What's Changed
ML Load Prediction
The ML Load Prediction component uses a lightweight multi-layer perceptron (MLP) neural network implemented in pure NumPy. It learns from your historical energy consumption patterns and makes predictions about future load.
Key Features:
- Predicts 48 hours of load data in 5-minute intervals
- Learns daily and weekly patterns automatically
- Supports historical PV generation data as an input feature
- Supports temperature forecast data for improved accuracy
- Automatically trains on historical data (requires at least 1 day, recommended 7+ days)
- Fine-tunes periodically to adapt to changing patterns
- Model persists across restarts
- Falls back gracefully if predictions are unreliable
ML Load predictor
- ML Load predictor, external temperature components by @springfall2008 in #3310
- Enhance documentation with LoadML charts details by @springfall2008 in #3334
- Copyright date, LoadML chart fixes by @springfall2008 in #3335
- Temperature, catch bad latitude/longitude by @springfall2008 in #3339
Improvements
- Web UI improvements - new charts, more responsive status
- Axle update sensor every minute by @springfall2008 in #3333
Bug fixes
- Ensure solar data updates at midnight by @springfall2008 in #3331
- Fix partial IOG slots not marking entire 30-min periods as off-peak by @springfall2008 in #3332
- Fix bug with DB enable false by @springfall2008 in #3346
Full Changelog: v8.32.14...v8.33.0