Electric Vehicle Charging Pattern Prediction
- Sponsored by: sonnen GmbH
- Scientific Lead: Dr. Stefan König
- Project Lead: Dr. Ricardo Acevedo Cabra
- Term: Summer semester 2018
Sonnen's home storage system (the "sonnenBatterie") allows its users to store their self-generated energy for later consumption. Typically, the battery stores energy generated by the user's photovoltaic (=PV) installation during the day and allows the user to use this energy during evenings and nights. To maximize the payoff of the PV installation for the user, the battery comes with an intelligent energy management that adapts charging and discharging to the user’s needs. For this purpose, various predictions (e.g. PV production, energy consumption) are already in place and allow intelligent control of the battery. Via smart plugs and various smart home protocols, the battery can also turn on and off large appliances and further optimize the user’s energy use. In this setting, charging electric vehicles becomes a more and more common use case and heavily interferes with the energy management at home. Therefore, the goal of the present project is to build prediction models for charging patterns of electric vehicles. Given data about past charging sequences, the goal is to predict when the vehicle will be charged the next time, how much energy will be needed, how much time will be available for charging, etc.
The results of this project are sumarized in the final presentation and explained in detail in the final documentation.