Conference Info
Invited Speaker of Parallel Session for Online Non-destructive Testing Monitoring Technology Equipment at the 2026 FENDT Forum
Time：2026-06-07
Electromagnetic Acoustic Testing Technology and Instrumentation
Jinjie Zhou
High-temperature equipment such as pressure vessels, utility boilers, and pressure pipelines suffer from diverse structural configurations, harsh operating conditions, and severe obstructions, leading to significant acoustic energy attenuation, strong interference signals, and limited inspection accuracy. To address these challenges, our team has conducted systematic research on the excitation and reception characteristics of electromagnetic acoustic transducers (EMATs) and guided waves, theoretical modeling and sound field computation of transducers, segmented/fused time-reversal inspection methods, B-scan/SAFT/TFM imaging techniques for high-temperature pipelines, and key technologies for high-energy excitation and reception instrumentation. A series of equipment has been developed, including high-temperature EMATs and guided-wave sensors, EMAT flaw detectors, EMAT guided-wave flaw detectors, and scientific experimental instruments for EMATs and guided waves featuring arbitrary waveform excitation. For high-temperature inspection scenarios, circumferential/axial automatic scanning devices have been designed and implemented, and successful high-temperature online inspection engineering applications have been accomplished at multiple industrial sites.
Biography of Jinjie Zhou
Dr. Jinjie Zhou is a Professor and Doctoral Supervisor at the School of Mechanical Engineering, North University of China. He serves as the Director of the Shanxi Key Laboratory of Smart Manufacturing and Intelligent Testing for Pipeline and Hole Structures in Harsh Environments. His research interests include novel nondestructive testing methods and instrumentation, intelligent inspection robots, and automated measurement and control equipment. To date, he has published over 80 peer-reviewed papers, been granted more than 30 invention patents, authored one monograph, and contributed to the formulation of five national standards. He has led over 10 major research projects, including those funded by the National Natural Science Foundation, the National Key Research and Development Program, and defense pre-research programs. Dr. Zhou is recognized as a Leading Talent under the “Sanjin Talent” Program of Shanxi Province and a Top-Notch Young Innovative Talent of Shanxi Higher Education Institutions. He has received one first-class provincial/ministerial science and technology award and two first-class science and technology awards from industry associations. His applied research addresses critical engineering challenges in harsh-environment inspection and intelligent manufacturing, with particular emphasis on high-temperature online inspection of pressure vessels, utility boilers and pressure pipelines using electromagnetic acoustic transducers, development of automated scanning devices for complex structures, and intelligent detection instrumentation for industrial field applications.
Anomaly Detection and Intelligent Prognosis for Critical Industrial Equipment: A Data-Driven Approach with Deep Transfer Learning
Wei Guo
High-end critical equipment constitutes the foundational infrastructure of industrial production and intelligent manufacturing systems; its operational stability and functional safety are pivotal determinants of production efficiency and maintenance costs. Fault detection and remaining useful life (RUL) prediction remain persistently challenging: data-driven models demand extensive high-fidelity labeled datasets, exhibit poor generalization across varying operating conditions and equipment variants, and are highly sensitive to measurement noise and disturbances in real-world industrial environments. Critically, failure data for critical industrial equipment is inherently scarce and highly imbalanced. To address these challenges, this study introduces data intelligence-driven and knowledge transfer Learning and constructs a deep learning-based framework for anomaly detection and fault prognosis in critical industrial assets. By transferring fault-relevant knowledge from data-rich source domains to data-scarce target monitoring systems, the framework enables early identification of incipient anomalies and accurate RUL prediction. Meanwhile, it effectively addresses the challenges of training under data-scarce scenarios and low detection accuracy for new equipment and special operating conditions, greatly improves detection sensitivity for incipient anomalies, enhances cross-scenario generalization, and overcomes technical bottlenecks of single modeling approaches and practical deployment constraints of resource-intensive deep learning architectures. The related technology can be applied to critical components in rotating machinery and machining processes, leveraging the advantages of artificial intelligence and fault prediction. It promotes the practical implementation of anomaly detection and prognostics theories that integrate deep learning-based data fusion and cross-domain transfer, providing feasible technical solutions and engineering references for the large-scale deployment of industrial intelligent maintenance technologies.
