To address the issues of insufficient feature extraction ability and low efficiency of manual parameter optimization in traditional single deep learning models for processing data under variable operating conditions, this study proposes a physics-prior-based time-frequency dual-channel model, CNBiG-AM. The model is designed to handle the different impacts of variable conditions on the manifestation of time-frequency domain fault features by employing a dual-channel time-frequency approach for complementary feature learning. At the same time, a lightweight and efficient model and attention collaboration mechanism are established, utilizing decision-level deep feature fusion for diagnostic decisions. The Sparrow Search Algorithm (SSA) is used as an intelligent optimization strategy to ensure the optimal performance of the complex model under variable operating conditions. Ablation experiments using the CWRU and HNU datasets demonstrate that CNBiG-AM outperforms baseline models in all metrics, with accuracy improvements of 6.5% and 5.5%, respectively, proving the necessity of the BiGRU-attention collaboration mechanism. After intelligent optimization using SSA, the model performance is further enhanced, with accuracy improving by 2.5% on the CWRU dataset and 3.5% on the more generalized HNU dataset. The t-SNE visualization graph shows a significant improvement in the feature space aggregation effect. This study provides an efficient and robust ensemble model for bearing fault diagnosis under variable operating conditions, demonstrating the effectiveness of dual-frequency separation and intelligent optimization algorithms.
ZhangHCheWCaoY, et al. Condition monitoring and fault diagnosis of rotating machinery towards intelligent manufacturing: review and prospect. Iran J Sci Technol Mech Eng2025; 49(1): 1–34. https://doi.org/10.1007/s40997-024-00783-w
2.
ZhaoBZhangXZhanZ, et al. A robust construction of normalized CNN for online intelligent condition monitoring of rolling bearings considering variable working conditions and sources. Measurement2021; 174: 108973. https://doi.org/10.1016/j.measurement.2021.108973
3.
XieZZhanHWangY, et al. Meta-learning-based fault diagnosis method for rolling bearings under cross-working conditions. Meas Sci Technol2025; 36(1): 016218. https://doi.org/10.1088/1361-6501/ad916a
4.
PopescuTDAiordachioaieDCulea-FlorescuA. Basic tools for vibration analysis with applications to predictive maintenance of rotating machines: an overview. Int J Adv Manuf Technol2022; 118(9-10): 2883–2899. https://doi.org/10.1007/s00170-021-07703-1
5.
ZhaoXWangLYangM, et al. A novel small-sample fault diagnosis method for rolling bearings via continuous wavelet transform and Siamese neural network. IEEE Sens J2024; 24(15): 24988–24996. https://doi.org/10.1109/jsen.2024.3409768
6.
VanMKangH. Two-stage feature selection for bearing fault diagnosis based on dual-tree complex wavelet transform and empirical mode decomposition. J Mech Eng Sci2016; 230(2): 291–302.
7.
HuCLiYChenZ, et al. Research on fault diagnosis of rolling bearing based on multi-sensor bi-layer information fusion under small samples. Rev Sci Instrum2023; 94(11): 115106. https://doi.org/10.1063/5.0174359
8.
YuJZhaoKGuoZ, et al. RUL prediction of rolling bearings based on adaptive variational mode decomposition and double-discriminator conditional CycleGAN. Proc Inst Mech Eng C J Mech Eng Sci2024; 238(12): 6007–6017. https://doi.org/10.1177/09544062231220201
9.
WangHLiuZPengD, et al. Understanding and learning discriminant features based on multiattention 1DCNN for Wheelset Bearing Fault Diagnosis. IEEE Trans Ind Inform2020; 16(9): 5735–5745. https://doi.org/10.1109/tii.2019.2955540
10.
HakimMOmranAABAhmedAN, et al. A systematic review of rolling bearing fault diagnoses based on deep learning and transfer learning: taxonomy, overview, application, open challenges, weaknesses and recommendations. Ain Shams Eng J2023; 14(4): 101945. https://doi.org/10.1016/j.asej.2022.101945
11.
RenHLiuSWeiF, et al. A novel two-stream multi-head self-attention convolutional neural network for bearing fault diagnosis. Proc Inst Mech Eng C J Mech Eng Sci2024; 238(11): 5393–5405. https://doi.org/10.1177/09544062231216877
12.
