previous 24 h its increased predictive capability over regression model trained using automated machine learning demonstrated. The simulations then extended include satisfy needs market events ability successfully adapt predictions such authors conclude is critical importance viability success V2G by supporting trading utilisation decisions multiple events.