Affiliation:
1. CUMHURİYET ÜNİVERSİTESİ
Abstract
The aim of this study is to investigate the relationship between the pulse echo value of the probes used in the ultrasonic examination method. Ultrasonic inspection method is known as one of the non-destructive inspection methods. Thanks to this method, discontinuities in the material can be detected without damaging the material to be examined. It works with the principle of sending the sound waves produced by piezoelectric materials to the material surface and reading the sound waves coming reversibly. As the general working principle, sound frequencies are used in the ranges that the human ear cannot hear. These frequency values vary according to the structure of the material and the environment to be examined, but generally vary between 1MHz and 4MHz. In our study, train wheels were preferred as material. Train wheels are produced from medium carbon steel material by forging method. Inspections of train wheels, which are exposed to dynamic stresses during operation, are of great importance. The method of drilling holes on the train wheels with 3mm artificial errors were created. Inspection was carried out with total probes with 3 different diameters and 3 different frequency values on the train wheels on which artificial defects were found. A spring mechanism has been prepared for the probes to contact the material surface with the same pressure force. Experimental groups and a data set was created of 9 tests were carried out with 6 different. The sound waves given from the probes were recorded as db unit. Although the frequency of the probes is seen as the basic parameter in the selection of probes in the literature, it has been observed that different probe diameters have an effect on the inspection sensitivity. IBk algorithm was used to determine the relationship between probe diameters and incoming sound waves. This algorithm is based on the principle of classifying objects based on the closest training examples in the feature space. In the model made using the IBk algorithm, a high correlation of 89% was found in the probe diameter and pulse echo value. This value is important in terms of showing the relationship between the probe diameters, which cannot be considered as the main parameter, and the pulse echo value.
Publisher
Karadeniz Fen Bilimleri Dergisi
Reference23 articles.
1. Afrin, F., & Nahar, I. (2015). Incremental learning based intelligent job search system [PhD Thesis]. BRAC University.
2. Bahonar, M., & Safizadeh, M. S. (2022). Investigation of real delamination detection in composite structure using air-coupled ultrasonic testing. Composite Structures, 280, 114939.
3. Bilgin, M. (2017). Gerçek veri setlerinde klasik makine öğrenmesi yöntemlerinin performans analizi. Breast, 2(9), 683.
4. Bojović, N. J. (2002). A general system theory approach to rail freight car fleet sizing. European Journal of Operational Research, 136(1), 136-172.
5. Caruana, R., & Niculescu-Mizil, A. (2006). An empirical comparison of supervised learning algorithms. Proceedings of the 23rd international conference on Machine learning, 161-168.