Residual learning

Modelling the gap between what physics predicts and what is measured.

A first-principles model predicts what a plant should be doing under the current conditions. The residual is the difference between that prediction and the measurement. Residual learning applies data-driven methods to the residual rather than to the raw signal.

The advantage is that most of the variation caused by ordinary operation has already been removed by the physics. Load changes, ambient conditions and setpoint moves are explained rather than treated as anomalies, so what remains is closer to genuine unexplained behaviour.

This is one of the reasons a hybrid approach tends to produce fewer false alarms than a purely statistical one on assets that vary constantly by design.