1.State Key Laboratory of Reliability and Intelligence of Electrical Equipment, Hebei University of Technology, Tianjin 300401, China
2.Tianjin Keyvia Electric Co. , Ltd. , Tianjin 300392, China
3.Department of Building Environment and Energy Engineering, The Hong Kong Polytechnic University, Hong Kong 000000, China
4.Key Laboratory of Electromagnetic Field and Electrical Apparatus Reliability of Hebei Province, Hebei University of Technology, Tianjin 300401, China
Aiming at the current situation of long-time consuming and expert-experienced fault diagnosis of relay protection devices operation in rail transit feeder systems, a fault diagnosis model named SG-CSSA-1DCNN is proposed based on the Synthetic Minority Over-sampling Technique enhanced by Generative Adversarial Network (SMOTE-GAN) and one-Dimensional Convolutional Neural Network (1DCNN) optimized by the Combined Sparrow Search Algorithm (CSSA). Firstly, the local ideal minority samples are generated by SMOTE as the input of the GAN generator. Then, the local interpolation advantage of SMOTE and the global distribution learning ability of GAN are integrated to solve problems of insufficient original samples and low quality of the generated samples. Secondly, the CSSA algorithm that introduces Tent chaotic sequences and a Gaussian variation mechanism is adopted to improve global optimization efficiency, the automatic search of optimal hyperparameters for 1D-CNN is realized, and the classification performance of the model is optimized. Finally, the fault diagnosis model is constructed based on an actual dataset of 9 types of faults including 18 electrical features. The results indicate that in comparison with the original 1D-CNN model, the loss of the optimized model is reduced by 12.5%, the diagnostic accuracy of it is enhanced to 98.46%, and the classification accuracy of 9 types of faults reaches equilibrium. The method is an effective solution to the problem of fault identification in cases of data with unbalanced category, and can significantly improve the reliability of fault identification in relay protection devices operation.
HUIJizhuang, ZHANGZeyu, YEMin, et al. Review on Digital Twin Technology for Highway Construction and Maintenance Equipment [J]. Journal of Traffic and Transportation Engineering, 2023, 23 (4): 23-44. in Chinese
[3]
FENGD, SUNX J, SHANGC, et al. Cost-Effectiveness Oriented Intelligent Maintenance Scheduling Optimization for Traction Power Supply System of High-Speed Railway [J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23 (12): 23179-23193.
XIONGJiayang, SHENZhiyun. Rise and Future Development of China's High-Speed Railway [J]. Journal of Traffic and Transportation Engineering, 2021, 21 (5): 6-29. in Chinese
LUOJia, HUANGJinying, MAJiancheng, et al. Fault Diagnosis Method of Plunger Pump for Railway Switch Machine Based on C-DCGAN [J]. China Railway Science, 2024, 45 (6): 111-120. in Chinese
CHAILinguo, ZHANGJinghui, SHANGGUANWei, et al. Fault Classification and Diagnosis Method for CBTC on-Board Signal Equipment Based on Doc2vec-LightGBM [J]. Journal of the China Railway Society, 2024, 46 (4): 108-118. in Chinese
LIXiayang, LIUChang, YANGXiaofeng, et al. A Fault Diagnosis Strategy for Track Circuits Based on Data-Knowledge Collaboratively Driven [J]. Journal of Railway Science and Engineering, 2024, 21 (12): 5276-5288. in Chinese
ZHOULujie, DANGJianwu, WANGYuxin, et al. Research on Fault Classification Method for Onboard Equipment of Train Control System Based on Convolutional Neural Network [J]. Journal of the China Railway Society, 2021, 43 (6): 70-77. in Chinese
[14]
FAN S KS., CHENGC W, TSAID M. Fault Diagnosis of Wafer Acceptance Test and Chip Probing between Front-End-of-Line and Back-End-of-Line Processes [J]. IEEE Transactions on Automation Science and Engineering, 2022, 19 (4): 3068-3082.
[15]
CHATTERJEES, BYUNY C. Highly Imbalanced Fault Classification of Wind Turbines Using Data Resampling and Hybrid Ensemble Method Approach [J]. Engineering Applications of Artificial Intelligence, 2023, 126: 107104.
