Multimodal Intelligent Diagnosis Model and Adaptive Parameter Optimization of Deep Groove Ball Bearings in Packaging Machines

CHE Chang; LI Minghui; MA Chenpei; QI Mengyuan

doi:10.19554/j.cnki.1001-3563.2025.19.026

PDF(9385 KB)

Packaging Engineering ›› 2025, Vol. 46 ›› Issue (19) : 247-257. DOI: 10.19554/j.cnki.1001-3563.2025.19.026

Multimodal Intelligent Diagnosis Model and Adaptive Parameter Optimization of Deep Groove Ball Bearings in Packaging Machines

CHE Chang¹, LI Minghui^2*, MA Chenpei², QI Mengyuan²

Author information +

History +

Abstract

To solve the problem of difficult early fault detection of deep groove ball bearings in packaging machines under high-speed start/stop and high dust and humidity environments, the work aims to study a fault diagnosis method based on multimodal feature fusion and intelligent algorithm optimization to improve equipment operation and maintenance efficiency. Considering the problems of complex structure, difficult training, and inability to adaptively select parameters in traditional deep belief networks for state monitoring and fault diagnosis, a multimodal intelligent diagnostic model was constructed by combining wavelet packet transform (WPT), sparrow search algorithm (SSA), and deep belief network (DBN). WPT was used to extract time-domain, frequency-domain, and energy features, reducing input dimensions. Introducing SSA adaptive optimization of the network layers, hidden layer nodes, and learning rate of DBN to solve the problem of traditional DBN parameter dependence on experience achieved adaptive selection of parameters, more accurately and quickly identifying bearing fault states. The multimodal intelligent model had a diagnostic accuracy of 100% for the normal state of deep groove ball bearings in packaging machines. The average diagnostic accuracy of different degrees of inner ring, outer ring, and rolling element faults was improved to 98.58%, 97.75%, and 98.42%, respectively, and the training time was shortened by about one minute. The optimized model can effectively solve the diagnostic problem of deep groove ball bearings in packaging machines under complex working conditions, providing an intelligent diagnostic solution for predictive maintenance of packaging machines.

Key words

packaging machine / deep groove ball bearings / wavelet packet transform / sparrow search algorithm

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

CHE Chang, LI Minghui, MA Chenpei, QI Mengyuan. Multimodal Intelligent Diagnosis Model and Adaptive Parameter Optimization of Deep Groove Ball Bearings in Packaging Machines[J]. Packaging Engineering. 2025, 46(19): 247-257 https://doi.org/10.19554/j.cnki.1001-3563.2025.19.026

