Rational Selection of Experimental Data for Inverse Structural Problems

Inverse analysis has been established as an effective tool for parameter identification of physical models in many fields of civil engineering. One of the main issues in inverse analysis is defining the well-posedness of the problem when a limited set of data is considered. In fact as shown in previous work, the location and the number of the sensors providing the experimental data greatly affect the accuracy of the inverse procedure. In this paper it will be shown that, under certain circumstances, it is possible to approximate the global field as a linear combination of the experimental data. This provides a rational basis for the choice of the experimental equipment by minimising the effect of the measurement error on the solution of the inverse problem. A numerical application regarding the estimation of the main parameters of an advanced mesoscale model for masonry structures highlights the practicality of this study.