YCYu Chen

Yu Chen

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Yu Chen has a total of 38 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Robust Data-Driven Method for Data and Label Noise Mitigation in Synchronous Machine Winding Short-Circuit Fault Diagnosis

    by Yu Chen, Yuxuan Ding, Qianchao Wang, Chakhung Yeung, Chenguo Yao, Yaping Du, Zhongyong Zhao

  • Towards Transformer Winding Deformation Fault Diagnosis under Data Scarcity: Integrating Frequency Response Analysis with Metric Learning

    by Yu Chen, Zhe Chen, Yuxuan Ding, Yaping Du, Binghao Li, Jiangnan Liu, Chakhung Yeung, Chenguo Yao, Zhongyong Zhao

    Accurate diagnosis of transformer winding deformation faults is critical for ensuring the reliability and stability of power systems. However, the scarcity of real-world faulty frequency response analysis (FRA) data severely constrains the development of intelligent data-driven diagnostic models. To overcome this limitation, this study proposes a metric learning framework that integrates FRA with triplet loss and virtual adversarial training (VAT). The method first reformulates the original dataset of N FRA samples into N2 training pairs through the random pairing mechanism, thereby substantially expanding the effective training set. Triplet loss is then applied to further enrich sample diversity (≈ N3) and enhance model feature extraction. To address the inherent label imbalance arising from random pairing, VAT is incorporated. This regularization technique generates synthetic intra-class pairs and enforces local smoothness in the embedding space. Experimental validation on a specially designed 10 kV transformer demonstrates that the proposed approach achieves superior diagnostic accuracy. When applied to two practical transformers with confirmed winding deformations, the trained model accurately identifies fault types. Importantly, the method offers a potential solution for detecting previously unknown winding faults.


  • Miniaturization design study of a dipole magnet applied to proton therapy beamline

    by Xuchuan Liang, Yu Chen, Yicheng Liao, Qushan Chen, Ziyi Yang, Zhenyi Yang, Xu Liu, Bin Qin

    The miniaturization of magnets is crucial for reducing both the physical footprint and the construction cost of a proton therapy facility. The configuration of dipole magnets significantly influences the design of the compact gantry. However, current research demonstrates a degree of subjectivity in achieving an optimal balance between magnet compactness and saturation levels. This study introduces a novel design principle aimed at miniaturizing the yoke of dipole magnets. To be specific, two metrics (i.e. magnet efficiency and magnetization uniformity) are adopted to develop an optimized design that achieves significant weight reduction while maintaining essential magnetic performance characteristics. For the 57-degree dipole magnet discussed in this paper, finite element simulations demonstrate that the optimized design achieves a weight reduction of 39.6% for the yoke (from 5.38 to 3.24 tons) while maintaining transverse and integrated field homogeneity within the stringent [ ] tolerance threshold, and exhibits satisfactory excitation linearity. The design of this study can provide insight for the subsequent construction of a compact gantry.


  • Two-step Broadband Equivalent Circuit Modeling Method for Power Transformer Winding Based on Frequency Response Analysis and Bayesian Optimization

    by Yu Chen, Jiangnan Liu, Yuxuan Ding, Yaping Du, Binghao Li, Chenguo Yao, Zhongyong Zhao

    Frequency response analysis (FRA) is a well-established technique to detect transformer winding deformation faults. Its diagnostic application is based on the principle that a transformer winding can be represented by an equivalent circuit consisting of resistors, inductors, and capacitors. However, rapidly obtaining an accurate and physically meaningful broadband equivalent circuit model for windings remains challenging, limiting both the understanding of fault mechanisms and the generation of data for data-driven fault diagnosis methods. To address these difficulties, this study proposes a two-step broadband equivalent circuit modeling method for the transformer winding based on FRA and Bayesian optimization (BO), considering long-distance mutual inductances and capacitances. The proposed method is validated in a specially designed 10 kV power transformer. Subsequently, two kinds of winding deformation faults, including inter-disk short circuits (IDSCs) and disk space variations (DSVs), are simulated on the basis of the built model and compared with the experimental FRA data. The validation results confirm the accuracy and effectiveness of the proposed method in the construction of the equivalent winding circuit.


