Keywords: damage identification, soil-structure interaction, multilevel approach, nonlinear analysis.

Predicting building damage arising from ground movements is an important aspect for buildings located in an urban environment. Possible approaches range from empirical estimates to detailed finite element calculations. In the first part of the paper, the main existing methods for predicting damage to buildings are briefly presented, and their potential and limitations discussed. The methods can be classified as strictly empirical, semi-empirical or detailed. The empirical methods include recommendations on the allowable settlement of structures. Early semi-empirical methods consider the building as an equivalent beam and identify the onset of damage by a threshold value of the ratio between the deflection of the equivalent beam and its length being exceeded. Another semi-empirical approach models the building as a laminated beam. It is still a simplified method, but it permits more realistic evaluations to be made. The finite element analyses allow an accurate assessment of the structural behaviour, by taking into account the structural properties; however, they require large amounts of time and computational effort and, despite the complexity of the modelling, they are still affected by the uncertainty of the soil-structure interaction.

In order to evaluate the adequacy of the various models, the methods are applied to two case studies. In the first one, two semi-empirical methods are compared: the equivalent beam model and the laminated beam model. The comparison shows that the laminated beam model allows more characteristics of the structure to be taken into account and suggests the presence of damage in some situations that are not considered by the equivalent beam model.

In the second case study, the response of an existing reinforced concrete frame structure damaged by differential settlements of the foundations is analyzed considering both the laminated beam model and the three-dimensional finite element approach. The laminated beam does not require a large amount of computational time and effort, and is able to identify the onset of cracking; however it cannot give information about the redistribution of the loads and the stress state in the structural elements. On the other hand, by using the finite element approach, even if a very demanding model is needed, it is possible to understand also the effects of the interaction that can lead the structural members to unsafe conditions.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £130 +P&P)