A pattern matching identification method of cracks on cantilever beams through their bending modes measured by magnetoelastic sensors. (October 2019)

• Record Type: Journal Article
• Authors: Samourgkanidis, Georgios; Kouzoudis, Dimitris
• Is Part Of: Theoretical and applied fracture mechanics. Volume 103(2019)
• Publication Date: 2019-10
• Subjects: Crack detection -- Magnetoelastic sensors -- Bending modes -- Metglas -- Damage detection -- Modal analysis; Fracture mechanics -- Periodicals; 620.1126
• DOI: 10.1016/j.tafmec.2019.102266
• Languages: English
• ISSNs: 0167-8442

Abstract

Abstract: This work introduces a simple method on identification of single transverse cracks on cantilever beams, having as an input the first 8 bending modes of the beam. The beam modeling part of the method was developed based on a finite element analysis and fracture mechanics theory, and implemented using Matlab programming. Bending mode normalized frequencies were extracted from the model, versus crack location (CL) and crack depth (CD), and all possible normalized frequency ratios (NFRs) were calculated in order to be stored in a database. This database information was used to find the optimum CL and CD values from input data, through a pattern matching process. Two different approaches were presented to validate the method, one experimental and one numerical. In the case of the experimental approach a number of 19 beam specimens of aluminum alloy 6063 were used, with various fixed crack locations and depths, and the extracted CL and CD errors were 1.7 % and 11.4 %, respectively. The measurement of the 8 bending modes was accomplished non-invasively by the use of magnetoelastic vibration sensors, which were composed of 29 μ m thin ribbons of magnetoelastic material Metglas 2826MB3. On the other hand, the numerical approach was performed using ANSYS 2016 Workbench software on beam designs with the same physical and geometrical characteristics as the beam specimens used in the experiment. The extracted CL and CD errors are found to be 0.7 % and 8.5 %, respectively. The above errors show that the proposed method is extremely capable of predicting the crack location and quite capable of predicting the crack depth. This work sets the basic prospects for the design of a new Structural Health Monitoring (SHM) technique combining the efficiency of the magnetoelastic sensors together with the proposed method.