Biography of Wei Guo
Dr. Guo Wei is an Associate Professor at the School of Mechanical and Electrical Engineering, University of Electronic Science and Technology of China. Her research interests include condition monitoring and fault diagnosis of mechanical systems, intelligent fault diagnosis and remaining useful life prediction, multi-source data analysis and intelligent operation & maintenance, etc. To date, she has authored or co-authored 66 peer-reviewed journal and conference papers, accumulating 1,215 SCI citations. She has led or contributed significantly to over 20 research and industrial application projects, securing total funding exceeding RMB 5 million. Dr. Guo serves on the Editorial Board of ‘Measurement Science and Technology’, a Senior Member of the IEEE, a Standing Committee Member of the Southwest Branch of the Dynamic Testing Committee, China Society of Vibration Engineering, and has been recognized as Outstanding Reviewers by both IEEE Transactions on Instrumentation and Measurement and Measurement Science and Technology. His applied research addresses critical engineering challenges across multiple high-reliability domains, including vibration-based health assessment of nuclear power plant rotating equipment, real-time vibration monitoring systems for nuclear piping networks, multi-sensor signal acquisition and analyses, condition monitoring of oil sands pump impellers, vibration characterization and diagnostics of key wind tunnel equipment, and heterogeneous data fusion from CNC machines, etc.
Research on 3D Measurement of Deep-Hole Inner Surfaces and Robotic Technology
Xiang Li
Addressing the technical challenges associated with the 3D topography measurement of ultra-long and narrow-diameter pipeline inner walls, this report introduces a multi-sensor measurement method that integrates structured light vision, panoramic imaging, and synchronous scanning technologies. This approach enables online 3D measurement of such inner surfaces based on structured light vision principles. The main research contents include: the design of annular single-camera/multi-camera structured light sensors and the establishment of their 3D measurement mathematical models; the optimal configuration between the sensor and the scanning drive system; calibration methods for sensor measurement model parameters, along with high-precision and robust algorithms for image feature extraction; the cross-modal fusion mechanism between annular structured light sensors and multi-view vision sensors; and the demand analysis of a motion platform for deep-hole inspection, as well as the development of a prototype 3D measurement robotic system. The deep integration of 3D measurement technology with robotic platforms is driving the evolution of deep-hole inner wall topography measurement from "mechanized execution" toward an intelligent inspection paradigm that features "perception-motion-decision" integration. In line with the development trend of embodied intelligence, deep-hole inspection robots equipped with environmental perception and autonomous operation capabilities are expected to demonstrate broad application prospects in the inspection, maintenance, and operation of key equipment involving complex piping systems and non-circular cross-section deep holes.
Biography of Xiang Li
Li Xiang, Ph.D., Associate Professor, School of Aeronautics and Astronautics, University of Electronic Science and Technology of China (UESTC). Dr. Li is engaged in research on nondestructive testing methods, with primary research interests including structural fatigue state detection and evaluation, stress detection, defect detection, digital twin-based monitoring and inspection technologies and corresponding data processing methods, as well as specialized robotics for deep-hole inspection, among others. He has served as principal investigator for five national-level scientific research projects, including the National Natural Science Foundation of China (NSFC) and major projects commissioned by a ministry or commission. He has also participated as a core member in key research and development programs. He is a recipient of the Third Prize of Sichuan Provincial Science and Technology Progress Award and has been recognized for a Sichuan Provincial First-Class Undergraduate Course. He has published over 20 academic papers and participated in the formulation of two standards.
