Empirical models for the health monitoring of high-rise buildings: the case of Palazzo Lombardia

This paper deals with the monitoring of Palazzo Lombardia, one of the tallest skyscrapers in the city of Milan. The monitoring system collects both dynamic and static data in order to assess the health condition of the whole structure. Accelerometers are used for collecting dynamic data, while static data are obtained from clinometers. Moreover, environmental conditions are acquired by employing different further transducers. The aim of this paper is to develop an empirical model to predict the values of the first eigenfrequencies of the structure as a function of the environmental conditions. Such a model can be then used to assess the health condition of the building. Indeed, deviations between the values expected from the model and those identified by means of an operational modal analysis of the acceleration data may indicate the occurrence of a structural change. However, it is not straightforward to build the mentioned empirical model because of some difficulties in managing the environmental data. As an example, it is hard to understand which temperature must be considered since it changes significantly along the structure. The paper shows how to overcome these problems, avoiding the use of the environmental data and employing, in place of them, the data coming from the clinometers and accelerometers. This approach allows to build the mentioned model successfully. Different empirical models are compared in the paper, evidencing which one is able to describe the trends of the eigenfrequencies at best. Furthermore, the uncertainty associated to the predictions of the model is also discussed. The whole work has been carried out on data acquired by the monitoring system for a time duration of almost 2 years.