PSI - Issue 44

Valentina Giglioni et al. / Procedia Structural Integrity 44 (2023) 1948–1955 Valentina Giglioni et al. / Structural Integrity Procedia 00 (2022) 000–000

1951

4

afterwards perform damage detection by computing the percentage of damaged short sequences within a macro sequence exceeding a pre-determined threshold $ ( ) .

Fig. 1. Flowchart of the methodology

Note that such a value should be evaluated on the basis of damaged sequences’ probability distribution during the training period.

4. Case study: post-earthquake assessment of the Z24 bridge 4.1. Description of the bridge

The Z24 benchmark bridge, dated back to 1963, was built to link the villages of Koppigen and Utzenstorf by passing over the main national highway A1 (Bern/Zurich). The structure was a post-tensioned RC bridge composed of a main span of 30 m and two side spans of 14 m. The cross-section of the bridge’s girder was characterized by two box cells for a global width of 8.6 m. The piers, described with a rectangular cross-section and a height of 6 m, were clamped into the girder and located and the end of the main span. Detailed information about the geometry and general characteristics of the structure are given by Masciotta et al. (2016). Before complete demolition, a long-term continuous monitoring took place throughout almost one year, from November 1997 to September 1998, by means of several sensors measuring both vibration and environmental data. In particular, accelerations were recorded on an hourly basis for about 10 minutes at a frequency of 100 Hz. In order to study the structural dynamic behavior during healthy and damage scenarios, a series of progressive destructive testing activities was carried out in August 1998. Further descriptions on the whole monitoring system and damage scenarios classes are provided by Peeters and De Roeck (2000) and Teughels and De Roeck (2004).