In-service Non-stop Inspection and Testing Technology and Application
Huan Li
In-service non-stop inspection and testing represents a key future direction in the field of special equipment inspection. This technology enables inspection operations under continuous industrial operation conditions, efficiently performing critical tasks such as material performance analysis, wall thickness reduction assessment, corrosion status evaluation, crack defect identification, and leakage detection. This paper reviews advanced in-service non-stop inspection techniques—including high-precision nano-indentation technology, magnetically induced acoustic stress testing, electromagnetic ultrasonic guided wave technology, array eddy current testing, and alternating current electromagnetic field detection—while detailing their respective application characteristics and limitations. Furthermore, through case studies from practical applications in pressure vessels and pipelines, the paper examines the practical effectiveness and value of these technologies in special equipment inspection practices.
Biography of Huan Li
Li Huan, Chairman of Anhui Huaxia High-Tech Development Co., Ltd., currently serves as Deputy Director of the Non-Destructive Testing Working Committee and Member of the Testing Technology Application and Evaluation Working Committee of the China Special Equipment Inspection Association; Vice Chairman of the Non-Destructive Testing Special Committee of the Anhui Mechanical Engineering Society; and a member of the expert database of the Anhui Provincial Market Supervision Administration. With long-term dedication to the field of special equipment safety, he has specialized in non-destructive testing for nearly three decades, having led or participated in over 20 key national research projects related to special equipment. He has received the Third Prize of Jiangxi Provincial Science and Technology Progress Award, published more than 10 academic papers, contributed to or edited 15 national, industry, and group standards, and holds over 30 patents.
Research on Residual Stress Detection Method Based on Electromagnetic Acoustic Surface Waves and Deep Learning
Binpeng Zhang
Residual stress directly determines the service life and safety of large steel structures, making its high-precision nondestructive evaluation of great engineering significance. Conventional piezoelectric ultrasonic testing relies heavily on couplants and imposes strict requirements on surface finish, making it poorly suited for complex industrial environments. Electromagnetic acoustic techniques offer a non-contact approach, yet suffer from low transduction efficiency and susceptibility to environmental interference. To address these issues, this paper integrates multiphysics simulations and deep learning to develop an electromagnetic acoustic surface wave method for residual stress detection, proposing an intelligent approach with high accuracy and strong generalization.
First, the excitation mechanism and propagation behavior of surface waves in metals were analyzed through steady-state and transient multiphysics simulations. An optimized separated dual-magnet transducer configuration was designed and validated, achieving a 20%–30% increase in received signal amplitude. At a center frequency of 1.8 MHz, the optimal spatial period of the meander coil was determined to be 1.5 mm, the maximum lift-off distance 0.8 mm, and the minimum specimen thickness 3.0 mm, laying a foundation for acquiring high signal-to-noise ratio signals.
Second, an experimental system was established, verifying the simulated attenuation of signal amplitude with increasing lift-off and identifying the optimal lift-off distance as 0.5 mm. To overcome the failure of the conventional cross-correlation algorithm in extracting time-of-flight from rough surfaces, a Gaussian envelope peak method combined with a second-order nonlinear compensation model was introduced, reducing the time-of-flight error from 127.2 ns to 10.5 ns and effectively mitigating roughness interference. Furthermore, the time-of-flight versus stress curve and acoustoelastic coefficient of Q235 steel were calibrated through uniaxial tensile tests, providing a real dataset and a baseline for the traditional calibration method.
Finally, breaking away from the limitations of classical acoustoelastic theory, a deep learning-based stress prediction scheme was proposed. A hybrid neural network integrating a one-dimensional convolutional neural network, bidirectional gated recurrent units, and an attention mechanism was constructed. Simulation and experimental data covering the 0–230 MPa range were collected, and a probabilistic combination data augmentation strategy was employed to address the scarcity of training samples. Test set results show a mean absolute error of only 0.40 MPa and a maximum error strictly within 0.98 MPa, representing an error reduction of approximately 97.8% compared with the 18.34 MPa mean absolute error of the traditional calibration method. In generalization experiments with unseen stress data, the incorporation of the attention mechanism dramatically lowered the prediction error from 4.37 MPa to 0.90 MPa, demonstrating excellent generalization capability and offering a novel pathway for intelligent residual stress detection.
