Gross Error Detection and Variable Classification in Dynamic Systems

Download or read book Gross Error Detection and Variable Classification in Dynamic Systems written by Joao S. Albuquerque and published by . This book was released on 1995 with total page 27 pages. Available in PDF, EPUB and Kindle. Book excerpt: Abstract: "Gross error detection plays a vital role in parameter estimation and data reconciliation. Data errors due to either miscalibrated or faulty sensors or just random events nonrepresentative of the underlying statistical distribution, can induce heavy biases in the parameter estimates and in the reconciled data. In this paper we concentrate on robust estimators and exploratory statistical methods which allow us to detect the gross errors as the data reconciliation is performed. These robust methods have the property of being insensitive to departures from ideal statistical distributions and therefore are insensitive to the presence of outliers. Once the regression is done, the outliers can be detected by using exploratory statistical techniques. Also important is the ability to classify the variables according to their observability and redundancy properties. Here an observable variable is an unmeasured quantity which can be estimated from the measured variables through the physical model while a nonredundant variable is a measured variable which cannot be estimated other than through its measurements. Variable classification can be used as an aid to design instrumentation schemes. Both variable classification and gross error detection relate to each other in the sense that gross error detection can only be performed on redundant measurements. In this paper we develop an efficient method for this classification for dynamic systems."