Visualization of internal defects using a deep generative neural network model and ultrasonic nondestructive testing

Introduction. The development of machine learning methods has given a new impulse to solving inverse problems in mechanics. Many studies show that along with well-behaved techniques of ultrasonic, magnetic, and thermal nondestructive testing, the latest methods are used, including those based on neural network models. In this paper, we demonstrate the potential application of machine learning methods in the problem of two-dimensional ultrasound imaging. Materials and Methods. We have developed an experimental model of acoustic ultrasonic non-destructive testing, in which the probing of the object under study takes place, followed by the recording of the response signals. The propagation of an ultrasonic wave is modeled by the finite difference method in the time domain. An ultrasonic signal received at the internal points of the control object is applied to the input of the convolutional neural network. At the output, an image that visualizes the internal defect is generated.Results. In the course of the performed complex of numerical experiments, a data set was generated for training a convolutional neural network. A convolutional neural network model, which is developed to solve the problem of visualizing internal defects based on methods of ultrasonic nondestructive testing, is presented. This model has a small size, which is 3.8 million parameters. Its simplicity and versatility provide high-speed learning and a wide range of applications in the class of related problems. The presented results show a high degree of information content of the ultrasonic response and its correspondence to the real form of an internal defect located inside the test object. The effect of geometric parameters of defects on the accuracy of the neural network model is investigated.Discussion and Conclusion. The results obtained have established that the proposed model shows a high operating accuracy (F1 > 0.95) in cases when the wavelength of the probe pulse is tens of times less than the size of the defect. We believe that the combination of the proposed methods in this approach can serve as a good starting point for future research in solving flaw defection problems and inverse problems in general.

К расчету поля конечного магнитного цилиндра с внутренним соосным цилиндрическим дефектом

Using the grid method, a numerical solution of the direct problem of magnetostatics for calculating the field of a finite cylinder with a constant magnetic permeability containing an internal inclusion in the form of a coaxial cylinder with a different magnetic permeability is carried out. The algorithm is created for an arbitrary external field. In order to assess the reliability and accuracy of the solution method, the results were tested using precisely solved problems. A comparison is also made with the results of the previously solved problem of a finite defect-free cylinder. The coordinate dependences of the components of the resulting field are constructed for different source data. The program adds to the library of magnetic control problems and can be used for high-quality verification with the results of model experiments, as well as for evaluating the geometric characteristics of an internal defect.

Effect of Carburizing and Nitriding on Fatigue Properties of 18Cr2Ni4WA Steel in Very High Cycle Fatigue Regime

The work aims to study the influence of carburizing and nitriding on fatigue properties of 18Cr2Ni4WA high strength steel in very high cycle fatigue regime. Very high cycle fatigue tests were carried out on 18Cr2Ni4WA Steel after carburizing and nitriding respectively. The micro morphology of fatigue fracture was observed by scanning electron microscope, the failure mode and failure mechanism were discussed. The relationship between fatigue life and defect size, FGA size, fish eye size of fracture was analyzed. The characteristic size of defects is evaluated by Gumbel, Weibull and GEV distribution functions, and a modified Akiniwa fatigue life prediction model considering the relationship between FGA size and inclusion size was established. The results showed that, nitriding and carburizing treatment improve the surface fatigue limit of the steel. The fatigue life decreases with the increase of internal defect size and FGA size. After carburizing and nitriding treatment, the internal fatigue strength of the specimen decreases slightly. When the failure probability is 99%, the internal defect sizes of nitrided specimens calculated by Weibull, Gumbel and GEV distributions are 141.5 μm, 148.4 μm and 211.7 μm respectively. The calculated internal defect sizes of carburized specimens are 47 μm, 67.8 μm and 40 μm respectively. Compared with the experimental data, the fatigue strength predicted by GEV is the most appropriate. carburizing and nitriding treatment can improve the surface fatigue strength of 18Cr2Ni4WA steel, but slightly reduce the internal fatigue strength. The prediction result of the new model is conservative when the failure probability is 99%, which is suitable for engineering application.

Threshold-Optimized Swarm Decomposition Using Grey Wolf Optimizer for the Acoustic-Based Internal Defect Detection of Arc Magnets

The acoustic-based internal defect detection is essential to ensure the quality of arc magnets efficiently. Swarm decomposition (SWD) is conducive to processing acoustic signals, but it is still confronted with threshold optimization problems. Especially, the existing optimization methods for the SWD thresholds are merely available for a single signal with exclusive characteristics, instead of the various signals with similar characteristics. Therefore, a threshold-optimized SWD using grey wolf optimizer (GWO) is proposed to solve these issues and applied to detect the internal defects of arc magnets. In this method, a fitness function is designed to indicate the relationship between the SWD thresholds and the overall decomposition effect of similar signals. The minimum value of it corresponds to the threshold setting yielding the optimal decomposition. GWO is used for searching such a minimum value, and the obtained optimal threshold setting allows SWD to decompose any signal into a series of oscillatory components. The frequency information in the two oscillatory components with the highest energy ratio is extracted as the internal defect features. Random forest is carried out to identify these features. Experimentally, the detection accuracy reaches above 97%, and the detection speed per single arc magnet does not exceed 3.4 seconds. The proposed method cannot only determine the unified threshold setting of SWD for similar signals but also achieve an accurate, rapid detection for the internal defects.

The Influence of a Novel Hydrophobic Agent on the Internal Defect and Multi-Scale Pore Structure of Concrete

In high humidity areas, it is necessary to improve the impermeability of concrete to water and other erosion solutions. The internal defect and pore channel of concrete are the main factors affecting the impermeability and durability. In this paper, a novel hydrophobic agent named Yellow River Engineering Consulting (YREC) was prepared. The relative internal defect degree of concrete with different curing ages and YREC contents was evaluated by ultrasonic non-destructive testing as qualitative characterization method, and the effect of YREC on hydration reaction was investigates using X-ray powder diffraction (XRD). Water permeability and contact angle tests were used to analyze the internal and external hydrophobicity induced by YREC addition, respectively. The pore structure changes of concrete mortar matrix induced by YREC were further discussed applying low-temperature liquid nitrogen adsorption (LT-NA) and mercury intrusion/extrusion porosimetry (MIP). The results indicated that YREC not only improves the impermeability of water, but also greatly enhances the mechanical strength. In the case of mixing YREC, the porosity of concrete mortar matrix decreases accompanied with the more advantage pores (micropores and transition pores) developed. Based on the relative internal defect degree and the changes of multi-scale pore structure, the functionality and durability of concrete with 4% YREC addition are the most desirable.