Two-phase drift monitoring and automatic retraining (TPDM-AR) for electric load forecasting
Michel B. Oliveira, Leandro A. Silva · 2026
Read Two-phase drift monitoring and automatic retraining (TPDM-AR) for electric load forecasting by Michel B. Oliveira, Leandro A. Silva (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13226265.txt.
Description
Michel Oliveira and Leandro Silva propose a two-phase system for spotting drift in power-use forecasts, in a 2026 paper. Smart buildings rely on forecasts of electric load, yet habits and seasons shift over time and stale models lose accuracy. Their workflow watches the raw power stream first, then checks whether forecast errors are truly getting worse.
Three change detectors scan the incoming data for early warnings. A second phase tracks smoothed forecast error against a baseline and demands steady, agreed proof before acting. Only then does the system retrain its neural network forecaster. Tests used both made-up drift patterns and a real household power dataset. The design aims to retrain at the right moment, not at every small wobble in the data.
Catalog record
| CREATOR | Michel B. Oliveira, Leandro A. Silva |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 73 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |