17. Robust Sensor and Exciter Design for Linear Structures

A wide variety of model-based modal test design methodologies have been developed over the past two decades using a non-validated baseline model of the structure of interest. Due to the presence of lack of knowledge, this process can lead to less than optimal distributions of sensors and exciters due to the discrepancy between the model and the prototype behaviors. More recent strategies take into account statistical variability in model parameters but the results depend strongly on the hypothesized distributions. This paper provides a decision making tool using a robust satisficing approach that provides a better understanding of the trade-off between the performance of the test design and its robustness to model form errors and associated imprecisions. The latter will be represented as an info-gap model and the proposed methodology seeks a sensor and exciter distribution that will satisfy a given design performance while tolerating a specified degree of modeling error. The evolution of this performance for increasing horizons of uncertainty is an important information for the test planner in choosing the total number of sensors. The methodology will be illustrated on an academic but practically useful example under severe uncertainty.