Application of the Maximum Entropy Method in Ultrasonic Flaw Detection with Allowance for Echo Signal Shape Variability

  • Евгений [Evgeniy] Геннадиевич [G.] Базулин [Bazulin]
  • Антон [Anton] Сергеевич [S.] Вовк [Vovk]
Keywords: eultrasonic flaw detection, super resolution, inverse scattering problem, maximum entropy method

Abstract

In order to reconstruct the image of reflectors with super resolution for more accurately determining the size of discontinuities in the examined item, it is suggested to use the maximum entropy (ME) method for processing echo signals measured in the TOFD mode. The effect of super resolution will make it possible to simplify the procedure for determining the phase of echoes from the crack upper and lower edges, a circumstance that is of much importance for their recognition. Conventionally, application of the ME method involves the need to calculate the circulant matrix correlating the reflection coefficient in the examined item and the echo signal measured in the TOFD mode. However, the echo signal shape depends on the reflector occurrence depth, a circumstance due to which phantom flare spots appear in the reconstructed image. A program for calculating the modified matrix has been developed, which takes into account the altered shape of the echo reflected from a point reflector lying at an arbitrary depth. For checking the efficiency of the echo signal calculation program, the obtained echo signals were compared with the echo signals calculated using the CIVA program. The obtained echoes have been found to differ from each other by less than 5%. The images obtained in a numerical experiment were used to study the effect the echo signal variation factor has on the results obtained from applying the ME method. It has been found that the echo signal waveform variability imposes certain limitation on using the ME method with a circulant matrix: the "side lobes" make around 40% of the reflector flare amplitude on the reconstructed image. The use of the modified ME method for reconstructing the reflector image made it possible to significantly reduce the level of "side lobes" and increase the resolving power by more than a factor of 3. Model experiments have also confirmed the effectiveness of the ME method with the modified matrix. The head wave impulse and the impulse reflected once from the examined item bottom have become narrower by more than a factor of 5 with a small level of "side lobes". Owing to this circumstance, the impulses were easily found to be in antiphase.

Information about authors

Евгений [Evgeniy] Геннадиевич [G.] Базулин [Bazulin]

Science degree:

Dr.Sci. (Techn.)

Workplace

Electrical Engineering and Introscopy Dept., NRU MPEI; Research and Production Center of Non-destructive Testing «ECHO+»

Occupation

Professor; Leading Researcher

Антон [Anton] Сергеевич [S.] Вовк [Vovk]

Workplace

Electrical Engineering and Introscopy Dept., NRU MPEI

Occupation

Student

References

1. ISO 16828:2012. Неразрушающий контроль. Ультразвуковой контроль. Времяпролетный дифракционный метод в качестве метода обнаружения и определения размера несплошностей.

2. Darmon M., Ferrand A., Dorval V., Chatillon S. Recent Modelling Advances for Ultrasonic TOFD Inspections // Rev. Prog. Quant. Nondestruct. Eval. 2015. V. 34. Pp. 1757—1765.

3. Базулин Е.Г. О возможности использования в ультразвуковом неразрушающем контроле метода максимальной энтропии для получения изображения рассеивателей по набору эхосигналов // Акустический журнал. 2013. Т. 59. № 2. С. 235—254.

4. Ковалев А.В. и др. Импульсный эхо-метод при контроле бетона. Помехи и пространственная селекция // Дефектоскопия. 1990. № 2. C. 29—41.

5. Lingvall F. Time-domain Reconstruction Methods for Ultrasonic Array Imaging: A Statistical Approach. Uppsala: Uppsala University, 2004.

6. Schmerr Jr., Lester W. Fundamentals of Ultrasonic Phased Arrays. Springer, 2015.

7. Hansen P.C. The L-curve and Its Use in the Numerical Treatment of Inverse Problems // Computational Inverse Problems in Electrocardiology. Southampton: WIT Press, 2001. Pp. 119—142.

8. Данилов В.Н. О диаграмме направленности наклонного преобразователя в режиме приема // Дефектоскопия. 2011. № 4. С. 33—49.