[16]
GAMELS A, GHONEIMS S M, SULTANY A. Improving the Accuracy of Diagnostic Predictions for Power Transformers by Employing a Hybrid Approach Combining SMOTE and DNN [J]. Computers and Electrical Engineering, 2024, 117: 109232.
[17]
NAHELIYAB, KUMARK, REDHUP. A Deep Extreme Learning Machine Approach Optimized by Sparrow Search Algorithm for Forecasting of Traffic Flow [J]. Physica Scripta, 2024, 99 (12): 125288.
WANGYankuai, MENGJiadong, ZHANGYu, et al. Turnout Fault Diagnosis Method Based on GADF and 2D CNN-Improved SVM [J]. Journal of Railway Science and Engineering, 2024, 21 (7): 2944-2956. in Chinese
[20]
LIL L, XIONGJ L, TSENGM L, et al. Using Multi-Objective Sparrow Search Algorithm to Establish Active Distribution Network Dynamic Reconfiguration Integrated Optimization [J]. Expert Systems with Applications, 2022, 193: 116445.
[21]
HOSSEINZADEHM, ALI S, RAHMANIA M, et al. A Smart Filtering-Based Adaptive Optimized Link State Routing Protocol in Flying Ad Hoc Networks for Traffic Monitoring [J]. Journal of King Saud University-Computer and Information Sciences, 2024, 36 (4): 102034.
[22]
JUZ D, CHENY S, QIANGY K, et al. A Systematic Review of Data Augmentation Methods for Intelligent Fault Diagnosis of Rotating Machinery under Limited Data Conditions [J]. Measurement Science and Technology, 2024, 35 (12): 122004.
[23]
ALZUBAIDIL, BAIJ S, AL-SABAAWIA, et al. A Survey on Deep Learning Tools Dealing with Data Scarcity: Definitions, Challenges, Solutions, Tips, and Applications [J]. Journal of Big Data, 2023, 10 (1): 46.
[24]
LIUD, ZHONGS S, LINL, et al. Feature-Level SMOTE: Augmenting Fault Samples in Learnable Feature Space for Imbalanced Fault Diagnosis of Gas Turbines [J]. Expert Systems with Applications, 2024, 238: 122023.
[25]
WANGZ Y, LIUT, WUX, et al. Application of an Oversampling Method Based on GMM and Boundary Optimization in Imbalance-Bearing Fault Diagnosis [J]. IEEE Transactions on Industrial Informatics, 2024, 20 (2): 1931-1940.
SONGJindong, LUANShicheng, LIShanyou, et al. Interference Signal Identification Method of Earthquake Early Warning for High-Speed Railway Based on Generative Adversarial Network and Convolutional Neural Network [J]. China Railway Science, 2025, 46 (1): 225-232. in Chinese
[28]
GHAREHCHOPOGHF S, NAMAZIM, EBRAHIMIL, et al. Advances in Sparrow Search Algorithm: a Comprehensive Survey [J]. Archives of Computational Methods in Engineering, 2023, 30 (1): 427-455.
ZHANGJunxiao, GAOChong, LAOYongzhao, et al. Fault Recovery Method for Distribution Network Based on Chaotic Binary Sparrow Search [J]. High Voltage Engineering, 2023, 49 (): 247-253. in Chinese
ZHOUHuan, CHENJianyun, WANRuoan, et al. Multi-Signal Fusion Fault Location of All Parallel AT Traction Network Based on TCNN-MADLSTM [J]. China Railway Science, 2023, 44 (4): 206-218. in Chinese
WANGWeidong, ZHANGChenlei, HUWenbo, et al. Fine-Grained Measurement of Ballastless Track Slab Cracks Based on Improved Faster R-CNN and Orthogonal Projection [J]. China Railway Science, 2023, 44 (6): 46-56. in Chinese
[35]
YET, ZHANGJ, ZHAOZ Y, et al. Foreign Body Detection in Rail Transit Based on a Multi-Mode Feature-Enhanced Convolutional Neural Network [J]. IEEE Transactions on Intelligent Transportation Systems, 2022, 23 (10): 18051-18063.
[36]
XUY, ZHAOY, KEW, et al. A Multi-Fault Diagnosis Method Based on Improved SMOTE for Class-Imbalanced Data [J]. The Canadian Journal of Chemical Engineering, 2023, 101 (4): 1986-2001.