Data mining and visualization for anomaly detection and diagnosis in civil structures

James M W Brownjohn, Piotr Omenzetter, Pilate Moyo

Research output: Contribution to conferenceUnpublished paperpeer-review

61 Downloads (Pure)


Management of data generated by SHM systems is a major issue to be addressed in future developments. Even with data compression and embedded systems to convert large quantities of data to more manageable amounts of information, there remains the need for procedures to manage the data and in particular to present it to various levels of user. Experience with a number of SHM systems has shown the need to condense data, to develop simple interfaces for quick visual inspection, to provide second and third levels of inspection via statistical analysis tools to identify performance anomalies, more sophisticated parametric modelling and data mining techniques to characterise the anomalies and links to validated structural models for diagnosis. The paper presents experiences with a combination of dynamics-based structural assessment and continuous remote monitoring of static and dynamic effects and response and some of the tools that have been developed to manage and interpret the data.
Original languageEnglish
Number of pages6
Publication statusPublished - 31 Jan 2005


Dive into the research topics of 'Data mining and visualization for anomaly detection and diagnosis in civil structures'. Together they form a unique fingerprint.

Cite this