SAYISAL GÖRÜNTÜ İŞLEME İLE TİTREŞİM YER DEĞİŞTİRMELERİNİN ÖLÇÜMÜ VE MODAL PARAMETRE TAYİNİ

İnşaat mühendisliği yapılarında deneysel yöntemler ile elde edilen yer değiştirme gibi fiziksel büyüklükler ile modal parametreler yapının mevcut durumu hakkında önemli bilgiler içermektedir. Bu parametrelerin tayini genellikle ivme, hız ve yer değiştirme ölçerler ile toplanan verilerin zaman veya frekans alanında analizi ile yapılmaktadır. Fakat bu verilerin alışılagelmiş yöntemler ile elde edilmesi çoğu kez pratik olmamaktadır. Son yıllarda ise sayısal görüntü işleme teknikleri titreşim verilerinin çok daha ucuz ve kolay bir şekilde toplanmasına olanak tanımaktadır. Bu çalışmada, sayısal görüntü işleme teknikleri ile yapı titreşimlerinin ölçülmesi ve deneysel modal parametrelerin (doğal frekans, mod şekilleri v.b.) tayini üzerinde durulmuştur. Alüminyum alaşımdan inşa edilmiş dört katlı tek açıklıklı model bir yapı üzerinde üç yöntem test edilmiştir. Bu yöntemler sırasıyla Kanade-Lucas-Tomasi (KLT) algoritması, korelasyon tabanlı şablon eşleştirme ve görsel akış yöntemleridir. İlk iki yaklaşımda, titreşim kaynaklı yer değiştirmeler ölçülürken diğer yaklaşımda hız büyüklükleri ölçülebilmektedir. Ayrıca tüm yöntemlerde doğal hedeflerin takibi ile bu büyüklükler bulunabilmektedir. Bu çalışma kapsamında yapısından ivmeölçerler ve LVDT yardımıyla toplanan veriler ile karşılaştırma yapılarak değerlendirilmiştir

Anahtar Kelimeler:

Sayısal görüntü işleme, Operasyonel Modal Analiz, Görüntü tabanlı şablon eşletirme, Kanade-LucasTomasi algoritması, Görsel akış algoritması

COMPUTER VISION BASED VIBRATIONAL DISPLACEMENT MEASUREMENT AND MODAL IDENTIFICATION

Physical parameters such as displacements and modal parameters that are obtained from experimental tests contain crucial information about the current state of civil engineering structures. Identification of those parameters has been usually carried out by the time and the frequency domain analysis of the data obtained by means of accelerometers, velocity and displacement transducers. However, it is not always practical to obtain such experimental quantities via conventional methods. Recently, digital image processing techniques have made it Optical Flow Algorithmpossible to measure vibrational data in a way that is much cheaper and easier. In this study, computer vision based measurement of structural vibrations and identification of experimental modal parameters (natural frequency, mode shapes etc.) are emphasized. Three popular methods are introduced and tested on a 4-storey single-span laboratory structure, which is made of aluminum alloy. The investigated methods are KanadeLucas-Tomasi (KLT) algorithm, a correlation-based template matching method and an optical flow method, respectively. While the first two methods are used to measure the vibrational displacements, the last approach is suitable for velocity field measurements. In addition, all methods are capable of capturing vibrational quantities by tracking natural targets located on the structure. In this study, the presented methods are discussed and their performances are evaluated by comparing the results with conventional accelerometers and LVDT measurements

Keywords:

Computer vision processing, Operational modal analysis, Vision based template matching, Kanade-LucasTomasi algorithm,

