An improved PCA algorithm for anomaly detection of hydropower units
Author:
- 1 April 2017
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 494-498
Abstract
This paper proposes an improved principal component analysis (PCA) algorithm for anomaly detection of hydropower units (HUs). Operation conditions of HUs are identified first. Then PCA model is updated by two adaptive updating methods under different operation conditions. And in steady operation conditions, the proper window size, an important parameter, is obtained by the estimation of model stability. The improved method is applied to detect a simulated constant deviation anomaly of the swing measurement sensor in the experiment part. The result shows this improved method has a higher precision rate of detection and a satisfactory detection rate of anomaly compared with the traditional method.Keywords
This publication has 4 references indexed in Scilit:
- A novel KICA–PCA fault detection model for condition process of hydroelectric generating unitMeasurement, 2014
- Recursive PCA for adaptive process monitoringJournal of Process Control, 2000
- The process chemometrics approach to process monitoring and fault detectionJournal of Process Control, 1996
- Statistical process control of multivariate processesControl Engineering Practice, 1995