Russian Journal of Nondestructive Testing最新文献_第4页

AlignDisc: A Discriminative Approach for Welding Anomaly Detection in X-ray Images AlignDisc：一种用于焊接x射线图像异常检测的判别方法

IF 1.3 4区材料科学

Russian Journal of Nondestructive Testing Pub Date : 2026-08-27 DOI: 10.1134/S1061830925605768

Haochen Zhao, Xingyu Wen, Zhihao Yang, Hongbing Wang, Jiang Zhu

{"title":"AlignDisc: A Discriminative Approach for Welding Anomaly Detection in X-ray Images","authors":"Haochen Zhao, Xingyu Wen, Zhihao Yang, Hongbing Wang, Jiang Zhu","doi":"10.1134/S1061830925605768","DOIUrl":"10.1134/S1061830925605768","url":null,"abstract":"X-ray radiographic inspection, as a core non-destructive testing technique, plays an indispensable role in ensuring weld quality and structural integrity across critical industrial applications. Accurate anomaly detection in X-ray weld images is crucial for preventing missed detection of high-risk defects (e.g., cracks, lack of fusion) that may cause catastrophic failures. Due to advances in welding processes and quality control, such critical defects occur infrequently in modern manufacturing, making limited anomalous samples from historical inspection records particularly valuable for understanding defect patterns. However, existing unsupervised methods rely solely on normal samples, failing to exploit these scarce but informative anomalous cases, thereby limiting detection accuracy. We propose AlignDisc, a discriminative anomaly detection framework that integrates dual feature enhancement mechanisms with few-shot discriminative learning. Specifically, we design a discriminator module that performs classification training using limited anomalous and normal samples to enhance anomaly discrimination. To strengthen discriminative features, we propose a joint local-global feature guidance mechanism: local feature guidance amplifies distinctions between normal and anomalous regions by leveraging spatial correlations, while global feature guidance aligns features from diverse scenarios through adaptive prototypes. This dual enhancement improves feature discriminability and enables robust detection across varying X-ray imaging conditions. Experimental results on the SWRD X-ray weld defect dataset demonstrate significant improvements over PatchCore: Detection AUROC increases from 82.9 to 94.7%, Segmentation AUPRO rises from 67.7 to 82.4%, and Segmentation AUROC improves from 88.6 to 92.9%. Comparative experiments on real-world industrial datasets further validate the superiority and effectiveness of the proposed method.","PeriodicalId":764,"journal":{"name":"Russian Journal of Nondestructive Testing","volume":"62 5","pages":"674 - 692"},"PeriodicalIF":1.3,"publicationDate":"2026-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148823046","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Determination of the Height of Planar Reflectors from Multi-Mode DAF-Images 多模daf图像平面反射面高度的确定

IF 1.3 4区材料科学

Russian Journal of Nondestructive Testing Pub Date : 2026-08-27 DOI: 10.1134/S1061830926601182

A. E. Bazulin, E. G. Bazulin, A. A. Efimovskaya

引用次数: 0

Identification of the Optimal Wavelet Type for Structural Health Monitoring: A Case Study on Concrete Filled Steel Tubular Columns 结构健康监测中最优小波类型的识别——以钢管混凝土柱为例

IF 1.3 4区材料科学

Russian Journal of Nondestructive Testing Pub Date : 2026-08-27 DOI: 10.1134/S1061830925605707