9. Левин М.Н., Татаринцев А.В., Ахкубеков А.Э. Метод Laplace-DLTS с выбором параметра регуляризации по L-кривой // ФТП. 2009. Т. 43. Вып. 5. С. 613—616.

10. Тихонов А.Н., Арсенин В.Я. Методы решения некорректных задач. М.: Наука, 1986.

11. Kullback S. Information Theory and Statistics. N.-Y.: Dover Publ., 1968.

12. Сумин М.И. Метод регуляризации А.Н. Тихонова для решения оптимизационных задач. Нижний Новгород: Изд-во Нижегородского гос. ун-та, 2016.

13. Extende CIVA [Офиц. сайт] http://www.extende. com/fr (дата обращения 20.02.2018).
---
Для цитирования: Базулин Е.Г., Вовк А.С. Применение метода максимальной энтропии в ультразвуковой дефектоскопии с учетом переменной формы эхосигнала // Вестник МЭИ. 2018. № 5. С. 111—119. DOI: 10.24160/1993-6982-2018-5-111-119.
#
1. ISO 16828:2012. Nerazrushayushchiy Kontrol'. Ul'trazvukovoy Kontrol'. Vremyaproletnyy Difraktsionnyy Metod v Kachestve Metoda Obnaruzheniya i Opredeleniya Razmera Nesploshnostey. (in Russian).

2. Darmon M., Ferrand A., Dorval V., Chatillon S. Recent Modelling Advances for Ultrasonic TOFD Inspections. Rev. Prog. Quant. Nondestruct. Eval. 2015;34: 1757—1765.

3. Bazulin E.G. O Vozmozhnosti Ispol'zovaniya v Ul'trazvukovom Nerazrushayushchem Kontrole Metoda Maksimal'noy Entropii dlya Polucheniya Izobrazheniya Rasseivateley po Naboru Ekhosignalov. Akusticheskiy Zhurnal. 2013;59;2:235—254. (in Russian).

4. Kovalev A.V. i dr. Impul'snyy Ekho-metod pri Kontrole Betona. Pomekhi i Prostranstvennaya Selektsiya. Defektoskopiya. 1990;2:29—41. (in Russian).

5. Lingvall F. Time-domain Reconstruction Methods for Ultrasonic Array Imaging: A Statistical Approach. Uppsala: Uppsala University, 2004.

6. Schmerr Jr., Lester W. Fundamentals of Ultrasonic Phased Arrays. Springer, 2015.

7. Hansen P.C. The L-curve and Its Use in the Numerical Treatment of Inverse Problems. Computational Inverse Problems in Electrocardiology. Southampton: WIT Press, 2001:119—142.

8. Danilov V.N. O Diagramme Napravlennosti Naklonnogo Preobrazovatelya v Rezhime Priema. Defektoskopiya. 2011;4:33—49. (in Russian).

9. Levin M.N., Tatarintsev A.V., Akhkubekov A.E. Metod Laplace-DLTS s Vyborom Parametra Regulyarizatsii po L-krivoy. FTP. 2009;43;5: 613—616. (in Russian).

10. Tikhonov A.N., Arsenin V.Ya. Metody Resheniya Nekorrektnykh Zadach. M.: Nauka, 1986. (in Russian).

11. Kullback S. Information Theory and Statistics. N.-Y.: Dover Publ., 1968.

12. Sumin M.I. Metod Regulyarizatsii A.N. Tikhonovadlya Resheniya Optimizatsionnykh Zadach. Nizhniy Novgorod: Izd-vo Nizhegorodskogo Gos. Un-ta, 2016. (in Russian).

13. Extende CIVA [Ofits. Sayt] http://www.extende. com/fr (Data Obrashcheniya 20.02.2018).
---
For citation: Bazulin E.G., Vovk A.S. Application of the Maximum Entropy Method in Ultrasonic Flaw Detection with Allowance for Echo Signal Shape Variability. MPEI Vestnik. 2018;5:111—119. (in Russian). DOI: 10.24160/1993-6982-2018-5-111-119.
Published
2018-10-01
Section
Radio Engineering and Communications (05.12.00)