The size of high-rise buildings and the significant expenses associated to using repeatedly ad hoc excitation sources make appealing the adoption of natural phenomena as sources of the excitation even if their randomness makes quite challenging the assessment of the accuracy of the damage identification algorithms implemented to check the structural health. in this paper the sensitivity of the interpolation damage detection method (IDDM) to the randomness of the base excitation is studied with reference to the case of an high-rise wall-frame building lately designed to be built in Salerno (Italy). a detailed numerical model of the building was used to simulate several damage scenarios and a recently proposed numerical procedure was applied to calculate a large set of realistic base inputs, corresponding to low intensity earthquakes complying with the Italian code for the location of the building. structural responses to the entire set of input calculated by the numerical model were used to check the reliability of the damage identification algorithms. results show that the interpolation damage detection method allows to take into account the effect of the variability of the input and provides a reliable detection of a damaged location both in the case of single and of multiple damage scenarios in case of medium to severe damages. The investigation on multiple damaged locations pointed out also a relationship between the severity of damage, the global damage pattern and the values of the damage index that can influence the reliability of the method in case of multiple light damages.
