Which are the best techniques to detect damages in civil structures particularly in Bridges?
Whether model based or non model based techniques work better?
Like non-model based we have statistical ar feature extraction and malnobis distance based ,PCA methods.
And in model based methods we have model updating based damaged detection methods.
We also have strain energy methods and many more like Hilbert–Huang transform methods.
So out of all these techniques which are most robust and accurate methods.Please suggest if any other good methods you know.
Hi Suman,
I employ neural networks to detect stiffness degradation in structures. I attach a paper with the technique that we're using. I hope this helps.
Luiza
As far as i know, there is no any method to really detect damage in real scale structures. All methods i have seen are just for laboratory scale models. Once you go out in real world, they cannot be reliably used. This issue still is in research phase.
I have used wavelet analysis for damage detection of nonlinear structures employing restoring force.
https://www.researchgate.net/publication/267482608_Wavelet-Based_Method_for_Damage_Detection_of_Nonlinear_Structures
Article Wavelet-Based Method for Damage Detection of Nonlinear Structures
We test structures in real scale and there is no damage detection method which would cover it all. It depends on what you want to detect - not all damage is equal... So, you should elaborate more on what your aim is.
My aim is to detect and localize different kind of faults in railway bridges(open web girder and plate girder steel) where train moves at a speed of 45-140 km/hr.
We have accelerometer and strain gauges as sensor.
Then what are the best techniques to make a shm system for these kind of openweb girder and plate girder steel bridges
I think the NDE methods (Schmidt hammer test - ultrasonic wave) and semi NDE tests (Core test) may be effective techniques to detect damages in a structure.
I have heard about the identification and location of destructive processes in reinforced concrete structures using the method based on the measurements of acoustic emissions generated by those processes. Maybe it will be usefull for You. Some related papers:
http://www.sciencedirect.com/science/article/pii/S1644966512000052
http://trid.trb.org/view.aspx?id=1247344
http://www.ndt.net/article/jae/papers/20-083.pdf
Regards, WojtekK
Hello,
first of all if you would like to analyze something like bridges, etc. you need many sensors, e.g. Piezo sensors. Usually it is not enough to use just a few. The procedure which is used e.g. in the above mentioned application or in airplanes also today is based on several 10 to some hundred of sensors and a Sensor Fusion method which works quite nicely. We have applied a special version of Hilbert-Huang-Transform which is based on empirical mode decomposition, analyzed the empirical mode functions, designed as features some statistical parameters from histograms and applied a fuzzy-pattern-classifier.
I have to say that also other methods will work as long you are able to describe your object behavior with a proper model and you have to understand how the features behave in certain error/flaw cases (important !!) physically. Just trying to analyze data does not help!
A company which is designing excellent sensors for the application you have in mind is: http://www.invent-gmbh.de/en/start/3e3cb996718aa6aa891cd835d0bb750a/
Maybe it helps....
To identify the damage Electromechanical Impedance (EMI) technique can be used. Piezo sensors are used in this technique. In the beginning one can get the admittance signatures in the pristine state. Again the signatures are taken after the desired time. Any alternation in admittance signatures with reference to the admittance signatures in pristine state indicate that the damage has occurred in the structure.
Dear Suman Maroju,
in order to detect damage on RC structures we are using the modal curvature variation (evaluated before, during and after an earthquake). The absolute value of the curvature variation is strictly related to the maximum interstory drift. Then, using this kind of approach, it is possible to quantify and localize the damge occurred on a framed structure. It is my pleasure, to send you some of my recent papers about this topic. For any question, do not esitate to contact me.
Sincerely,
Rocco
Article Analysis of non-stationary structural systems by using a ban...
Article Localization of Damage Occurred on Framed Structures: Analys...
Article Automatic evaluation of the fundamental frequency variations...
Article A fast method for structural health monitoring of Italian re...
Wavelet based methods can be used for damage detection
The effectiveness of vibration-based methods in damage detection of a typical highway structure is investigated. Two types of full-scale concrete structures subjected to fatigue loads are studied: (1) Portland cement concrete pavements on grade; and (2) a simply supported prestressed concrete beams. Fast Fourier transform (FFT) and continuous wavelet transform (CWT) are used in the analysis of the structures’ dynamic response to impact, and results from both techniques are compared. Both FFT and CWT can identify which frequency components exist in a signal. However, only the wavelet transform can show when a particular frequency occurs. Results of this research are such that FFT can detect the progression of damage in the beam but not in the slab. In contrast, the CWT analysis yielded a clear difference between the initial and damaged states for both structures. These findings confirm the conclusions of previous studies conducted on small-scale specimens that wavelet analysis has a great potential in the damage detection of concrete. The study also demonstrates that the approach is applicable to full-scale components of sizes similar or close to actual in-service structures.
Read More: http://ascelibrary.org/doi/abs/10.1061/(ASCE)0733-9399(2003)129:5(571)
http://ascelibrary.org/doi/abs/10.1061/(ASCE)0733-9399(2003)129:5(571)
Another method is statistical process control proposed by scientists from
Los Alamos National Laboratory
Currently, vibration-based damage detection is an area of significant research activity. This paper attempts to extend the research in this field through the application of statistical analysis procedures to the vibration-based damage detection problem. The damage detection process is cast in the context of a statistical pattern recognition paradigm. In particular, this paper focuses on applying statistical process control methods referred to as ‘control charts’ to vibration-based damage detection. First, an autoregressive (AR) model is fit to the measured acceleration–time histories from an undamaged structure. Residual errors, which quantify the difference between the prediction from the AR model and the actual measured time history at each time interval, are used as the damage-sensitive features. Next, the X-bar and S control charts are employed to monitor the mean and variance of the selected features. Control limits for the control charts are constructed based on the features obtained from the initial intact structure. The residual errors computed from the previous AR model and subsequent new data are then monitored relative to the control limits. A statistically significant number of error terms outside the control limits indicate a system transit from a healthy state to a damage state. For demonstration, this statistical process control is applied to vibration test data acquired from a concrete bridge column as the column is progressively damaged.
http://institutes.lanl.gov/ei/shm/pubs/MSSP_00_1323.pdf
- Badges
- Science topic
- Similar topics
- Computer Communications (Networks)
- Simulators