Using a wavelet network for reconstruction of fatigue crack depth profile from AC field measurement signals

Ravan, Maryam; Sadeghi, S.H.H.; Moini, R.

doi:10.1016/j.ndteint.2007.04.001

Cited by 13 publications

(10 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To address the problem of sizing cracks in a cluster a machine learning approach is proposed where the crack pocket length is predicted by an ANN which is trained to learn the inverse relationship between the ACFM signal and crack pocket length based on a training database obtained via modelling. An earlier study by Ravan, et al [57,58] proposed a neural network approach to reconstruct isolated cracks of multi-humped profile from ACFM measurements, although this network would not generalise to the characterisation of clustered cracks. In the present study, the inversion aims to map inputs from domain to through a non-linear network to predict the crack knowledge of the RCF cracks.…”

Section: Comparison With Single Crack Sizing Methodsmentioning

confidence: 99%

Characterisation of clustered cracks using an ACFM sensor and application of an artificial neural network

Rowshandel

Nicholson

Shen

et al. 2018

NDT & E International

View full text Add to dashboard Cite

The alternating current field measurement (ACFM) technique can be applied for surface-breaking fatigue crack detection and sizing; the link between the ACFM signal and crack size is well understood for individual cracks. However, the ACFM response to multiple clustered cracks is significantly different to that of isolated cracks. In railway rails the high wheel-rail forces can lead to rolling contact fatigue (RCF) cracks. Often cracks appear together in small clusters or in long stretches. The accurate characterisation of such fatigue cracks is essential for carrying out efficient and safe repair and maintenance. This paper presents a method for sizing the important sub-surface section of multiple cracks using ACFM via the application of an artificial neural network (ANN). The approach is demonstrated using a railway case study: a simulation-based dataset of signal response covering the range of RCF cracks typically seen in in-service railway tracks has been generated to give a thorough representation of the effect of clustered crack parameters on the ACFM response. A 5×5×2×1 multi-layer ANN has been optimised and trained using the validated simulation database to learn the inverse relationship between the crack pocket length (desired output) and the ACFM signal for a given cluster of RCF cracks. The network has been evaluated on a set of experimental data to size cracks of known dimensions from ACFM measurements and also on unseen simulation data. Results from both simulation and experiment show that the approach presented can be used to size clustered cracks to approximately the same degree of accuracy as is possible for isolated cracks.

show abstract

Section: Comparison With Single Crack Sizing Methodsmentioning

confidence: 99%

Characterisation of clustered cracks using an ACFM sensor and application of an artificial neural network

Rowshandel

Nicholson

Shen

et al. 2018

NDT & E International

View full text Add to dashboard Cite

show abstract

“…This effect produces strong peaks and troughs in B Z at the ends of defect, whereas B X shows peaks at the ends and trough at the deepest point of defect. The detecting coils of the ACFM probe generally pick up the signals of B X and B Z for defect recognition and quantification [12][13][14]. The lift-off value affects the interaction between the ACFM probe and the workpiece, and causes error in defect recognition and quantification.…”

Section: U-shaped Acfm Probementioning

confidence: 99%

“…According to the ACFM principle, the maximum distortion of B X above crack (M Xmax ), and the maximum distortion of B Z above crack (M Zmax ) [12][13][14] are selected as the characteristic vectors to determine the depth of the crack [9]. The simulation results show that the characteristic vectors decrease exponentially as lift-off value (H) increases.…”

Section: Lift-off Effect Analysismentioning

confidence: 99%

Analysis of the lift-off effect of a U-shaped ACFM system

Chen

Yin

et al. 2013

NDT & E International

View full text Add to dashboard Cite

“…X direction shows the direction of the scan path, so the detection coil of 1-D probe is arranged along Y direction. [4], the electromagnetic signal components of the magnetic field, the B X and B Z , are the important components for crack quantitative recognition. Through simulation and calculation, 100 rows of the magnetic field data near the crack in the surface of simulation model are picked, and the contour map of the magnetic flux density B Z near the crack is plot in Figure 2.…”

Section: Study On Acfm Angle Detection Algorithmmentioning

confidence: 99%

Study on ACFM Crack Angle Detection with 1-D Array Probe

Li¹,

Chen²,

Liu³

et al. 2011

2011 Third International Conference on Measuring Technology and Mechatronics Automation

View full text Add to dashboard Cite

For the development of ACFM intelligent and visualization inspection system, based on a large number of finite element simulation results about ACFM crack angle detection, the calculate method of ACFM crack direction by 1-D array probe is present, and the experimental test is designed and conducted with the real crack. Test results show that, when tested by the 1-D probe composed of 4 couples of detecting coils, the accuracy of crack angle is more than 80%, which can meet the engineering needs, and the ACFM crack angle detection method is effective. Moreover, compared with the simulation analysis results, it is easy to know the calculation accuracy of the angle detection can be further improved by increasing detection coils or decreasing distance between two coils.

show abstract

Using a wavelet network for reconstruction of fatigue crack depth profile from AC field measurement signals

Cited by 13 publications

References 20 publications

Characterisation of clustered cracks using an ACFM sensor and application of an artificial neural network

Characterisation of clustered cracks using an ACFM sensor and application of an artificial neural network

Analysis of the lift-off effect of a U-shaped ACFM system

Study on ACFM Crack Angle Detection with 1-D Array Probe

Contact Info

Product

Resources

About