DAMAGE IDENTIFICATION THROUGH THE USE OF HIGHORDER STATISTICS

Structural Health Monitoring is based on the development of reliable and robust indicators capable to detect, locate, quantify and predict damage. Studies related to damage detection in civil engineering structures have a noticeable interest for researchers in this area. Indeed, the detection of structural changes likely to become critical can avoid the occurrence of major dysfunctions associated with social, economic and environmental consequences. Recently, many researchers have focused on dynamic assessment as part of structural diagnosis. Most of the studied techniques are based on time or frequency domain analyses to extract compressed information from modal characteristics or based on indicators built from these parameters. This work has as its main interest the use of highorder statistics (HOS) coupled with clustering techniques i.e. the k-means algorithm to detect structural modification (damage). The approach is applied directly to dynamic measurements (accelerations) obtained on site. In order to attest the efficiency of the proposed methodology,two investigations are carried out: a numerical model of a simply supported beam and a real case railway bridge, in France. It is shown that HOS coupled with clustering methods is able to distinguish structural conditions with adequate rates