PDF

___

Mukhopadhyay, S.C., ed. 2011. New Developments technology for structural health monitoring, Lecture Notes in
Electrical Engineering, Springer, Berlin. sensing Brincker R, Andersen P, Jacobsen NJ., 2007. Automated frequency domain operational
Proceedings of IMAC-XXIV: A Conference and Exposition on Structural Dynamics. Society for Experimental Mechanics, Florida, USA. for analysis. Van Overschee P., De Moor B. 1996.
Subspace identification for linear systems: Theory - Implementation Applications. Kluwer Academic Publishers, Netherlands The Rainieri, C. ve Fabbrocino,G. 2014.
Operational Modal Analysis of Civil Engineering Introduction Applications, Springer, Berlin. An for Guide Catbas, N.F., Correa,K.T ve Aktan,A.E. Identification Systems Approaches, Methods, and Technologies for Effective Practice of St-Id, American Society of Civil Engineers, ASCE,USA. Structural Constructed Feng, M.Q., Fukuda, Y., Feng, D. and Mizuta, M. 2015 Nontarget Vision
Sensor for Remote Measurement of Bridge Dynamic Response, Journal of Bridge Engineering, Cilt. 20(12), s. 1-11. Khuc, T. and Catbas, F.N. 2017.
Completely contactless structural health monitoring of real-life structures using cameras and computer vision, Structural Control and Health Monitoring,Cilt. 24(1), s. e1852, DOI: 10.1002/stc.1852
Feng, D., Feng, M. Q., Ozer E. ve Fukuda Y. 2015. A Vision-Based
Sensor for Noncontact Structural Displacement Sensors, Cilt. 15, s. 16557-16575. doi:10.3390/s150716557
Dworakowski Z., Kohut, P., Gallina, A., Holak K. ve Uhl, T. (2016) Vision-based damage detection and localization in monitoring,Structural Control and Health Monitoring, Cilt. 23, s. 35– 10.1002/stc.1755 for structural health
Yoon, H., Elanwar, H., Choi, H., Fard,M.G., Spencer, B.F. 2016.
Target-free approach for vision- based identification using consumer- grade cameras, Structural Control and Health Monitoring, Cilt. 23, s. –1416. DOI: 10.1002/stc.1850 Oh, B.K., Hwang, J.W., Kim,Y.,
Cho,T., Park, H.S. 2015. Vision- based technique for building structures using a motion capture system,
Journal of Sound and Vibration, Cilt. http://dx.doi.org/10.1016/j.jsv.20 07.011 –85. Feng, D., Feng, M.Q. 2017.
Experimental validation of cost- effective vision-based structural health monitoring, Mechanical Systems and Signal Processing, Cilt. , http://dx.doi.org/10.1016/j.ymssp 2016.11.021 –211.
Kohut, P., Holak,K., Uhl,T., Ortyl, L., Owerko,T., Kuras, P. And Kocierz, R. 2013. Monitoring of a civil structure’s noncontact
Structural Health Monitoring Cilt. (5-6), 1177/1475921713487397
Kim, S.W., Kim N.S. 2013. Dynamic characteristics bridge hanger cables using digital image
International, Cilt. 59, s. 25–33. http://dx.doi.org/10.1016/j.ndtein t.2013.05.002 suspension processing, NDT&E
Ye, X.W., Dong, C. Z., Liu, T. 2016. A
Review of Machine Vision-Based Structural Methodologies and Applications, Journal of Sensors, 7103039, 10 pages.http://dx.doi.org/10.1155/2 /7103039 Monitoring: Shi, J., Tomasi.C. 1994. Good features to track. IEEE Conference on Computer Vision and Pattern
Recognition (CVPR'94), 593 – 600. Fredriksson, K., Ukkonen, E. 2001
Faster template matching without FFT. In: Proc. IEEE Intl. Conf. on Image Processing (ICIP). Volume 1, –681. Barron, J.L., Fleet, D.J., Beauchemin, S.S. Performance of optical flow techniques. International Journal of Computer Vision, Cilt. 12, s. 43- DOI: 10.1007/BF01420984
Ullah,F. Kaneko,S 2004. Using orientation codes for rotation- invariant
Pattern Recognition Cilt.37, s. 201 – (03)00184-5 matching, doi:10.1016/S0031
Ji,Y.F., Chang,C.C. 2008. Nontarget image-based technique for small vibration cable
Journal of Bridge Engineering Cilt.13, doi:10.1061/_ASCE_10840702_2008_ :1_34_ http://avesis.yildiz.edu.tr/serhate/ dokumanlar
Ewnis,D.J. 2000. Modal Testing: theory, practice and application, nd edition, Research Studies Press, England