Vibration-based post-Earthquake Rapid Damage Identification in historic masonry towers
ERIES-VERDI
Dataset Description
The ERIES-VERDI (Vibration-based post-Earthquake Rapid Damage Identification in historic masonry towers) project aims to assess the effectiveness and scalability of innovative vibration-based Structural Health Monitoring (SHM) methods for damage localization and quantification in earthquake-excited masonry constructions. The project proposed a dynamic investigation of the behaviour of masonry towers embedded within building aggregate, using shake table testing to validate SHM methods on a stone masonry tower subjected to different seismic inputs.
The dataset contains the results of the experimental campaign carried out at the LNEC facilities, in Lisbon. The masonry tower was built with calcareous stone and lime mortar, with boundary conditions reproducing the surrounding aggregates, at a 1:7 scale, featuring a total height of about 6.30 m, a square cross section of 1.0 x 1.0 m2, 0.2 m thick walls and small openings. The material was characterized by vertical compression tests on masonry wallets, three-point bending and compression tests on mortar specimens, and compression tests on stone specimens, all available at the dataset.
The tests on the scaled tower were carried out in two different configurations. In the first configuration, ambient vibration tests (AVT) were performed while it was standing on the floor of the testing hall. In the second configuration, shake table (ST) tests were performed where it underwent dynamic identification through ambient and forced vibration tests, followed by seismic tests with different input motions. The dataset includes recordings from a dense monitoring system collected during all shaking tests, as well as under white- and pink-noise excitations at different damage levels. It also provides detailed information on the damage state at each intensity level, enabling the development and validation of SHM algorithms for earthquake-induced damage.
Specimens
1. Masonry tower
2
In order to study the behaviour of slender masonry towers, a 1:7 scale model representative of this structural typology present in Italy, specifically inspired by the Sciri Tower in Perugia, was built at LNEC.
The geometry of the tower was defined after statistical analysis on historic towers and taking into account scaling laws. The cross section of the tower is square shaped with outer dimensions of 1.00 x 1.00 m2 and 0.20 m thickness, with a total height of 6.30 m. In addition, two side masonry walls were built to represent parts of adjacent buildings, providing boundary conditions similar to those of historic towers within an aggregate, and rising up to a height of 1.80 m.
The specimen walls were made of three leaves of rough shaped stones with average dimensions of 60-70 x 60-70 x 130-140 mm3, mortar joints of about 3-4 mm, and regular horizontal courses. Different layout patterns were combined and alternated throughout the stone layers, making a total of four layouts at the base level where the side walls were present and ten at the tower level, identified as BL# and TL#, respectively. Through stones were only provided at the edge zones or in a single layer each nine courses of stone, namely BL3, BL4, TL7 and TL8. Rectangular openings were distributed across all four façades, aligned at the same elevation on opposite sides and vertically offset between adjacent façades.
A good connection between the concrete slab foundation and the specimen was ensured by using a higher strength mortar in the first three courses. Moreover, interlocking between the side walls and the tower walls was provided through larger stones every two courses and the presence of an FRP grid.
The specimen was built with materials that represent the current conditions of historic masonry towers typical of the architectural heritage present across Europe, especially in the southern countries, namely using natural limestone blocks and a low-strength hydraulic-lime mortar.
The construction of the model lasted approximately 1.5 months, taking 13 days to reach up to the side walls total height and the remaining 28 days for the isolated tower above. On average, about 2.5 rows of the isolated tower cross section were built per day. Photos of the construction stages are provided below.
1. Ambient vibration tests
Ambient vibration tests were carried out before placing the tower on the shake table to characterise its dynamic behaviour, identifying the first and second modes of the structure, thus, enabling the selection of excitation frequencies for the shake table tests. The recorded signals were firstly pre-processed to remove filter initialisation artifacts (first three samples in each channel) and to remove artificial offset/trend (a high-pass second-order Butterworth forward-backward filter with cut-off of 0.1 Hz was applied).
Instrumentation
Seventeen measurement points were defined for the off-table ambient vibration test taking into account preliminary numerical modal analysis and the expected behaviour of towers. Three points were defined at the concrete slab, for control, and the other fourteen were distributed along the height of the tower in seven levels. At each level, two sensors were installed in one corner and a third sensor at the opposite corner, measuring the two directions in the horizontal plan. This distribution of sensors allowed capturing torsional vibration modes. Moreover, at the top level, an accelerometer in the vertical direction was also installed at each measurement point to better catch axial vibration modes coupled with the floor mode shapes. In total, 28 DOF’s were measured with this sensor configuration.
The instrumentation system consisted of uniaxial piezoelectric accelerometers connected to a multi-channel data acquisition system (DAQ) system by cables. Anti-aliasing filters were applied to the signals acquired with a frequency of 200 Hz and 10 min duration. For the ambient vibration tests before placing the tower on the shaking table, 28 uniaxial accelerometers with a nominal sensitivity of 10 V/g and a measurement range of ±0.5 g were used.
2. Shake-table tests
The shake-table tests comprised 97 steps in between seismic tests, dynamic identification tests, ambient vibration tests and other operations performed on the shake table. A set of two ground motion records with increasing intensities was applied to the shake table in both x and y directions. All data regarding the definition of the seismic input is shared inside the folder Seismic_input, shared online with DOI https://doi.org/10.60756/lnec-pq12.
The ground motions were generated to match a code-compliant design elastic response spectrum (ERS) defined according to the Italian Standard for the seismic hazard at the “Sciri” tower coordinates in Perugia. The target spectrum was calculated for the Damage Limit State, corresponding to a peak ground acceleration (PGA) of 0.094 g (50-year return period), using the Rexel software provided by ReLUIS. The target real scale ERS was interpolated for time history generation with 200 logarithmically spaced frequencies between 0.25Hz and 50Hz. Moreover, because the tested specimen was built at a 1:7 scale, the target ERS was scaled using the Cauchy similitude law, i.e. by dividing the period and multiplying the spectral accelerations by a factor of 7 but keeping the original sampling frequency. More details about the signal generation are presented in the file “ERIES-VERDI_signal_generation.pdf”, shared online with DOI https://doi.org/10.60756/lnec-pq12
Instrumentation
Nineteen measurement points were defined for the shake table tests taking into account preliminary numerical modal analysis and the expected behaviour of towers. Three points were defined at the concrete slab, for control, two were placed at each adjacent side wall’s mid-height, measuring the out-of-plane direction, and the other fourteen were distributed along the height of the tower in seven levels, as shown in the attached “ST_Instrumentation_plan.pdf” file. At each level, two sensors were installed in one corner and a third sensor at the opposite corner, measuring the two directions in the horizontal plan. This distribution of sensors allowed capturing torsional vibration modes. Moreover, at the top level, an accelerometer in the vertical direction was also installed at each measurement point to better catch axial vibration modes coupled with the floor mode shapes. In total, 30 DOF’s were measured with this sensor configuration.
The instrumentation system consisted of uniaxial piezoelectric accelerometers connected to a multi-channel data acquisition system (DAQ) system by cables. 30 uniaxial accelerometers with a nominal sensitivity of 1 V/g and a measurement range of ±5 g were used. Sensors with lower sensitivity were adopted for the ambient vibration tests on the shake table to keep the same instrumentation set-up between subsequent seismic tests during the following shaking table campaign, for which sensors with a higher dynamic range are required.
Project Metadata
Rights
Creative Commons Attribution 4.0 International.
CC BY 4.0
82 sessions
10 downloads
82 views
0 metadata
9 file previews
Feedback
We are always looking to improve the quality of our data and metadata. If you have any feedback or suggestions, please let us know.
Send Feedback