Electric car charging already uses intelligent functions. But we can surely do even better.
You have surely noticed that our cars are becoming more and more “intelligent”. You will also have certainly noticed that artificial intelligence is one of the biggest topics today, not only for fans of technology and innovation, but for the entire society, which it promises to profoundly transform. .
AI is revolutionizing many sectors, and electric car charging should be no exception. There is no doubt that various scenarios are already being studied within the R&D departments of large operators, and that AI building blocks are now integrated into the processes that manage charging.
However, could we imagine other applications of AI in charging, particularly from the user’s point of view? Here are 5 possible scenarios which could perhaps improve the efficiency, reliability and profitability of charging infrastructures, to offer a better experience to electric motorists. Note that these are only forward-looking hypotheses that we believe are feasible given the current state of technology. Some may seem far-fetched to you, and we do not claim that they are all destined to develop.
The charging station chases away squatters
In this first scenario, the charging station is equipped with a camera which identifies the car model and registration. If it is a thermal car that comes to squat in the charging space, a warning message is displayed on the terminal screen, and an alarm sounds for a few seconds. Second step, a fine is automatically generated after a few minutes of squatting, in fact following the operating principle of automatic speed cameras. Same process for suction cup cars, with the difference that it is triggered X minutes after the end of the recharge. Is it brutal? Certainly, but ultimately no more than any automatic verbalization system, right?
The charging station recognizes the car… and the driver
The idea is to offer an adapted and personalized service based on the driver’s habits. Once the car is connected and identified, its history is analyzed in order to launch a service that corresponds to its latest uses. If it is someone who is used to stopping for short periods of charging, the flow rate increases in order to provide the most energy in the least amount of time. If, on the contrary, it is a driver who takes his time, the flow modulates to provide the same quantity over a longer time. Advantage for the load point manager, the load is modulated according to the cases and the distribution is therefore better distributed and more optimal. There is no point in sending 270 kW to someone who is going to lunch. A modularity obviously adapted to the occupancy rate of the station.
The charging station communicates remotely with the car and its planner
Like Waze which displays geolocated advertisements when you pass near a center of interest, the charging point is connected to a Google Maps type API and analyzes the surrounding traffic and traffic in real time. state of charge of batteries in circulation to send a message encouraging motorists to come and charge, for example by highlighting a promotional rate per kWh. Other promotions could be activated, such as restaurants or businesses near the terminal. The ultimate: electric bicycles or scooters would be made available at the charging point – free of charge or rental – to allow electromobilists to move around the area during charging. All managed by an unlocking system by recognition of the car which would prevent theft and damage.
The charging point is connected to news and weather
AI could play a key role in optimizing electricity networks for more efficient charging. By analyzing consumption data, weather and other parameters, AI algorithms could predict peaks in demand and dynamically adjust electricity distribution. This approach would avoid overloading networks, optimize the use of renewable energies and minimize costs for end users. Better still, the charging station would be able to understand and analyze its environment and current events to predict charging needs, for example in the event of strikes, severe bad weather, vacations, etc. or road blockages following a social conflict.
The charging station… discharges you
AI can help optimize smart charging by analyzing data from charging stations, EVs, the power grid and renewable energy sources. It can thus adjust the power and timing of charging, for example to promote the use of green energies by synchronizing charging with solar or wind production, or to manage charging priorities between different EVs depending on their autonomy, their use (or their identified intention of use), their location, etc. But better, we could also imagine that it contributes to the flexibility of the network by intelligently and discerningly suggesting to cars with a full battery without having a planned use of it in the short term to reinject current into the network, in return for a payment in in cash or in another form (purchase vouchers, etc.).
There are probably many other scenarios to imagine, which will be integrated into our lives as electromobilists in the medium term, as Aurélien de Meaux, CEO of Electra, suggested during our last interview. Furthermore, certain features offered by networks such as that of Tesla or, even, Electrademonstrate that a dose of intelligence is already injected into the processes.
One pitfall, however, is that all these features assume that the charging points have access to private, and for some, quite confidential, user data. This would therefore only be possible with the informed consent of the latter, and provided that these devices are solidly governed by the rules of the GDPR, which should be taken into account and obviously respected. The management of the recharge could then be done in a process identical to that which we know with cookies on our computers, or with our conversation history in ChatGPT.
A sensitive subject, but unfortunately we don’t get much from the machines if we don’t give them some data to digest in exchange…
rewrite this content and keep HTML tags