In-flight structural identification:input/output versus output-only data processing

The problem of in-flight data analysis, for the purpose of structural model identification under both measured and uncontrolled non-stationary excitation, is addressed. Input/output and output-only eigenstructure identification methods are described and compared, within two classes of methods: subspace-based and prediction error. In particular, different types of relevant projections for handling the known (measured) and unknown (uncontrolled) inputs are discussed. The relevance of the methods is emphasized through numerical results obtained on flight test data sets.