Adel Younesi

{"title":"Identification of the Optimal Wavelet Type for Structural Health Monitoring: A Case Study on Concrete Filled Steel Tubular Columns","authors":"Adel Younesi","doi":"10.1134/S1061830925605707","DOIUrl":"10.1134/S1061830925605707","url":null,"abstract":"Damage detection plays an important role in maintaining the health, safety, and optimal performance of infrastructure, especially civil infrastructure. Among various signal processing techniques, wavelet transform has emerged as a powerful tool for detecting and localizing damage in structural components. This study aims to identify the most suitable type of wavelet for effective SHM, with a specific focus on concrete-filled steel tubular (CFST) columns, which are widely used due to their high load-bearing capacity and ductility. Several commonly used wavelet families such as Daubechies, Symlets, Coiflets, Haar, Biorthogonal are evaluated based on their ability to extract damage-sensitive features from vibration signals obtained under simulated loading conditions. The performance of each wavelet type is assessed in terms of signal clarity, noise reduction, and damage localization accuracy. The results reveal that the choice of wavelet significantly influences the detection capability, and an optimal wavelet can enhance the reliability of SHM systems. The findings of this case study provide practical insights for engineers and researchers in selecting appropriate wavelets for monitoring CFST columns and similar composite structural elements. After conducting a detailed error analysis and averaging the results, it was found that the Bior5.5 with a scale factor of 4 yielded the minimum number and magnitude of maximum errors in accurately localizing damage within CFT columns.","PeriodicalId":764,"journal":{"name":"Russian Journal of Nondestructive Testing","volume":"62 5","pages":"616 - 632"},"PeriodicalIF":1.3,"publicationDate":"2026-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148823048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Spectral Method for Estimating the Velocities of Normal Waves in Tube Blanks of Deep Rod Pumps Taking into Account the Dispersion 考虑频散的深抽杆泵管坯中法向波速度的频谱估计方法

IF 1.3 4区材料科学

Russian Journal of Nondestructive Testing Pub Date : 2026-08-27 DOI: 10.1134/S1061830926600966

V. A. Strizhak

{"title":"A Spectral Method for Estimating the Velocities of Normal Waves in Tube Blanks of Deep Rod Pumps Taking into Account the Dispersion","authors":"V. A. Strizhak","doi":"10.1134/S1061830926600966","DOIUrl":"10.1134/S1061830926600966","url":null,"abstract":"The value of the acoustic wave velocity is a necessary parameter during flaw detection, structroscopy and thickness measurement. In some cases, the exact calculation of speed is complicated by the presence of its variance, in which the speed value changes with frequency. The paper presents a method for calculating the velocities of a dispersionless torsional wave and a dispersive zero symmetric mode of a Pochammer wave based on the full spectrum of an echogram of multiple reflections along the length of a pipe. The method is based on the detection of spectral lines in the spectrum of the recorded signal that repeat with an interval Δf on the frequency axis corresponding to the repetition frequency of the bottom pulses. Subsequent interpolation of the obtained values of the velocities of the L(0.1) mode into the region of absence of dispersion, taking into account the known density of the material, to estimate the dynamic elastic modules. Using the proposed method, the velocities of torsional and rod waves on 20 tubes – billets of cylinders of deep-rod pumps made of 38X2MYA steel were determined.","PeriodicalId":764,"journal":{"name":"Russian Journal of Nondestructive Testing","volume":"62 5","pages":"575 - 583"},"PeriodicalIF":1.3,"publicationDate":"2026-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148823009","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Dual-Branch Framework for Bearing Fault Diagnosis Based on Time-Frequency Analysis and Kolmogorov–Arnold Network 基于时频分析和Kolmogorov-Arnold网络的双分支轴承故障诊断框架

IF 1.3 4区材料科学

Russian Journal of Nondestructive Testing Pub Date : 2026-08-27 DOI: 10.1134/S1061830925605665

