Title :
System Status Feature Extraction Based on Recursive Lattice and Block Linear Predictive Coding Algorithm
Author :
Liu, Wei ; Doraiswami, R.
Author_Institution :
Student Member, IEEE, Department of Electrical Engineering, University of New Brunswick, Fredericton, NB, Canada
Abstract :
The control system status features are extracted by using fast recursive least-squares lattice (RLSL) and block linear predictive coding algorithm (LPCA), respectively. A short time record of control system signal is captured and the future trend of the system is predicted by analyzing the variations of the estimated auto regressive and moving average (ARMA) model parameters. The comparison of two methods applied in system performance monitoring with real-time data cases are given and their performance are evaluated.
Keywords :
Control system synthesis; Control systems; Feature extraction; Lattices; Linear predictive coding; Monitoring; Prediction algorithms; Predictive models; Signal analysis; System performance;
Conference_Titel :
American Control Conference, 1992
Conference_Location :
Chicago, IL, USA
Print_ISBN :
0-7803-0210-9
