Research Article

Subspace Method Aided Data-Driven Fault Detection Based on Principal Component Analysis

Algorithm 1

SMI-PCA fault detection.
Generate and with fault-free data
Perform PCA on
Calculate and according to (22)
Construct residual according to (12)
Calculate in the fault-free case
Generate and with on-line operating data
Construct residual according to (12), then calculate
ā€ƒā€‰ the testing statistic by (13)
Determine the threshold
Evaluate the residual according to the detection logic