Yongjian Li, Yi Gan, Yuyuan Wu, Zhaoyang Chen, Ligang Cheng, Qing Xiong

{"title":"A Dual-Branch Framework for Bearing Fault Diagnosis Based on Time-Frequency Analysis and Kolmogorov–Arnold Network","authors":"Yongjian Li, Yi Gan, Yuyuan Wu, Zhaoyang Chen, Ligang Cheng, Qing Xiong","doi":"10.1134/S1061830925605665","DOIUrl":"10.1134/S1061830925605665","url":null,"abstract":"Time-frequency analysis methods are widely employed for feature extraction. However, most existing approaches depend on fixed basis functions or predetermined window settings, making them inadequate for adapting to the dynamic variations of different signals. Additionally, the linear nature of traditional Fully Connected layers struggles with nonlinear high-dimensional feature mapping. To address these issues, this paper proposes a dual-branch framework based on time-frequency analysis and the Kolmogorov–Arnold Network (DKAN), aiming to overcome the deficiencies of time-frequency analysis in dynamic adaptability and to enhance the nonlinear fitting capacity, thereby improving fault diagnosis accuracy. The dual-branch architecture of the DKAN flexibly integrates time-frequency information with frequency details to fully utilize the complementary strengths. Specifically, the Time-Frequency Enhanced Branch (TEB) is designed to extract and reinforce time-frequency interaction information dynamically, while the Frequency Enhanced Branch (FEB) captures and optimizes fine-grained features in the frequency domain. By jointly leveraging the representations from both branches, DKAN significantly improves the dynamic adaptability and diversity of feature extraction. Moreover, through the incorporation of the KAN layer with nonlinear mapping capabilities, DKAN achieves precise fitting of high-dimensional features and superior fault classification performance compared to Fully Connected layers. Experimental results demonstrate that DKAN attains accuracies of 96.72% and 95.42% on the JNU and Ottawa datasets, respectively, thereby validating the effectiveness and innovation of DKAN and its key modules in fault diagnosis tasks.","PeriodicalId":764,"journal":{"name":"Russian Journal of Nondestructive Testing","volume":"62 5","pages":"599 - 615"},"PeriodicalIF":1.3,"publicationDate":"2026-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148822972","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Slip off Wave Pulse Emitted the Side-Drilled Hole: Experiments and Evaluations 侧钻孔发射的滑移波脉冲：实验与评价

IF 1.3 4区材料科学

Russian Journal of Nondestructive Testing Pub Date : 2026-08-27 DOI: 10.1134/S1061830926601170

V. G. Shevaldykin

引用次数: 0

The Reaction Field of a Uniformly Magnetized Rectangular Parallelepiped Arbitrarily Located in Space 任意位置均匀磁化的矩形平行六面体的反作用力场

IF 1.3 4区材料科学

Russian Journal of Nondestructive Testing Pub Date : 2026-08-27 DOI: 10.1134/S1061830926601169

V. Y. Raevskii, O. V. Kudryashova

引用次数: 0

Application of Poisson’s Ratio for Determining Microdamage in Metastable Austenitic Steel under Fatigue Fracture 用泊松比测定疲劳断裂下亚稳奥氏体钢的微损伤

IF 1.3 4区材料科学

Russian Journal of Nondestructive Testing Pub Date : 2026-08-27 DOI: 10.1134/S1061830926601194

V. V. Mishakin, A. V. Gonchar, A. A. Solovyov, S. A. Sorokina

{"title":"Application of Poisson’s Ratio for Determining Microdamage in Metastable Austenitic Steel under Fatigue Fracture","authors":"V. V. Mishakin, A. V. Gonchar, A. A. Solovyov, S. A. Sorokina","doi":"10.1134/S1061830926601194","DOIUrl":"10.1134/S1061830926601194","url":null,"abstract":"The effect of fatigue fracture on the Poisson’s ratio of metastable austenitic steel 12Kh18N10T (analogous to AISI 321) has been investigated in the low-cycle fatigue (LCF) and high-cycle fatigue (HCF) regions. It is shown that the main degradation processes governing the change in the Poisson’s ratio of the studied steel are microdamage accumulation and strain-induced martensitic transformation. A method is proposed for calculating the microdamage parameter based on the combined measurement of the Poisson’s ratio by the ultrasonic method and the volume fraction of strain-induced martensite by the eddy current method. It is established that the intensity of microdamage accumulation in the LCF region exceeds the corresponding value in the HCF region by an order of magnitude. A non-monotonic behaviour of the Poisson’s ratio has been observed and interpreted at a strain amplitude of (Delta varepsilon {text{/}}2 = 0.2% ) in the high-cycle fatigue region, where martensitic transformation and microdamage accumulation alternately dominate during cyclic loading. A regression expression has been obtained that relates Palmgren–Miner damage to micro-damage and the volume fraction of deformation martensite, characterizing the balance between two competing degradation processes.","PeriodicalId":764,"journal":{"name":"Russian Journal of Nondestructive Testing","volume":"62 5","pages":"693 - 705"},"PeriodicalIF":1.3,"publicationDate":"2026-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148822978","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Research on Crack Estimation Method of Structure based on Pulsed Vibration Excitation Method 基于脉冲振动激励法的结构裂纹估计方法研究

