Both issues are caused by the assumption that once an EV is plugged in to the charging station it should receive power. An alternative configuration would be that many simple charging stations share one power cable from the grid, and then the question is how to distribute this power onto the many attached EVs. Consider for instance one person who will not need her EV before the next morning, whereas the neighbour will need the car for a sports event two hours after coming home from work. Then the power could be used to charge the second person's EV first, and then the overnight charging could happen at lower power to increase the number of cars that can be charged simultaneously. There are currently systems available that will allow the reservation of a charging station at a particular time slot, but tomorrow's charging systems needs to be intelligent and base the power distribution on three factors: What is the current state of charge (SoC) for the EV? What is the time the EV will be needed? What is the desired SoC when the EV is needed?
What will I do?
- Develop a model for EV batteries and charging characteristics of different EV models
- Develop an app for EV users to provide information on the three factors over a period of time
- Develop an algorithm based on learning and collaborative game theory for a district charging station controller aiming to maximise the number of EVs that can be charged in day based on the data collected from real EV users
- Investigate by simulation the impact of different pricing models or rewards for collaborative behaviour on the number of EVs that can be charged in a day.