Electric Vehicle Supply Equipment Day-Ahead Power Forecast Based on Deep Learning and the Attention Mechanism

Transports is one of the sectors that produce thehighest emissions of CO2; in the last ten years, there has beena process of decarbonization which has led to a considerableincrease in Electric Vehicles (EVs). However, the sudden intro-duction of a large number of Electric vehicle supply equipment(EVSE) supplying electrical energy to EVs could cause problemsin the management of the electric grid which must cope with theconsequent increase in the electrical load demand. In this context,the 24 hour ahead forecast of the power curve associated withthe recharge of EVs becomes of vital importance to ensure thereliability of the electric grid. In this paper, different MachineLearning models based on Recurrent Neural Networks (LSTM,GRU) and with different architectures, are compared based ontheir capability to accurately predict the power curve of an EVcharging station one day in advance. A Sequence to Sequencemodel has been implemented and a thorough analysis of anAttention layer has been detailed. The models are tested on areal world open dataset