Inscriere cercetatori

Site nou !

Daca nu va puteti recupera parola (sau aveti alte probleme), scrieti-ne la pagina de contact. Situl vechi se gaseste la adresa


Damage-patterns-based method to locate discontinuities in beams

Domenii publicaţii > Stiinte ingineresti + Tipuri publicaţii > Articol în volumul unei conferinţe

Autori: Gilbert-Rainer Gillich; Zeno-Iosif Praisach

Editorial: Tribikram Kundu, Proc. SPIE 8695, Health Monitoring of Structural and Biological Systems 2013, 8695, 2013.


The paper presents a method to locate discontinuities in form of transversal cracks in beams, based on vibration measurements. Patterns characterizing frequency changes of the first ten weak-axis bending vibration modes are determined for all possible locations on the structure, using a relation contrived by the authors. It base on the correlation between the strain energy stored in a segment of the beam, which is proportional with the square of the mode shape curvature of the considered vibration mode at that location, and the frequency change for this mode by the occurrence of a discontinuity on that segment. The patters consist from a series of ten values representing the normalized relative frequency shifts for the first ten vibration modes. For a structure similarly supported, by continuous or periodical measurements, potential frequency changes can be detected. By processing these data the so-called damage location index for that crack is found out, also as a series of ten values representing the relative frequency shifts of the ten vibration modes. To precise locate the crack a pattern matching method involving the database with all possible patterns and the damage location index is used. Knowing the location, it is easy to determine by analytic calculus the crack depth. The method is easy to be used, provide accurate results, demands modest computational effort and has the advantage that the measurements may be carried out in situ with rather simple equipment. The method was validated by experiments.

Cuvinte cheie: damage detection, Euler-Bernoulli beam, pattern recognition