Method to compare autocorrelation analysis results obtained using SmartHUMS unit

Lee, E, Forsyth, G and Bil, C 2005, 'Method to compare autocorrelation analysis results obtained using SmartHUMS unit', in Proceedings of the 11th Australian International Aerospace Congress, Melbourne, Australia, 13-17 March 2005.


Document type: Conference Paper
Collection: Conference Papers

Title Method to compare autocorrelation analysis results obtained using SmartHUMS unit
Author(s) Lee, E
Forsyth, G
Bil, C
Year 2005
Conference name Australian International Aerospace Congress
Conference location Melbourne, Australia
Conference dates 13-17 March 2005
Proceedings title Proceedings of the 11th Australian International Aerospace Congress
Publisher Engineers Australia and Royal Aeronautical Society, Australia
Place of publication Melbourne, Australia
Abstract In the context of this paper the autocorrelation process is described as a type of Detection Index (DI). This DI is currently being proposed to be implemented in a miniature and low cost Health and Usage Monitoring System (HUMS) called SmartHUMS. DI is a specific example of Partial Information Extraction, where the information extraction algorithm, such as autocorrelation, has been used to achieve the principle of â¿¿moderate data, high informationâ¿¿. The primary application of autocorrelation analysis in this research is to detect differences between two processed autocorrelation results gathered from the same mechanical system. The degree of discrepancies between the two results will determine whether the monitored mechanical system has a behaviour change or not. In order to quantify the differences between two autocorrelation results, a simple and effective method is required. This paper presents a number of experimental results obtained by SmartHUMS. SmartHUMS is currently under development by the Defence Science and Technology Organisation (DSTO) in cooperation with GPS Online Pty Ltd. The experimental process started with SmartHUMS collecting data from a monitored test rig; once the data has been obtained it is then analysed by the autocorrelation process. During the experimental process artificial disturbances were introduced to demonstrate the ability of autocorrelation to detect these changes. This paper demonstrates a method that verifies the affiliation between two autocorrelation results, which then enables the indication of the overall behaviour of a monitored mechanical system.
Subjects Statistical Theory
Copyright notice © Engineers Australia and RAeS 2005
Versions
Version Filter Type
Access Statistics: 181 Abstract Views  -  Detailed Statistics
Created: Fri, 09 Oct 2009, 08:09:01 EST by Catalyst Administrator
© 2014 RMIT Research Repository • Powered by Fez SoftwareContact us