Biography of Binpeng Zhang
Binpeng Zhang, Senior Engineer, China Special Equipment Inspection and Research Institute (CSEI). In recent years, his research has primarily focused on new techniques in ultrasonic nondestructive testing (NDT), involving stress detection and lithium-ion battery inspection. He has proposed several advanced fundamental theories in ultrasonics and developed multiple novel ultrasonic NDT methods. He has published over 10 academic papers in internationally renowned journals such as the Journal of Power Sources, Ultrasonics, Applied Acoustics, and Sensors and Actuators A: Physical (including 7 SCI-indexed papers as first author). He has filed over 10 national patents and been granted 5. As a project core member, he has participated in 2 National Science and Technology Major Projects, 1 National Key Research and Development Program of China, 1 Key Project of the National Natural Science Foundation of China (NSFC), 2 General Projects of the NSFC, and 1 Key Project jointly funded by the Beijing Municipal Natural Science Foundation and the Beijing Municipal Commission of Education. He has contributed to the development of 2 national standards for NDT (English version) and 1 CSTM group standard. He currently serves as a member of the Acoustic Instruments Technical Committee of the China Instrument and Control Society, as Secretary of the Intelligent Testing Technical Committee under the NDT Technology and Equipment Standardization Field Committee, as a Youth Editorial Board Member of the Journal of Measurement Science and Instrumentation (JMSI), and as an adjunct master's supervisor at North University of China.
The electromagnetic eddy current high-sensitivity detection and defect assessment technology driven by a regulated circuit
Qiuping Ma
Eddy current sensors play an important role in equipment inspection and structural health monitoring, where improving their response to micro-defects is critical for achieving high-sensitivity detection and reliable evaluation. Traditional studies on eddy current sensors have mainly focused on optimizing coil structures, magnetic field distributions, and array configurations, while relatively less attention has been paid to the role of conditioning circuits in signal enhancement, noise suppression, and feature regulation. This report focuses on the requirements of sensitivity enhancement and defect evaluation in eddy current testing. A systematic analysis is conducted from two aspects: sensor structural design and conditioning circuit optimization. The significant influence of conditioning circuits on signal quality, system stability, and defect response characteristics is clarified. Furthermore, different conditioning circuit configurations can modify the amplitude, phase, frequency response, and impedance matching characteristics of the detected signals, thereby affecting the extraction and representation capability of defect-related features. Therefore, incorporating conditioning circuits as an essential part of the collaborative design of eddy current sensors is of great significance for improving micro-defect detection sensitivity, enhancing the reliability of defect identification, and enabling quantitative evaluation.
Biography of Qiuping Ma
Dr. Qiuping Ma is a Senior Engineer at the China Special Equipment Inspection and Research Institute, specializing in electromagnetic nondestructive testing and intelligent sensing instrumentation. Her research focuses on advanced electromagnetic sensing mechanisms, high-sensitivity defect detection, and intelligent inspection systems for engineering structures. She has authored more than 30 peer-reviewed papers in leading international journals, including IEEE Transactions on Industrial Informatics, IEEE Transactions on Industrial Electronics, and NDT & E International. She has led research projects funded by the National Natural Science Foundation of China, the China Postdoctoral Science Foundation, and the Sichuan Provincial Science and Technology Innovation Talent Program. She has also served as a key contributor to a National Key R&D Program of China and has participated in multiple industry-sponsored projects. Beyond her research achievements, she has contributed to the development of one group standard and holds multiple invention patents, including seven Chinese invention patents filed or granted and one granted U.S. invention patent. She actively serves the academic community as a reviewer for several international journals, including IEEE Sensors Journal, Measurement, Scientific Reports, Nondestructive Testing and Evaluation, Results in Engineering, and Journal of Pipeline Science and Engineering. She is also a Topic Advisory Panel member of Sensors and a young editorial board member of the Journal of Measurement Science and Instrumentation.
