List of publications in 2017, in which our employees participated:
Vera Barat, Peter Kostenko, Vladimir Bardakov, and Denis Terentyev. Acoustic Signals Recognition by Convolutional Neural Network. International Journal of Applied Engineering Research ISSN 0973-4562 Volume 12, Number 12 (2017) pp. 3461-3469. https://ripublication.com/ijaer17/ijaerv12n12_67.pdf (full text). eLibrary ID: 31066184
Abstract
Acoustic emission is non-destructive method which is widely and successfully used to assess the technical condition of industrial facilities. It is challenging to divide useful signals and a noise in the flow of acoustic emission data. In this paper we describe the application of convolutional neural networks to classify acoustic emission signals by their time frequency structures. Convolutional neural network was learned to classify acoustic emission signals with 95% accuracy. Acoustic signal recognition by convolutional neural networks can greatly simplify the data processing.
S. Elizarov, V. Bardakov, A. Shimanskiy, A. Alyakritskiy, D. Terentyev, V. Barat, A. Gogin, and V. Koltsov. UNISCOPE: Instrument Integrating NDT Methods. Springer Proceedings in Physics. Volume 218. Advances in Acoustic Emission Technology. Proceedings of the World Conference on Acoustic Emission-2017, pp. 65-74. DOI: 10.1007/978-3-030-12111-2_7. eLibrary ID: 41689952
Abstract
This article describes the use of the multifunction device UNISCOPE. Tightness test of shutoff and control valves, evaluation of the integral thickness of test object, and concrete strength forecasting and monitoring of the movement of pipeline pig are the main topics described in the paper.
Sergey Elizarov, Vladimir Bardakov, Denis Terentyev, Vera Barat and Dmitry Chernov. AE Testing of Equipment on Operating Mode. Approaches and Results. World Conference on Acoustic Emission (WCAE-2017, Xi'an, China, October 10–13, 2017) (unpublished)
Abstract
The possibility of organizing a continuous monitoring AE without decommissioning of the following objects is shown: a walking excavator of the dragline type, a high-temperature process pipeline, and a 7-span motor bridge across the river. For each of the objects, the optimal operating frequency range, the threshold level and the distance between the sensors were determined, and a data analysis technique was developed that makes it possible to identify useful signals against a background of high noise. When testing the walking excavator AE, two defects were found that turned out to be 20 and 30 mm cracks, while monitoring the high-temperature process pipeline - flow through the flanges.
Barat V.A., Shimanskij A.G. Method of non-threshold automatic intellectual registration of acoustic emission signals by non-destructive control device. Russian patent №2660403 (2018). eLibrary ID: 37376547
Abstract
Field: physics. Substance: using: for non-destructive testing with the acoustic emission method (AE). Summary of the invention is that registration of acoustic emission pulses is performed without the use of a threshold limitation in the estimation of the parameters of the pulses of acoustic emission and noise in time intervals of a certain duration, the acoustic emission pulses are recorded on the basis of a combined criterion if one of the two statistical hypotheses is satisfied - the hypothesis that the sixth-order moment is zero and the hypothesis of the difference in the dispersion of acoustic emission signal samples calculated in neighbouring time windows, the time of the onset of the acoustic emission pulse is determined within the allocated time interval by the cumulative sum method. Effect: technical result: simplification of the procedure for conducting AE control through automatic non-threshold data recording, an increase in the sensitivity of the AE method due to the recording of pulses with an amplitude on the order of the noise level, as well as an increase in the accuracy of the AE sources coordinates determination by determination of the AE pulse beginning time.
Bardakov, V. V.; Sagaidak, A. I. Forecasting the strength of concrete during its hardening by the acoustic-emission method. Russian Journal of Nondestructive Testing, 2017. V. 53, pp. 436-443. DOI: 10.1134/S106183091706002X. eLibrary ID: 31100657
Abstract
Concrete structure formation has been studied by the acoustic-emission method. Using the results of the experimental research, a relationship has been revealed between the acoustic-emission data registered during the first 24 hours of hardening of a concrete composition and its strength at the age of 28 days. Informative parameters of acoustic-emission data have been determined that correlate with the strength of the concrete composition at the age of 28 days.
Vasilev I.E., Matvienko Yu.G., Elizarov S.V., Chernov D.V. Method for monitoring degradation of material structure and determining residual strength of article. Russian patent No. 2649081 (2018). eLibrary ID: 38150525
Abstract
Field: measurement technology. Substance: use to monitor the degree of degradation of a material structure and to determine an article residual strength. Core of the invention is performing acoustic-emission (AE) monitoring using location groups of active emission converters, preamplifiers and a system of collection-processing of registered arrays of active emission pulses, herewith in real time pulses of active emission are graded as per the relative energy level and the average emission frequency, the lower, the middle and the upper clusters are formed within the field of the specified parameters, and the percentage of pulses in each cluster is calculated reflecting micro-, meso-, and macrostructural fractures of the material, wherein as informative and stable acoustic emission parameters for the cluster separation of signals the relative pulse energy is used measured in decibels and corresponding to the number of emissions in units, as per which when compared with the results of testing the material for destruction the degree of degradation and the residual strength of the article in the zone of acoustic emission monitoring is determined, herewith the boundaries of the formed clusters are established as per the results of the preliminary testing the material of the article basing on the nature of the pulses radiation sources and the used signals discrimination threshold level. Effect: ensuring the possibility in the process of acoustic-emission diagnostics of the article to monitor the degree of degradation of the material structure and to determine the residual strength of the article in the AE monitoring zone without involving other methods of technical diagnostics and non-destructive testing.
Kachanov, V. K.; Sokolov, I. V.; Matyunin, V. M.; Barat, V. A.; Bardakov, V. V.; Marchenkov, A. Yu. Evaluation of the Fracture Toughness of Titanium Nitride Hardening Coatings According to Kinetic Indentation and Acoustic Emission Parameters. Measurement Techniques, 2017, 60, 7, pp. 706—710. DOI: 10.1007/s11018-017-1258-4. eLibrary ID: 31048648
Abstract
The possibility of usage of kinetic indentification method combined with acoustic emission for TiN-based hardening coatings crack resistance evaluation is revealed. Appropriate compliance between the moments of the first сrack appearance in the coating, the first notch on kinetic indentification curve registration and high-energetic acoustic emission impulse registration is established. The ability of much accurate and sound evaluation of critical indentation load causing the first ring crack appearance around the indent in the coating is shown.