The goal is to be able to charge the home battery with cheap energy during the night. But only if the expected solar production of the following day is not sufficient. Or if the battery is almost empty and some energy is needed to cover the (expensive) peak prices in the morning. It is Not intended for trading purposes or peak shaving, the only goal is to optimise self use at the best prices.
The following components are needed to setup this service. I mentioned the version numbers I have used during the development of this code.
- Home Assistant 2026.5.1
- Alpha ESS HA integration, version 0.8.4
- Forecast.Solar, with personal account. This gives more frequent updates (every 30 mins) compared to the free version (1 hour) and also a longer forecast.
- Nordpool for price info, based on Epex. I found this to be more stable than Entso-e which had crashed my HA install a couple of times.
- Pyscript version 2.0.1, needed to run the script.
- Terminal & SSH for troubleshooting and testing.
Add this to configuration.yaml to enable pyscript to import the required libraries
pyscript:
allow_all_imports: true
hass_is_global: trueAnd
logger:
default: warning
logs:
custom_components.pyscript: infoTo enable the logger to log all activities. This is needed for debugging and to check what happened. HA needs to be restarted after this.
Make changes to smart_battery_charge.py such that it matches your system and wishes. If you have renamed your Alpha ESS you need to change that here as well. I just use the default name with the serial numbers. Also update the NordPool sensor for your situation. The script uses the 15, 30, and 60 minute charge buttons which are available in the Alpha ESS integration. It basically presses these buttons during the night if the requirements (low price and low solar power next day) are met.
The Charge Rate is measured on two different occasions. Close enough to the theoretical maximum of 5.04 kWh/hour of my setup, a 10 kW convertor and 9.3 kWh battery.
Nightly consumption is relevant as there is some time between planning the best charging moment and the actual charging. During that period the battery drains due to base consumption by stand-by equipment, boiler, refrigerator and such. I found this to be around a constant use of 350 W. This is (much) higher during the winter as the heatpump is running to keep the house warm. Currently estimated at a constant use of 800 W during the night. Charging an electric car is not included in this. Also be aware this is an average number which may fluctuuate a lot due to additional use by for example a running dishwasher.
BATTERY_CAPACITY_KWH = 9.3
CHARGE_RATE_KW = 4.8 # Your alphaESS max charge rate (kW)
NIGHT_START_HOUR = 23 # Start of cheap night window
NIGHT_END_HOUR = 6 # End of cheap window (next morning)
MIN_CHARGE_KWH = 0.5 # Skip if less than this is needed
MAX_PRICE_EUR = 0.30 # Never charge above this price (€/kWh)
MIN_SOC_FLOOR_PCT = 30 # Always charge to at least this SOC %
# Seasonal base consumption
MONTH = datetime.now().month
if MONTH in [11, 12, 1, 2, 3]: # Winter months
BASE_CONSUMPTION_KW = 0.80 # Higher due to heatpump
else:
BASE_CONSUMPTION_KW = 0.35 # Summer base load
# ── Sensor entity IDs — adjust these to match YOUR setup ──────────────────────
SOLAR_SENSOR = "sensor.energy_production_tomorrow"
SOC_SENSOR = "sensor.alpha_ess_energy_statistics_ald071026xxxxxx_ald071026xxxxxx_instantaneous_battery_soc"
NORDPOOL_SENSOR = "sensor.nordpool_kwh_nl_eur_3_095_021"
SERIAL = "ald071026xxxxxx"
BTN_BASE = f"button.alpha_ess_energy_statistics_{SERIAL}_{SERIAL}"
BTN_15 = f"{BTN_BASE}_15_minute_charge"
BTN_30 = f"{BTN_BASE}_30_minute_charge"
BTN_60 = f"{BTN_BASE}_60_minute_charge"
BTN_RST = f"{BTN_BASE}_reset_charge_discharge"Make sure the pythin file is on the correct location, Home Assistant (the Pyscript integration) expects it in pyscript/, so the folder should be at the same level as configuration.yaml. Do not create subfolders because the script will not be found.
Copy the script to the correct location and activate it via Developer Tools → YAML → Pyscript Python scripting. This needs to be done after each change to the code.
This script shall run in the background, and twice per day it makes a charging plan for the night, one at 13:30 when the new prices are known and one at 22:00 at which time the latest solar forecast data is available. Only the last one will be used for charging. The other is just for info.
If you want to have this on a dashboard you will need to make two helpers: Settings → Devices & Services → Helpers → Create Helper → Text
| Name | Automatically created Entity ID |
|---|---|
| BatteryOpt Status | input_text.batteryopt_status |
| BatteryOpt Detail | input_text.batteryopt_status |
These helpers can be used in your dashboard to show the planned info on a MarkDown card.
Example MarkDown card:
type: markdown
title: Battery Charging Plan
content: |
{{ states('input_text.batteryopt_status') }}
{{ states('input_text.batteryopt_detail') }}Which looks like this:
Functions that are preceeded by @service are available to run manually from Developer Tools → Actions. If the functions are not there, there is an issue with starting the script, maybe it is in the wrong location or pyscript did not start. pyscript.simple_test, as part of debug_test.py, is a function which you can use for testing if everything works. pyscript.plan_night_charging_afternoon can also be run safely, it plans the nightly chage based on the available data but does not actually start charging. It does update the markdown card so it is a complete test.
If all is okay the charging plans are visible in the logfiles. These can be accessed on different ways but I prefer via the Terminal/SSH as that enables quick filtering. ha core logs | grep BatteryOpt gives the most recent plan but make sure to do this quickly after sending the command otherwise is will be gone again (so a bit tricky)
Example logfile entry:
