Pribory i Metody Izmerenij (Dec 2020)

Smartphone-Based Automated Non-Destructive Testing Devices

  • V. F. Petryk,
  • A. G. Protasov,
  • R. M. Galagan,
  • A. V. Muraviov,
  • I. I. Lysenko

DOI
https://doi.org/10.21122/2220-9506-2020-11-4-272-278
Journal volume & issue
Vol. 11, no. 4
pp. 272 – 278

Abstract

Read online

Currently, non-destructive testing is an interdisciplinary field of science and technology that serves to ensure the safe functioning of complex technical systems in the face of multifactorial risks. In this regard, there is a need to consider new information technologies based on intellectual perception, recognition technology, and general network integration. The purpose of this work was to develop an ultrasonic flaw detector, which uses a smartphone to process the test results, as well as transfer them directly to an powerful information processing center, or to a cloud storage to share operational information with specialists from anywhere in the world.The proposed flaw detector consists of a sensor unit and a smartphone. The exchange of information between the sensor and the smartphone takes place using wireless networks that use "bluetooth" technology. To ensure the operation of the smartphone in the ultrasonic flaw detector mode, the smartphone has software installed that runs in the Android operating system and implements the proposed algorithm of the device, and can serve as a repeater for processing data over a considerable distance (up to hundreds and thousands of kilometers) if it necessary.The experimental data comparative analysis of the developed device with the Einstein-II flaw detector from Modsonic (India) and the TS-2028H+ flaw detector from Tru-Test (New Zealand) showed that the proposed device is not inferior to them in terms of such characteristics as the range of measured thicknesses, the relative error in determining the depth defect and the object thickness. When measuring small thicknesses from 5 to 10 mm, the proposed device even surpasses them, providing a relative measurement error of the order of 1 %, while analogues give this error within 2–3 %.

Keywords