  • Data-Efficient Synchronous Machine Winding Short-Circuit Faults Diagnosis Based on Frequency Response Analysis and Active Learning

    by Yu Chen, Zhongyong Zhao, Jiangnan Liu, Wei Wang, Chenguo Yao

    Winding short-circuit (SC) faults are a prevalent issue in synchronous machines, and the accurate and timely identification of these faults is critical for maintaining power system stability. Current methods for inspecting synchronous machine windings often rely on periodic inspections based on human expertise. Numerous studies have explored the application of deep learning (DL) models for detecting synchronous machine winding SC faults. However, these models often exhibit excessive complexity and overlook physical overhead, leading to inefficient utilization of computational power and data resources. To address these limitations, this study proposes a dual-channel deep learning (DL) model integrated with the active learning (AL) query strategy. In this study, winding SC faults are manually simulated on a 5 kVA synchronous machine, and corresponding frequency response analysis (FRA) data are recorded. Subsequently, the proposed method is validated on the test set and benchmarked against previous studies. Experimental results demonstrate that the proposed method significantly reduces the data annotation effort and accelerates model training to the order of seconds (under 5 seconds) while maintaining satisfactory accuracy (≥ 95%). Comparative experimental results further indicate that the proposed model, requiring only 1/20th of the labeled training samples used in previous studies, achieves a substantial reduction of approximately 99.8% in both model parameters and training time.


  • A data-efficient surrogate modeling method for a cyclotron-based proton therapy beamline based on active learning

    by Yu Chen, Bin Qin, Xu Liu, Wei Wang, Yicheng Liao

    High-fidelity beamline models typically involve particle-tracking and particle–matter interaction, which are intensive computationally demanding and time-consuming. This has led researchers to adopt data-driven surrogate models as an alternative to complex physics simulations. Training a data-driven model requires many labeled data, prompting researchers to simulate diverse beamline settings and obtain corresponding labels. However, the required dataset grows increasingly large as the number of adjustable parameters increases. Therefore, this study proposes a data-efficient surrogate modeling method that employs a student–teacher framework combined with active learning (AL) query strategies to minimize labeled samples while ensuring model accuracy. The proposed method is evaluated on the energy selection system (ESS) design of the Huazhong University of Science and Technology proton therapy facility (HUST-PTF). The results show that: (i) Training the surrogate model using 684 labeled samples selected via query strategy achieves a relative error below 5% for 90% of the samples in the test set. (ii) Compared to the beamline model built by Beam Delivery Simulation (BDSIM), the computational efficiency of the surrogate model is enhanced by a factor of [ ].


  • Improved Interpretation of Impulse Frequency Response Analysis for Synchronous Machine Using Life long Learning Based on iCaRL

    by Yu Chen, Zhongyong Zhao, Yueqiang Yu, Yunlong Guo, Chao Tang

    Winding inter-turn and ground short circuit (SC) faults are common fault types of synchronous machines. The sweep frequency response analysis (SFRA) has been recently introduced to detect winding SC faults. Additionally, impulse FRA (IFRA) provides an alternative that serves the same objective. Whether SFRA or IFRA, so far, there is still no standard and reliable interpretation code. The present interpretation of frequency response still calls for experienced personnel, which could be subjective. Thus, many researchers have used machine or deep learning-based models to detect winding SC faults from the frequency response automatically. However, most proposed models cannot provide real-time feedback and update. Therefore, this study proposes an improved model using life long learning strategy based on incremental classifier and representation learning (iCaRL) to interpret and analyze the IFRA curves. This study artificially simulates and records winding SC faults on a 5 kW synchronous machine. The proposed method is then verified on the test set and compared with other traditional life long learning strategies. The experimental results show that the accuracy of the proposed model is higher than 90% under all types of fault data streams in real-time. The comparative experimental results show that the proposed model performs better than other life long learning strategies. (Github code: https://github.com/cy1034429432/Impro ved-Interpretation-of-IFRA-based-on-iCaRL/tree/main).


  • Understanding IFRA for Detecting Synchronous Machine Winding Short Circuit Faults Based on Image Classification and Smooth Grad-CAM++

    by Yu Chen, Zhongyong Zhao, Yueqiang Yu, Wei Wang, Chao Tang

    Fault diagnosis of the synchronous machine is essential for the safe operation of the modern power system. Recently, impulse frequency response analysis (IFRA) has been used to detect synchronous machine winding short circuit (SC) faults since the failure will significantly alter the IFRA signature of machines. However, there is no standard and reliable code for interpreting the IFRA signatures. It often refers to the frequency response analysis of power transformer standards, the mathematical-statistical indicator of relative factor is directly used to perform the judgment by the standards’ threshold, which might not be appropriate. Besides, the internal mechanism of frequency response analysis diagnosis is unclear. Therefore, this study proposes a technique based on image classification and smooth grade and gradient-based class activation maps (Smooth Grad-CAM++) to understand and interpret the IFRA method. It trains and analyzes the visualization results of the fault data set of a 5kW synchronous machine’s winding. The experimental results show that the average accuracy of the image classification model based on Resnet18 reaches 99.63%. We performed an IFRA difference analysis according to the visualization results and gave some suggestions about IFRA. The training and detection process can be accelerated based on these suggestions. To illustrate the generalization of the suggestions, another 8kW synchronous machine is used for the case study, and the experimental results show that these suggestions are still effective. The main contributions of this study are understanding the internal mechanism of IFRA diagnosis from another perspective and providing some conclusions different from previous research results on the IFRA method for the synchronous machine.