battery_forecast¶

Discharge rate and runtime forecast from Home Assistant’s history.

uv run tools/battery_forecast.py [–since 2026-09-27T08:12] [–hours 72]

Takes the battery level recorded since the board last stopped charging (sensor.epaper_ladezustand), or since –since (local time), fits a line through it and projects when the levels that change the update pace are reached.

Attributes¶

CHARGE_STATE

Functions¶

`parse_time`(text)	Parse an ISO time from Home Assistant, including a "Z" suffix.
`discharge_start`(states)	Find when the board last stopped charging.
`fit`(points)	Least-squares line through [(hours, percent)]: (slope %/h, intercept).
`forecast`(levels, start, now)	Rate and projections from [(time, percent)] since `start`.
`history`(entity_ids, since, until)	Load the state history of `entity_ids` from Home Assistant.
`main`()	Run the command line tool.

Module Contents¶

battery_forecast.CHARGE_STATE = 'sensor.epaper_ladezustand'¶

battery_forecast.parse_time(text)¶: Parse an ISO time from Home Assistant, including a “Z” suffix.

battery_forecast.discharge_start(states)¶

Find when the board last stopped charging.

From [(time, state)] this is the first report after the last “charging”, or the first report at all if none was charging; None if it is charging now. (On battery the board mostly reports “stable”: the discharge in the deep-sleep cycle is below the gauge’s resolution.)

battery_forecast.fit(points)¶: Least-squares line through [(hours, percent)]: (slope %/h, intercept).

battery_forecast.forecast(levels, start, now)¶: Rate and projections from [(time, percent)] since start.

battery_forecast.history(entity_ids, since, until)¶: Load the state history of entity_ids from Home Assistant.

battery_forecast.main()¶: Run the command line tool.