Quantify pixel-level detection of dam surface crack using deep learning

An increasing number of detection methods based on computer vision are applied to detect cracks in water conservancy infrastructure. However, most studies directly use existing feature extraction networks to extract cracks information, which are proposed for open-source datasets. As the cracks distribution and pixel features are different from these data, the extracted cracks information is incomplete. In this paper, a deep learning-based network for dam surface crack detection is proposed, which mainly addresses the semantic segmentation of cracks on the dam surface. Particularly, we design a shallow encoding network to extract features of crack images based on the statistical analysis of cracks. Further, to enhance the relevance of contextual information, we introduce an attention module into the decoding network. During the training, we use the sum of Cross-Entropy and Dice Loss as the loss function to overcome data imbalance. The quantitative information of cracks is extracted by the imaging principle after using morphological algorithms to extract the morphological features of the predicted result. We built a manual annotation dataset containing 1577 images to verify the effectiveness of the proposed method. This method achieves the state-of-the-art performance on our dataset. Specifically, the precision, recall, IoU, F1_measure, and accuracy achieve 90.81%, 81.54%, 75.23%, 85.93%, 99.76%, respectively. And the quantization error of cracks is less than 4%.

Control of fatigue failure mechanisms in multilayer coatings by varying the architectural parameters of an intermetallic interlayer

A multilayer overlay coating system containing an intermediate intermetallic layer (designated 2IML) is an architecture expected to show good fatigue resistance. Experimental characterisation and modelling simulations were carried out to classify the different crack initiation mechanisms occurring during fatigue of this coating system and to reveal how changes in the layer architecture lead to fatigue improvement. Fatigue improvement is achieved by decreasing the IML-Top layer thickness due to the increased surface crack initiation resistance. However subsurface initiation mechanisms inhibit the improvement (dominated by surface initiation mechanism) achieved by locating the IML-Top layer closer to the top surface.

Determination of parameters of surface defects for quality control of repair work on submerged crossings

Внутритрубная диагностика позволяет обеспечить высокую выявляемость поверхностных трещиноподобных дефектов стенки трубопровода. Для их устранения в условиях проведения ремонтных работ на подводных переходах магистральных нефте- и нефтепродуктопроводов традиционно используется установка герметизирующей камеры, изолирующей дефектный участок и позволяющей в полной мере провести обследование области с опасным дефектом и дальнейший ремонт с применением методов неразрушающего контроля. Данная технология является безальтернативной, когда требуется вырезка-вварка катушки или другая сварочно-монтажная операция, однако представляется избыточной для устранения поверхностного дефекта методом вышлифовки. Современные технологии позволяют устранить поверхностный трещиноподобный дефект без установки герметизирующей камеры и тем самым значительно уменьшить трудоемкость и стоимость ремонта. Однако их широкое применение сдерживается отсутствием технической возможности проведения контроля качества отремонтированного участка под водой методами неразрушающего контроля с эквивалентной чувствительностью и выявляемостью относительно обычных условий. В этой связи выполнены экспериментальные исследования с целью определения пороговых значений параметров поверхностных трещиноподобных дефектов, выявляемых методами неразрушающего контроля в подводных условиях с учетом качества зашлифованной поверхности. Такие значения получены, в частности, для метода ультразвукового контроля с технологией фазированной решетки, модуляционного метода контроля с использованием вихретоковой матрицы, люминесцентного магнитопорошкового метода. Сформулированы требования к характеристикам оборудования для проведения неразрушающего контроля в условиях подводных работ. In-line diagnostics makes it possible to secure high detectability of surface crack-like defects of the pipeline wall. To eliminate them in conditions of repair work performed on submerge crossings of main oil and oil products pipelines, the installation of a sealing chamber is traditionally used to isolate the defective area and to enable a full-investigation of the area with a hazardous defect and further repairs, by non-destructive testing methods. This technology has no alternatives when cutting and welding of the coil or other welding and assembly operation is required, but it seems redundant for elimination of a surface defect by grinding method. Modern technologies make it possible to eliminate a surface crack-like defect without installing a sealing chamber and thereby can substantially cut manpower input and cost of repair. Their widespread use is deterred by the lack of technical possibility for quality control of the repaired area under water by nondestructive testing methods with equivalent sensitivity and detectability versus the normal conditions. In this regard, experimental research has been conducted to determine the threshold values of the parameters of surface crack-like defects detected by non-destructive testing methods in submerged conditions, taking into account the quality of the sanded surface. Such values were obtained, in particular, for the ultrasonic testing method with phased array technology, the modulation control technique using eddy current matrix, and the luminescent magnetic powder method. Requirements to performance characteristics of equipment for nondestructive testing in submerged operations have been formulated.

Calculation model for the evaluation of tired defect development in the freight wagon side frame

The reliability and safe operation of mechanical elements of rail transport is an important and relevant scientific and technical issue since high-loaded units and elements of its chassis are exposed to prolonged operation and their failure can lead to damage with catastrophic consequences. To prevent the possible failure of such objects, there is a necessity for a reliable estimation of their residual life. Among the cast parts of the freight car trolley, the side frames are one of the most loaded elements, which take on dynamic loads that cause vibrations of the unsprung parts of the freight car. The side frame of the 18-100 trolley, as a typical representative of a number of trolleys of freight cars, does not have a sufficient margin of resistance to fatigue and survivability, so it is sensitive to some deviations and defects (different wall thickness, sinks, and pores, residual stresses) that are detected during the operation process. Based on the energy approach of the mechanics of materials fatigue failure, the calculation models for estimating the dynamics of the development of crack-like defects under the action of operational load are developed in this work. The calculation models constructed using modern information technologies, and the software developed for their numerical implementation allow predicting the impact of irregular cyclic loading and complex stress on the growth of surface fatigue crack in the side frame of the carriage, which significantly brings the problem closer to real operating conditions. Numerical calculations were performed using a computer program of our own development in the Python programming language. At the first stage of the program functioning the spectrum of amplitudes of irregular cyclic loading is built, at the second - the program module of numerical solution construction for systems of usual differential equations of the proposed mathematical model of fatigue defect development is started. Calculations of the fatigue crack growth dynamics taking into account the action of shear stresses in the section with the crack of the side frame showed a slight effect of shear stresses on the residual durability of the frame. It is demonstrated that the dynamics of surface crack development significantly depend on its initial geometry. The proposed calculation method for determining the initial shape of the surface crack with a minimum period to critical growth can be effectively used to predict the residual durability during the technical diagnosis of the running gear parts of railway rolling stock. Keywords: wagon side frame; fatigue crack; probability distribution function; load amplitude spectrum; numerical method.

Three-Dimensional Surface Crack Growth of Maraging Steel Spherical Pressure Shell

This study focused on the three-dimensional surface crack growth of a spherical pressure shell. Eight maraging steel 18Ni (250) samples were fabricated and tested, and the fatigue crack growth rate curves were obtained. Considering the influence of plastic closure effect and sample thickness on crack growth, the fitting formula of fatigue crack growth only related to materials was obtained. Based on the three-dimensional crack closure theory and the strip yield model, a three-dimensional surface crack growth model of spherical pressure shell was established. By using a self-written program and FRANC3D, the three-dimensional surface crack growth simulations of the spherical shell were completed. The influence of the initial shape ratio and initial depth of the crack on the crack growth and the fatigue life of the spherical shell was analyzed.