References

[1] 郭志明, 张鹏敏, 李兆丰, 等. 食品智能仓储、包装技术与高端装备研究进展[J]. 中国食品学报, 2025, 25(1): 12-25.
GUO Z M, ZHANG P M, LI Z F, et al.Advance of Food Intelligent Storage and Packaging Technology and High-End Equipment[J]. Journal of Chinese Institute of Food Science and Technology, 2025, 25(1): 12-25.
[2] 中国包装联合会. 包装业全年营收超1.15万亿元[N]. 中国新闻出版广电报, 2024-06-19(006).
China Packaging Federation.Packaging Industry Revenue Exceeds 1.15 Trillion Yuan[N]. China Press, Publication, Radio, Film and Television News, 2024-06-19(006).
[3] 张岩, 丁文亮, 李文慧. 包装装置系统效率提升研究[J]. 中国氯碱, 2025(5): 55-58.
ZHANG Y, DING W L, LI W H.Research on Efficiency Improvement of Packaging System[J]. China Chlor-Alkali, 2025(5): 55-58.
[4] 李友家, 张忠伟, 焦宗豪, 等. 强噪声干扰下基于SVMD-FFCNN的深沟球轴承故障分类模型[J]. 机电工程, 2025, 42(4): 686-696.
LI Y J, ZHANG Z W, JIAO Z H, et al.Fault Classification Model of Deep Groove Ball Bearing Based on SVMD-FFCNN under Strong Noise Interference[J]. Mechanical & Electrical Engineering, 2025, 42(4): 686-696.
[5] 张秋欣, 李华. 基于“rbio5.5” 小波函数小波包分解的滚动轴承故障诊断研究[J]. 机械工程学报, 2022, 58(10): 150-158.
ZHANG Q X, LI H.Research on the Fault Diagnosis of Rolling Bearings Based on the Wavelet Packet Decomposition with the "rbio5.5" Wavelet Function[J]. Journal of Mechanical Engineering, 2022, 58(10): 150-158.
[6] ZHANG D Q, DING X X, HUANG W B, et al.Transient Signal Analysis Using Parallel Time-Frequency Manifold Filtering for Bearing Health Diagnosis[J]. IEEE Access, 2019, 7: 175277-175289.
[7] 周涛涛, 朱显明, 彭伟才, 等. 基于CEEMD和排列熵的故障数据小波阈值降噪方法[J]. 振动与冲击, 2015, 34(23): 207-211.
ZHOU T T, ZHU X M, PENG W C, et al.A Wavelet Threshold Denoising Method for Fault Data Based on CEEMD and Permutation Entropy[J]. Journal of Vibration and Shock, 2015, 34(23): 207-211.
[8] HINTON G E, OSINDERO S, TEH Y W.A Fast Learning Algorithm for Deep Belief Nets[J]. Neural Computation, 2006, 18(7): 1527-1554.
[9] 柴保明, 周涛涛. 概率神经网络在包装设备轴承故障诊断中的应用[J]. 包装工程, 2020, 41(9): 156-162.
CHAI B M, ZHOU T T, Application of Probabilistic Neural Network in Fault Diagnosis of Bearings in Packaging Equipment. Packaging Engineering, 2020, 41(9): 156-162.
[10] JIA F, LEI Y G, LIN J, et al.Deep Neural Networks: A Promising Tool for Fault Characteristic Mining and Intelligent Diagnosis of Rotating Machinery with Massive Data[J]. Mechanical Systems and Signal Processing, 2016, 72: 303-315.
[11] 甄冬, 孙赫明, 冯国金, 等. 基于包络谱语义构建的零样本滚动轴承复合故障诊断方法[J]. 振动与冲击, 2024, 43(14): 189-200.
ZHEN D, SUN H M, FENG G J, et al.Zero-Shot Rolling Bearing Compound Fault Diagnosis Method Based on Envelope Spectrum Semantic Construction[J]. Journal of Vibration and Shock, 2024, 43(14): 189-200.
[12] 张亚雄, 李枝荣, 孙亚军. 小波包-偏最小二乘在滚动轴承故障检测中的应用[J]. 机电工程技术, 2023, 52(8): 181-183.
ZHANG Y X, LI Z R, SUN Y J.Application of Wavelet Packet Energy Spectrum and PLS for Ball Bearing Fault Detection[J]. Mechanical & Electrical Engineering Technology, 2023, 52(8): 181-183.
[13] 陈超, 许琦, 宋正华, 等. 结合深度学习和迁移学习的行星齿轮箱故障诊断算法研究[J]. 机床与液压, 2025, 53(10): 40-49.
CHEN C, XU Q, SONG Z H, et al.Research on Planetary Gearbox Fault Diagnosis Algorithm Combining Deep Learning and Transfer Learning[J]. Machine Tool & Hydraulics, 2025, 53(10): 40-49.