IF 1.3 4区材料科学

Russian Journal of Nondestructive Testing Pub Date : 2026-08-10 DOI: 10.1134/S1061830925605264

Un Hyok Pak, Myong Guk Won, Kyong Chol Kim

{"title":"Research on Crack Estimation Method of Structure based on Pulsed Vibration Excitation Method","authors":"Un Hyok Pak, Myong Guk Won, Kyong Chol Kim","doi":"10.1134/S1061830925605264","DOIUrl":"10.1134/S1061830925605264","url":null,"abstract":"Pulsed vibration excitation is a nondestructive technique that can assess the dynamic characteristics of materials (Young’s modulus, Poisson’s modulus) and the crack characteristics present in the specimen by measuring and analyzing the fundamental resonance frequency of an elastic hammer, and is constantly expanding in engineering practice. In this paper, a mathematical model is proposed to evaluate the crack characteristics present in various types of structural elements based on the excitation of pulsed vibrations with cracks. The relationship between the modal parameters and the crack characteristics is evaluated by applying the concept of modal strain energy. When a crack exists in a structure, the energy released varies with its size and location, which affects the natural vibration mode and natural vibration frequency of the structure. This effect was considered using the commercial software ABAQUS and experimentally verified using a high-temperature elastic modulus measurement device IET-1000 based on pulsed vibration excitation. The results show that the method suggested in this paper can fairly accurately identify the crack characteristics.","PeriodicalId":764,"journal":{"name":"Russian Journal of Nondestructive Testing","volume":"62 4","pages":"529 - 543"},"PeriodicalIF":1.3,"publicationDate":"2026-08-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148699697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Industrial Case Study: Rapid Diagnosis of Flow Maldistribution in an HF Alkylation Column using Radiometric Gamma Scanning 工业案例研究：使用放射伽马扫描快速诊断HF烷基化柱中的流动不均匀

IF 1.3 4区材料科学

Russian Journal of Nondestructive Testing Pub Date : 2026-08-10 DOI: 10.1134/S1061830925606038

Seyedeh Zahra Islami rad, Reza Gholipour Peyvandi, Behnam Sadeghi

{"title":"Industrial Case Study: Rapid Diagnosis of Flow Maldistribution in an HF Alkylation Column using Radiometric Gamma Scanning","authors":"Seyedeh Zahra Islami rad, Reza Gholipour Peyvandi, Behnam Sadeghi","doi":"10.1134/S1061830925606038","DOIUrl":"10.1134/S1061830925606038","url":null,"abstract":"Flow maldistribution in packed columns plays a crucial role in reducing mass transfer efficiency in industrial process equipment in refineries and petrochemicals. In this study, gamma-ray scanning was used as a rapid, non-contact, and cost-effective technique to diagnose hydrodynamic problems in a live industrial HF alkylation rerun column. In this technique, a 137Cs source with 300 mCi activity and a NaI (Tl) detector were positioned on opposite sides of the column. The scanning was performed across the three radial paths (0°, –30°, 30°) within two packed sections, including Nutter Ring Type 1 (NR1) and Type 3 (NR3) packings. According to the results, there is a flow misdistribution and potential local blockage in the NR1 section, which is characterized by a systematic asymmetry of the count rate of 35% and unusual peaks/valleys in the axial profile. In contrast, a uniform radial profile and a smooth axial density gradient were observed in the NR3 section, which indicates healthy operation. Consequently, the gamma radiation scanning technique provides reliable, critical, and actionable data for targeted column maintenance, optimizing performance, and preventing unnecessary shutdowns.","PeriodicalId":764,"journal":{"name":"Russian Journal of Nondestructive Testing","volume":"62 4","pages":"498 - 504"},"PeriodicalIF":1.3,"publicationDate":"2026-08-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148699808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0