EV charging
Your EV knows your habits – but does your charging system?
Why charge your car tonight if you will not need more energy until the day after tomorrow? Intelligent EV charging is not just about finding the cheapest hours, but understanding when the energy is actually needed.
↓You arrive home, park your EV and plug it in. The battery has 55% remaining, and you usually charge to 80%.
A conventional price-based solution now looks for the cheapest hours tonight.
But if you are only expected to use 10% of the battery tomorrow, why charge already?
The cheapest charging session may be the one you do not need today at all.
Cheap charging is not necessarily intelligent charging
Many chargers can already schedule charging around electricity prices. That is useful, but the price does not tell you how much energy you need or when you need it.
If the car needs 30 kWh and charges at 11 kW, charging takes just under three hours under ideal conditions. The cheapest quarter-hour periods are not necessarily consecutive, so it can make sense to spread charging across several periods.
But before the system chooses when to charge, it should ask: Does the car even need more energy yet?
What if electricity is cheaper the day after tomorrow?
Imagine the following example:
When tomorrow’s driving is already covered
| Battery level on arrival | 55% |
|---|---|
| Expected consumption tomorrow | 10 percentage points |
| Battery level after the trip | Approx. 45% |
| Electricity price tonight | DKK 2.50/kWh |
| Expected electricity price the following night | DKK 1.20/kWh |
If the car has plenty of energy for tomorrow’s driving, waiting to charge may be cheaper. And if the next few days also involve short trips, charging might be postponed even further.
That takes more than a list of cheap hours. The system needs to combine the car’s battery level, past driving patterns, expected future demand and upcoming electricity prices.
It also needs to account for uncertainty: an unexpected trip must not leave you without the energy you need.
Intelligent charging is not about always filling the car up. It is about having enough energy when it is needed.
Illustrative electricity prices over three days
Why should you have to enter your habits yourself?
Most people have recognisable patterns. Perhaps you leave home at 07:30 on weekdays, later at weekends, and rarely drive far on an ordinary Tuesday.
An intelligent charging system can learn from when the car is connected and disconnected, how much energy is delivered and the charge level at which the car normally stops.
If you usually charge to 80%, the system should be able to learn that. If you later change the car’s limit to 90%, it should gradually be able to detect the change through observations.
This is a fundamental principle in Fluks: The system should learn what it can observe instead of requiring yet another set of manual settings.
Learning starts with observations
Observes charging
When is the car connected and disconnected, and how much energy is delivered?
Learns the car’s habits
Actual charging power, usual charge level and typical disconnection times.
Adapts charging
Observations are used to plan around the available electricity prices.
The car needs to work with the rest of the home
The EV is often the home’s largest flexible electricity consumer. That is why it makes little sense to plan charging independently of solar panels, a home battery and the rest of the household’s consumption.
If the car starts charging during an expensive evening hour, the home battery may end up supplying the car, even though the car could easily wait for a cheaper period. That can be a poor deal.
The same applies to solar energy. Using every surplus solar kWh for the car is not automatically the best choice. Sometimes selling the electricity and charging later can pay off better. It depends on the purchase price, sale price and the household’s other needs.
A good plan needs to optimise the whole home – not just the charger.
One plan for the whole home
What if your plans change?
Nobody makes exactly the same trips every day. A charging system that learns habits must therefore also understand that expectations can be uncertain.
When the car already has plenty of energy, waiting may make sense. When there is a risk of greater demand, charging earlier may be wiser.
The aim is not to predict the future perfectly, but to make sound economic decisions without compromising the car’s usefulness.
Fluks – energy management that learns
Already in Fluks
Fluks treats EV charging as part of the household’s overall energy management. The system already learns from charging sessions, including the actual charging power, usual charge level and typical disconnection times, and uses those observations to plan charging around the available electricity prices.
Future extension
A future extension should be able to consider energy demand over several days and assess whether charging could usefully wait until tomorrow or the day after. This is a possibility, not a feature we claim is already finished.
The best charging is the charging you never need to think about
A conventional price-based charger asks: When is electricity cheapest?
An intelligent energy management system should ask: When do you actually need more energy – and what is the cheapest way to make sure the car is ready?
Sometimes the answer is tonight. Other times it is tomorrow or the day after.
That is the vision behind Fluks.
Your energy. Decides together.