Toogle Left

BATir banner

 

logo SMC logo SMC  DYNA logo SMC  DYNA  SHM Damage localization in civil engineering structures using dynamic strain measurements
   
Contact

Arnaud Deraemaeker, Gilles Tondreau

   
Keywords Modal filters, damage localization, Structural Health Monitoring, control charts, ambient vibrations
   
Current Collaborations

Prof G. De Roeck and Prof G. Lombaert, KULeuven, Belgium

Prof K. Worden, University of Sheffield, UK

Dr Ch. Farrar, Los Alamos National Labs, United States

   

Past Collaborations

Prof J. Kullaa, Helsinki University of Technology, Finland

Dr L. Mevel and Prof. M. Basseville, INRIA/IRISA

Dr B. Peeters, LMS International, Belgium

   

Structural health

monitoring

The current practice to monitor civil engineering structures consists in scheduled maintenance (visual inspection, radiography, ultrasounds,...), which requires that the vicinity of the damage is known, but are also expensive, tedious, and may require to stop the traffic. An alternative called Structural Health Monitoring (SHM) which has been widely studied during the last decades uses the ambient vibrations in order to diagnose damages continuously in (near) real time. The methods developed so far focus mainly on the damage detection of severe damages. The real interest of SHM relies however in the possibility to locate (and possibly quantify) damages at an early state, in order to allow intervention in time if necessary. This research topic aims at locating small damages in civil engineering structures using ambient vibrations.

 

bridges600

Figure 1 : Catastrophic consequences of non continuous monitoring of the structures.

     
Dynamic strains

Pandey et al. have shown in [1] that the curvatures are locally sensitive to damage, and this observation has been extended to any kind of structures with the strains in [2]. Thanks to recent advances in the field of instrumentation, it is now possible to equip structures with large network of dynamic strain sensors (fiber optics, strain gages, piezoelectric patches,...). The use of dynamic strain sensors is therefore a very promising solution to locate damages in structures. In particular, cheap PVDFs piezoelectric sensors have shown interesting results in experiments led in our research group.

Deformations

Figure 2: Effect of damage on low order strain mode shapes. 

   
Modal filters

A very simple method consisting in linear combination of the sensor responses of accelerometers has been first developed for damage detection [3], and then recently extended by processing locally the strain measurements from small groups separately [4]. This non-model based procedure enhances the visibility of very small damages (2% of stiffness reduction) and shows a nice robustness with respect to environmental changes (changes of excitation, effect of temperature, noise measurement). The method is applied in time domain, and the classical Fourier analysis allows to extract the corresponding frequency response, in which the damage will be responsible of the appearance of a spurious peak

 linear combinerLoc

Figure 3 : Modal filters and their use to locate damages.

     

Automated damage localization

The automation of the damage localization is essential for a successful application of the method in real applications. Indeed, avoiding false alarms (alarms triggered when there is no damage) and missing alarms (alarms missed when damage occurs) will make the technique reliable and particular attention should be paid to damage sensitive features as well as statistical analysis in order to encompass the variability due to environmental changes. The definition of adequate features sensitive to the appearance of damages, as well as the use control charts [5] gave excellent results, both numerically and experimentally. The procedure is fully automated, since after a first design step of the modal filters and of the control charts, the process does not need any user interaction, and fires alarms automatically when a damage is located.

 

PiezoSetUp

Figure 4: Automated damage localization using control charts.

   
Support

2009-2013: FRIA Grant : Damage localization in civil engineering structures using dynamic strain measurements

     
Selected publications

[1] A. K. Pandey, M. Biswas, and M. M. Samman. Damage detection from changes in curvature mode shapes. Journal of Sound and Vibration, 145:321–332, 1991.

[2] A. Deraemaeker. On the use of dynamic strains and curvatures for vibration based damage localization. In The Fifth European Workshop on Structural Health Monitoring, Sorrento, Italy, July 2010.

[3] A. Deraemaeker and A. Preumont. Vibration based damage detection using large array sensors and spatial filters. Mechanical Systems and Signal Processing, 20:1615–1630, 2006

[4] G. Tondreau, A. Deraemaeker, Local modal filters for automated data-based damage localization using ambient vibrations, Submitted to Mechanical Systems and Signal Processing.

[5] Douglas C. Montgomery. Statistical quality control: a modern introduction. John Wiley and Sons, New York, 2009.