PyTeslaPowerSet

Check current power prices and set a Tesla charge state accordingly:

• If the Tesla is in Trip Mode™ (charge target above 90%) then do nothing
• If the power price tonight is low, then set the Tesla to charge full
• If the power price tonight is high, then set the Tesla to charge minimum

Thanks!

Relies on Michiel Lowijs's Tesla-Api and Kimmo Huoman's Nordpool

Without their work, this wouldn't have been possible -- thanks guys!

Setup

Installation

Because of the external dependencies, this requires setuptools to be installed before the other requirements, so:

python -m venv venv
python -m pip install setuptools
python -m pip install -r requirements.txt

Configuration

Uses environmental variables for things like username/password.
Add a .env file in the root with:

TESLA_USER=your_email
TESLA_PASS=your_password
TESLA_TOKEN=your_access_token
MIN_PERCENT=desired_minimum_charge_limit (e.g. 50%)
MAX_PERCENT=desired_maximum_charge_limit (e.g. 90%)
BASE_CURRENCY=desired_currency_to_use

If you already have an access token that can be used to avoid storing the credentials locally (remember the access tokens expire after 45 days)

There are also a couple of behaviour toggles early in the script, set them according to your need:

• areas - the Nordpool area to check for, NB currently only supports 1 area
• cheap_threshold - the decision threshold for whether to charge min or max (e.g. 280), NB multiply by 1000 (conversion of cost/MWh to cost/kWh)

Scheduling

Add it to crontab for scheduling, for example:
30 22 * * * user python3.7 ~/PyTeslaPowerSet/main.py > /dev/null 2>&1

Basic logic

My original thoughts for the logic

• ideally, we want tomorrow's prices because it will have both the price of tonight (midnight) and tomorrow night (11pm)
• as a backup, we can use today's prices, because it will have the price tonight (11pm) and the useless one: yesterday night (midnight between yesterday and today)
  • Conclusion: So there are two interesting prices and one back up price:
  • #1: The price tonight -- which is prices_tomorrow[0]
  • #2: The price tomorrow night -- which is prices_tomorrow[-1]
  • #3: The price tonight (1 hr earlier) -- which is prices_today[-1]

So the logic should be as follows:

• If the car is set >90% charge limit -> do nothing, we're in Trip Mode™
• If the car is set to <= 90% charge limit -> decide whether to charge much (max_percent) or little (min_percent)
  • If the price is below threshold tonight -> set charge to max_percent
    • unless it's even lower tomorrow?