Development of a statistical method for validating ESDM strain-field output
Doktor, Christopher A.
MetadataShow full item record
This thesis addresses the problem of developing a statistical method for validating the ESDM strain-field output. Specifically, it addresses the problems of estimating the ESDM strain-field uncertainty, choosing an appropriate standard measurement, estimating the uncertainty in the standard measurement, and comparing the ESDM strain-fields against a finite set of standard measurements. A solution to each of these problems is presented. Specifically, it is shown that 1) the uncertainty in the strain-field may be estimated from the ESDM dynamic stiffness matrix; 2) strain gages are chosen to validate the ESDM strain-fields; 3) the uncertainty of the strain gage measurements may be estimated from the uncertainty in the system calibration, the output voltage of the system and the tolerance of the strain gage gage factor; and 4) the ESDM strain-fields may be validated against a set of strain gage measurements by a combining a 2-group z-test (hypothesis test on two means with known variances) with a qualitative analysis of the field. This statistical validation method is then evaluated in test-based simulation to determine if it provides reliable assessments of the ESDM strain-field accuracy. The results of the simulation show that the validation method does, indeed, provide reliable assessment of the ESDM strain-field accuracy. The results also show that the ESDM method should be capable of accurately estimating dynamic strain-fields, provided it is not subjected to substantial error in the calculated position and orientation of the laser or nonstationary behavior in the structure.
- Masters Theses