Statistical confidence bounds for structural health monitoring damage indicators

Detecting damage in a structure from changes in global dynamic parameters has received considerable attention from the civil, aerospace, and mechanical engineering communities. Although a number of promising damage detection techniques have been developed, very little work has been done on establishing statistical bounds for damage indicators. Statistical bounds are essential to establish how large a change is necessary to distinguish damage from the natural variability associated with any measurement. In this paper a False Positive Damage Indication Method (FPDIM) will be presented and applied to experimental data for a plate that was progressively damaged. The statistical bounds are determined using several sets of data from the undamaged structure. The method will be used with two non-model based damage detection methods.