Research on Dynamic Fault Injection of Bearing Digital Twin Model Based on Improved Dung Beetle Optimization Algorithm

doi:10.3901/JME.260097

Abstract

Abstract: In practical applications, the lack of fault samples and the imbalance of fault sample categories often limit the effectiveness of diagnosis. In order to solve the problem of unbalanced bearing fault samples, a digital twin system based on dynamic fault injection is designed, which can not only reflect the normal operation state, but also generate samples of different faults. The system is divided into an agent model part and a parameter identification part: the agent model part adopts a two-freedom bearing dynamics model of the impact response at the fault point, and uses the fourth-order variable-step-length Lunger-Kutta method to perform simulation calculations and generate the simulated vibration data; the parameter identification part combines the original algorithm with the Chebyshev chaotic mapping, the golden sinusoidal strategy, and the adaptive weighting factor method and improves its fitness function to propose an improved dung beetle optimisation algorithm, which can identify the dynamic parameters through the measured vibration data, construct a digital twin system, and realise dynamic fault injection. It is proved through experiments that the fault data generated by this system has a higher accuracy rate compared with the data generated using inverse physical information neural network, CycleGAN and GAN. In addition, after expanding the outer ring fault data by this system, the overall diagnostic accuracy of the diagnostic model is increased by 8.9%, and the diagnostic accuracy of the outer ring fault is increased by 26.7% to 99.8% and 99.5%, respectively, which provides a certain reference for the problem of unbalanced fault samples in the fault diagnosis of rolling bearings.

Key words: digital twin, bearing dynamics model, improved dung beetle optimisation algorithm, dynamic fault injection, sample imbalance

CLC Number:

TH133.33

LIU Suyan, DONG Yilin, QIAO Yiming, MA Zengqiang. Research on Dynamic Fault Injection of Bearing Digital Twin Model Based on Improved Dung Beetle Optimization Algorithm[J]. Journal of Mechanical Engineering, 2026, 62(3): 446-457.

References

[1] 王曦，侯宇，孙守光，等. 高速列车轴承可靠性评估关键力学参量研究进展[J]. 力学学报，2021，53(1)：19-34. WANG Xi，HOU Yu，SUN Shouguang，et al. Research progress on key mechanical parameters for reliability evaluation of high-speed train bearings [J] Journal of Mechanics，2021，53(1)：19-34.
[2] 阳建宏，黎敏，丁福焰，等. 滚动轴承诊断现场实用技术[M]. 机械工业出版社，2015. YANG Jianhong，LI Min，DING Fuyan，et al. Practical on-site diagnostic technology for rolling bearings [M]. China Machine Press，2015.
[3] 郭敏，陈鹏，周超，等. 基于坐标注意力关系网络的小样本轴承故障诊断[J]. 四川大学学报，2024，61(4)：338-346. GUO Min，CHEN Peng，ZHOU Chao，et al. Small sample bearing fault diagnosis based on coordinate attention relationship network [J]. Journal of Sichuan University，2024，61(4)：338-346.
[4] HOU Wenbo，ZHANG Chunling，JIANG Yunqian，et al. A new bearing fault diagnosis method via simulation data driving transfer learning without target fault data[J]. Measurement，2023，215：112879.
[5] PAN Haiyan，YIN Xueling，CHENG Jinan，et al. Periodic component pursuit-based kurtosis deconvolution and its application in roller bearing compound fault diagnosis[J]. Mechanism and Machine Theory，2023，185：105337.
[6] DAI Peng，LIANG Xingyu，WANG Kun，et al. Dynamic modeling and vibration analysis for the gear-rotor-bearing system with localized failures[J]. Mechanical Systems and Signal Processing，2023，200：110551.
[7] LI Xin，LI Shuhua，WEI Dong，et al. Dynamics simulation-driven fault diagnosis of rolling bearings using security transfer support matrix machine[J]. Reliability Engineering & System Safety，2024，243：109882.
[8] MA Leiming，JIANG Bin，XIAO Lingfei，et al. Digital twin-assisted enhanced meta-transfer learning for rolling bearing fault diagnosis[J]. Mechanical Systems and Signal Processing，2023，200：110490.
[9] SHI Huaitao，SONG Zelong，BAI Xiaotian，et al. A novel digital twin model for dynamical updating and real-time mapping of local defect extension in rolling bearings[J]. Mechanical Systems and Signal Processing，2023，193：110255.
[10] FENG Ke，XU Yadong，WANG Yulin，et al. Digital twin enabled domain adversarial graph networks for bearing fault diagnosis[J]. IEEE Transactions on Industrial Cyber-Physical Systems，2023(1)：113-122.
[11] QIN Yi，LIU Hongyu，MAO Yongfang. Faulty rolling bearing digital twin model and its application in fault diagnosis with imbalanced samples[J]. Advanced Engineering Informatics，2024，61：102513.
[12] WANG Hui，ZHENG Junkang，XIANG Jiawei. Online bearing fault diagnosis using numerical simulation models and machine learning classifications[J]. Reliability Engineering & System Safety，2023，234：109142.
[13] SUNNERSJÖ C S. Varying compliance vibrations of rolling bearings[J]. Journal of Sound and Vibration，1978，58(3)：363-373.
[14] 刘倩楠，郭瑜，伍星. 滚动轴承外圈剥落故障双冲击特征机理建模[J]. 振动工程学报，2017，30(4)：670-678. LIU Qiannan，GUO Yu，WU Xing. Modeling of dual impact characteristic mechanism for outer ring peeling fault of rolling bearings[J]. Journal of Vibration Engineering，2017，30(4)：670-678.
[15] 杨岗，徐五一，邓琴，等. 基于振动信号的滚动轴承复合故障诊断研究综述[J]. 西华大学学报，2024，43(1)：48-69. YANG Gang，XU Wuyi，DENG Qin，et al. Review of research on composite fault diagnosis of rolling bearings based on vibration signals[J]. Journal of Xihua University，2024，43(1)：48-69.
[16] XUE Jiankai，SHEN Bo. Dung beetle optimizer：A new meta-heuristic algorithm for global optimization[J]. The Journal of Supercomputing，2023，79(7)：7305-7336.
[17] CHATTERJEE S，ROY S，DAS A，et al. Secure biometric-based authentication scheme using Chebyshev chaotic map for multi-server environment[J]. IEEE Transactions on Dependable and Secure Computing，2016，15(5)：824-839.
[18] 徐慧玲，刘升，李安东. 基于双向局部开发和黄金正弦的异构导向的鲸鱼优化算法[J]. 计算机工程与科学，2024，46(6)：1128-1140. XU Huiling，LIU Sheng，LI Andong. Heterogeneous oriented whale optimization algorithm based on bidirectional local development and golden sine[J]. Computer Engineering and Science，2024，46(6)：1128-1140.
[19] QIN Yi，WU Xingguo，LUO Jun. Data-model combined driven digital twin of life-cycle rolling bearing[J]. IEEE Transactions on Industrial Informatics，2021，18(3)：1530-1540.