Parking lots Load prediction by LSTM

With the increase of electric vehicles (EVs) and plug-in electric vehicles (PEVs) in cities, EV parking lots face many challenges in estimating the electric charge and managing the optimal charge of the vehicles. To increase the profit of the owners of EV parking lots, they must have a correct estimate of the amount of their day ahead load so that they can request from the Day-Ahead Market(DAM) at a lower price than the Real-Time market (RTM). In this paper, using the LSTM network, the amount of EVs day ahead load is estimated and compared LSTM method with other conventional methods by buying from DAM and RTM. The simulation results show that the LSTM network gives a very accurate estimate of the load and performs well compared to the actual value. In this case, parking lots can have a higher profit.