Author(s)

WenMing Chen¹, Xian Zhou^1*, YunTao Yang²

Affiliation(s)

¹Hunan Electrical College of Technology, Xiangtan 411101, Hunan, China.

²School of Physics & Electronics, Hunan University, Changsha 410082, Hunan, China.

Corresponding Author

Xian Zhou

ABSTRACT

To address the challenge of fault diagnosis in elevators caused by limited sample data, this paper proposes a few-shot fault diagnosis method based on domain adaptive transfer learning. By constructing a feature extraction network incorporating multi-scale convolution and attention mechanisms, combined with a domain adaptation module that aligns both marginal and conditional distributions, and introducing meta-learning and data augmentation strategies, the diagnostic capability of the model under few-shot conditions in the target domain is effectively improved. Experimental results demonstrate that the proposed method outperforms traditional diagnostic models in terms of accuracy and cross-domain transfer performance, showing promising potential for practical engineering applications. This study provides an effective solution for few-shot fault diagnosis in elevators, contributing both theoretical insights and practical value to enhancing elevator operational safety.

KEYWORDS

Elevator fault diagnosis; Few-shot learning; Transfer learning; Adversarial training; Feature extraction

CITE THIS PAPER

WenMing Chen, Xian Zhou, YunTao Yang. Model transfer for few-shot fault diagnosis of elevators based on domain adaptation. Journal of Computer Science and Electrical Engineering. 2025, 7(7): 25-37. DOI: https://doi.org/10.61784/jcsee3098.

REFERENCES

[1] Qing Guangwei, Liu Xiaofan. Research on Prediction Method of City Elevator Entrapment Fault Causes Based on Machine Learning. Internet of Things Technologies, 2022, 12(10): 55–58.

[2] Qi Yongsheng, Shan Chengcheng, Gao Shengli, et al. Fault diagnosis strategy for wind turbine bearing based on AEWT-KELM. Acta Energiae Solaris Sinica, 2022, 43(8): 281–291.

[3] Zhao Wenqiang, Zhou Jun, Wang Zhengwei, et al. Fault diagnosis method of synchronous condenser used in UHV transmission system. Science Technology and Engineering, 2024, 24(9): 3683–3690.

[4] Jiang Wanlu, Zhao Yan, Li Zhenbao. Fault diagnosis method for rotating machinery based on multi-model stacking ensemble learning. Chinese Hydraulics & Pneumatics, 2023, 47(4): 46–58.

[5] Zhao Bochao, Ma Jiajun, Cui Lei, et al. Photovoltaic anomaly detection based on improved VMD-XGBoost-BILSTM hybrid model. Computer Engineering, 2024, 50(3): 306–316.

[6] Liu Y, Jiang H, Yao R, et al. Counterfactual-augmented few shot contrastive learning for machinery intelligent fault diagnosis with limited samples. Mechanical Systems and Signal Processing, 2024, 216: 111507.

[7] Chen Yuanqiong, Meng Yujia, Li Zhihao. Research on Distributed Fault Diagnosis System Based on Machine Learning. Computer Knowledge and Technology, 2024, 20(03): 22–24.

[8] Song Zhangting, Liu Yang, Guo Liang. Research on Causes and Solutions of Vertical Elevator Faults. China Plant Engineering, 2023(20): 170–173.

[9] Zhu Wenhua, Zuo Yi. Design of Elevator Control System Based on S7-1500PLC. Machine Building & Automation, 2023, 52(06): 182–186.

[10] Bi Lilong. Research on Fault Detection Method of Elevator Brake Based on Machine Vision. China Machinery, 2023(10): 116–119.

[11] Li Guodong. Discussion on Fault Diagnosis and Maintenance of Elevator Electrical Control System. Modern Industrial Economy and Informatization, 2023, 13(02): 246–248.

[12] Wu Cong, Li Mengnan, Li Kun. Elevator Bearing Fault Detection Based on PBO and CNN. Information Technology, 2023, 47(04): 73–78.

[13] Niu Dapeng, Guo Lei, Zhang Weiwei, et al. Operation performance evaluation of elevators based on condition monitoring and combination weighting method. Measurement, 2022, 194(1): 13–17.

[14] Ali Murad, Din Zakiud, Solomin Evgeny, et al. Open switch fault diagnosis of cascade H-bridge multilevel inverter in distributed power generators by machine learning algorithms. Energy Reports, 2021, 7(1): 13–20.

[15] Chen Zhiyu. An Elevator Car Vibration Fault Diagnosis Based on Genetic Optimization RBF Neural Network. China Science and Technology Information, 2023(10): 110–115.

[16] Feng Yongming. Research on Data Mining of Elevator Safety Big Data Based on Hadoop. Xi'an: Xi'an University of Science and Technology, 2023.

[17] Meng Lin, Wang Xiaoyang, Guo Qingliang. Analysis of Related Technical Issues in Elevator Fault Detection. Engineering Technology Research, 2020, 5(7): 62–63.

[18] Zhang Zhanyi, Zhang Baoquan, Wang Zhouli, et al. Data Augmentation Optimization for Multi-Tea CNN Image Recognition and Quantitative Evaluation of Class Activation Mapping. Journal of Tea Science, 2023, 43(3): 411–423.