[14] 贺思艳, 刘亚, 田新诚. 基于小波包-AR谱和深度学习的轴承故障诊断研究[J]. 计算机应用研究, 2019, 36(6): 1758-1761.
HE S Y, LIU Y, TIAN X C.Research on Bearing Fault Diagnosis Based on Wavelet Packet-Auto Regressive Model Spectrum and Deep Learning[J]. Application Research of Computers, 2019, 36(6): 1758-1761.
[15] 吴闻笛, 吴征天. 基于深度强化学习的整数规划算法优化[J]. 苏州科技大学学报(自然科学版), 2025, 42(2): 76-84.
WU W D, WU Z T.Optimization of Integer Programming Algorithms Based on Deep Reinforcement Learning[J]. Journal of Suzhou University of Science and Technology (Natural Science Edition), 2025, 42(2): 76-84.
[16] 唐磊, 茅晔辉, 蔡婧, 等. 光谱数据增强方法及其应用进展[J]. 分析测试学报, 2025, 44(6): 1227-1236.
TANG L, MAO Y H, CAI J, et al.Spectral Data Augmentation Methods and Their Application Progress[J]. Journal of Instrumental Analysis, 2025, 44(6): 1227-1236.
[17] 丁萍萍. 基于IDA优化极限学习机的轴承故障诊断研究[D]. 青岛: 青岛科技大学, 2020: 35-36.
DING P P.Research on Bearing Fault Diagnosis Based on IDA-Optimized Extreme Learning Machine[D]. Qingdao: Qingdao University of Science & Technology, 2020: 35-36.
[18] 万庆, 袁志鹏, 王二振, 等. 基于优化SVM的轻载工况滚动轴承故障诊断[J]. 教练机, 2025(1): 63-68.
WAN Q, YUAN Z P, WANG E Z, et al.Fault Diagnosis of Rolling Bearing under Light Load Operating Condition Based on Optimized Support Vector Machine(SVM)[J]. Trainer, 2025(1): 63-68.
[19] 潘晓铭, 彭自良, 李韦廷, 等. 基于ARM架构的多通道振动信号采集与故障诊断系统设计[J]. 机电工程技术, 2023, 52(10): 120-123.
PAN X M, PENG Z L, LI W T, et al.Design of Multi- channel Vibration Signal Acquisition and Fault Diagnosis System Based on ARM Architecture[J]. Mechanical & Electrical Engineering Technology, 2023, 52(10): 120-123.
[20] 韩东洋, 陈宏, 陈新财, 等. 短时傅里叶变换结合DRSN的滚动轴承故障诊断研究[J]. 中国测试, 2024, 50(10): 136-141.
HAN D Y, CHEN H, CHEN X C, et al.Research on Fault Diagnosis of Rolling Bearings Based on Short-Time Fourier Transform Combined with DRSN[J]. China Measurement & Testing Technology, 2024, 50(10): 136-141.
[21] 顾张清, 黄清岩, 任世锦, 等. 基于CELMDAN与IMOMEDA的微弱机械特征增强方法[J]. 常州大学学报(自然科学版), 2025, 37(3): 75-86.
GU Z Q, HUANG Q Y, REN S J, et al.Weak Mechanical Feature Enhancement Method Combining CELMDAN with IMOMEDA[J]. Journal of Changzhou University (Natural Science Edition), 2025, 37(3): 75-86.
[22] 胡耀斌, 谢静, 胡良斌. 基于神经网络与小波变换的滚动轴承故障诊断[J]. 机械设计与研究, 2013, 29(6): 33-35.
HU Y B, XIE J, HU L B.Fault Diagnosis of Rolling Bearing Based on Neural Network and Wavelet Transform[J]. Machine Design & Research, 2013, 29(6): 33-35.
[23] 周连弘. 基于多信息融合和深度学习的滚动轴承健康管理研究[D]. 沈阳: 沈阳大学, 2024: 28-32.
ZHOU L H.Research on Health Management of Rolling Bearings Based on Multi-Information Fusion and Deep Learning[D]. Shenyang: Shenyang University, 2024: 28-32.
[24] 聂傲男. 主动磁悬浮轴承系统保护轴承碰撞特性研究及减摩设计[D]. 洛阳: 河南科技大学, 2023: 163-177.
NIE A N.Research on Collision Characteristics and Anti-Friction Design of Protective Bearings in Active Magnetic Suspension Bearing Systems[D]. Luoyang: Henan University of Science and Technology, 2023: 163-177.
[25] 王洁, 刘天伦, 邱溢阳. 基于KPCA与IBWO优化SVM的滚动轴承故障诊断研究[J]. 软件工程, 2025, 28(5): 54-59.
WANG J, LIU T L, QIU Y Y.Research on Rolling Bearing Fault Diagnosis Based on KPCA and IBWO-Optimized SVM[J]. Software Engineering, 2025, 28(5): 54-59.