A Point Crack Source Location Method without Velocity Information in Anisotropic Plates

Round 1

Reviewer 1 Report

Authors proposed a technique to localize the acoustic source in an anisotropic solid by minimizing an objective function. This objective function is obtained by assuming the wave front shape to have an elliptical shape.  Please note that the acoustic source has been already localized in this manner by Sen and Kundu [1.2].  Apparently the authors are not familiar with these works since they did not even refer to these works.  Unless the authors can clearly point out what is new in this work and how it is better than Sen and Kundu’s approach I cannot recommend this paper for publication.

N. Sen nand T. Kundu “A new wave front shape-based approach for acoustic source localization in an anisotropic plate without knowing its material properties”, Ultrasonics, Vol. 87, pp. 20-32, 2018.

N. Sen and T. Kundu, “Acoustic source localization in a highly anisotropic plate with unknown orientation of its axes of symmetry and material properties with numerical verification”, Ultrasonics, Vol. 100, pp. 105977, https://doi.org/10.1016/j.ultras.2019.105977, 2020.

Author Response

Thank you for pointing out the papers [1,2]. The paper [1,2] also use minimization of an objective function to locate the source coordinates without any knowledge of the elastic properties of the material. However, those papers require to use a group of sensors arranged in an L-shape, while our study does not require any specific sensor alignments. In addition, the novelty of our study lies in devising the concept of time offset for TOA error correction due to signal noise and applying it to the objective function.

In the study of [1,2], the number of unknowns was reduced by differentiating the ellipse equation, and the differential elements according to the differentiation were replaced with information from the L-shape. Therefore, the L-shape arrangement in which the three sensors are arranged to form a right isosceles triangle was essentially used to reduce the number of unknowns. In our study, the number of unknowns was reduced by solving the system of two ellipse equations. Therefore, special sensor arrangement is not required. In addition, the novelty of our study lies in devising the concept of time offset for TOA error correction due to signal noise and applying it to the objective function.

The paper [1] and [2] was referred to as [24] and [23] in the introduction (on page 2, lines 39 - 43), respectively. In response to the reviewer, we provide a more detailed comparison with these paper in the discussion section.

Author Response File: Author Response.pdf

Reviewer 2 Report

In this paper, an acoustic emission source location method for anisotropic plates without velocity information is proposed. The author uses gradient descent method to minimize the objective function to get the final result. At the same time, the effect of noise is considered, and the effectiveness and robustness of the method are verified by numerical simulation and experiments. The reviewer believes that the author needs to make major revisions to the following issues.

1)      It is mentioned that the proposed method doesn’t need to know the velocity information in advance. However, in Section 3.1, the calculation of t_i uses the velocity. In addition, the elastic wave velocity in the x direction mentioned later is 3 times faster than that in the y direction. These data are obtained when velocity information is known.

2)      Is there any basis for proposing the time offset value s_i? The author needs to give a detailed explanation.

3)      In fact, the acoustic emission source location problem is transformed into a simple function optimization problem. The author used different numbers of sensors, such as 4, 6, 8, 10 and 12. When the sensors are arranged in a straight line, the proposed method may produce false location results due to symmetry. Therefore, only a specific sensor layout method can be used. Whether this is inconsistent with the method proposed in the paper that doesn’t require special sensor layout.

4)      The author needs to elaborate on the reasons for using broken pencil lead instead of crack failures in composite panels.

5)      The author should explain why 1MHz excitation signal is selected for simulation. This is quite different from the center frequency of broken pencil lead and the resonant frequency of R15 sensor used in the experiment. There is no good correspondence between simulation and experiment.

6)      In order to verify the noise immunity of the proposed algorithm, the author uses noise-containing signals with SNR of 10, 30 and 50 for experiments. The reviewer believes that in real engineering applications, the true SNR should be much lower than the SNR proposed by the author. It is suggested that the author adds experiments at or below 0dB SNR to verify the noise immunity of the proposed algorithm.

7)      The author has carried out experiments with different number of sensor, and the experimental results need to be displayed. At the same time, has the author analyzed the effect of different number of sensors on the location accuracy?

Comments for author File: Comments.docx

Author Response

Thanks for the feedback regarding our work. We revised the manuscript according to the comments. We attached the pdf for the answers.

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript of applsci-1972405 introduced a new AE location method, which was interesting and valuable. The reviewer has a suggestion that more realistic AE waves used in this study should be provided in the manuscript to make the experimental work more convincing.

Author Response

Thanks for the comment. In order to use more realistic AE waves, we changed the excitation signal according to PLB modeling studies[1,2,3]; A waveform that applies a Hanning filter to a sine (or cosine) waveform with central frequency of 140 kHz. According to the study of Falcetelli et al. [1], the burst signal generated by pencil-lead breaking generally had a frequency range of 50-250 kHz and a central frequency of about 140 kHz. We conducted numerical experiments according to the changed excitation signal, and modified the manuscript.

[1] Falcetelli, F., Romero, M. B., Pant, S., Troiani, E., & Martinez, M. (2018, July). Modelling of pencil-lead break acoustic emission sources using the time reversal technique. In Proceedings of the 9th European Workshop on Structural Health Monitoring, Manchester, UK (pp. 10--13).
[2] M. Hamstad, "Acoustic emission signals generated by monopole (pencil-lead break) versus dipole sources: finite element modeling and experiments," Journal of Acoustic Emissions, vol. 25, pp. 92-106, 2007.
[3] Falcetelli, Francesco, "Modelling of pencil-lead break acoustic emission sources using the time reversal technique." Master degree thesis, UNIVERSITA’ DI BOLOGNA. 2018.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I am satisfied with the authors' revision of the paper.  I recommend its publication.

Author Response

Thanks for your comments and acceptance.

Reviewer 2 Report

The author adds the relevant explanation of time offset value . Are the  of different signals the same? Is the definition of the  based on a large number of experiments? Is it universal? The author needs to explain this.

The author gives the definition of SNR in the manuscript. This is different from the common SNR. The author defines it as the ratio of the mean to the standard deviation of the signal arrival time. The formula doesn’t contain log. The unit of SNR in the manuscript is still dB. Please check whether this is reasonable. At the same time, the reviewer doesn’t recommend that the author use the name SNR. This will cause confusion to readers. The reviewer suggested changing the name and adding the calculation formula.

Author Response

Thanks for your comments. We revised the manuscript according to the comments. We attached the pdf for the answers.

Author Response File: Author Response.pdf