DuWNiuXWangH, et al. A fault diagnosis method for key transmission components of rotating machinery based on SAM-1DCNN-BiLSTM temporal and spatial feature extraction. Proc Inst Mech Eng C J Mech Eng Sci2025; 239(14): 5810–5822. https://doi.org/10.1177/09544062251328244
13.
JoungBGNathCLiZ, et al. Bearing anomaly detection in an air compressor using an LSTM and RNN-based machine learning model. Int J Adv Manuf Technol2024; 134(7–8): 3519–3530. https://doi.org/10.1007/s00170-024-14322-z
14.
GaoSLiTZhangY, et al. Fault diagnosis method of rolling bearings based on adaptive modified CEEMD and 1DCNN model. ISA Trans2023; 140: 309–330. https://doi.org/10.1016/j.isatra.2023.05.014
15.
AliYMDingLQinS. An efficient approach for diagnosing faults in photovoltaic array using 1D-CNN and feature selection Techniques. Int J Electr Power Energy Syst2025; 166: 110526. https://doi.org/10.1016/j.ijepes.2025.110526
16.
HaoSGeF-XLiY, et al. Multisensor bearing fault diagnosis based on one-dimensional convolutional long short-term memory networks. Measurement2020; 159: 107802. https://doi.org/10.1016/j.measurement.2020.107802
17.
Siami-NaminiSTavakoliNNaminAS. The Performance of LSTM and BiLSTM in forecasting time series. In: IEEE International Conference on Big Data (Big Data), Los Angeles, CA. 2019, pp. 3285–3292.
18.
WuJKongLKangS, et al. Aircraft engine fault diagnosis model based on 1DCNN-bilstm with CBAM. Sensors2024; 24(3): 780. https://doi.org/10.3390/s24030780
19.
MaKWangYYangY. Fault diagnosis of wind turbine blades based on one-dimensional convolutional neural network-bidirectional long short-term memory-adaptive boosting and multi-source data fusion. Appl Sci2025; 15(7): 3440. https://doi.org/10.3390/app15073440
20.
NacerSMNadiaBAbdelghaniR, et al. A novel method for bearing fault diagnosis based on BiLSTM neural networks. Int J Adv Manuf Technol2023; 125(3–4): 1477–1492. https://doi.org/10.1007/s00170-022-10792-1
21.
YaoXLvZCaoL, et al. Parallel CNN-BiLSTM fault diagnosis method based on multi-domain transformation. In: 2024 IEEE International Conference on Sensing, Diagnostics, Prognostics, and Control (SDPC). IEEE, 2024, pp. 42–46.
22.
HouPZhangJJiangZ, et al. A bearing fault diagnosis method based on dilated convolution and multi-head self-attention mechanism. Appl Sci2023; 13(23): 12770. https://doi.org/10.3390/app132312770
23.
ZhangJTanZZhangR, et al. Load identification and prediction based on feature fusion and lightweight neural network. In: 2024 IEEE 7th International Electrical and Energy Conference (CIEEC). IEEE, 2024, pp. 4565–4570.
24.
MaMQuCZhaoX, et al. Enhanced rolling bearing fault diagnosis using a multi-stage attention fusion network. Proc Inst Mech Eng C J Mech Eng Sci2025; 239(13): 5297–5309. https://doi.org/10.1177/09544062251328857
25.
XuZLiCYangY. Fault diagnosis of rolling bearings using an improved multi-scale convolutional neural network with feature attention mechanism. ISA Trans2021; 110: 379–393. https://doi.org/10.1016/j.isatra.2020.10.054
26.
ZhaoYWeiSXuX. Bearing fault diagnosis based on wavelet adaptive threshold filtering and multi-channel fusion cross-attention neural network. Rev Sci Instrum2024; 95(11): 114703. https://doi.org/10.1063/5.0223715
27.
WanMXiaoYZhangJ. Research on fault diagnosis of rolling bearing based on improved convolutional neural network with sparrow search algorithm. Rev Sci Instrum2024; 95(4): 045111. https://doi.org/10.1063/5.0192639
28.
KhanASohailAZahooraU, et al. A survey of the recent architectures of deep convolutional neural networks. Artif Intell Rev2020; 53(8): 5455–5516. https://doi.org/10.1007/s10462-020-09825-6
29.
ZhangSZhangSWangB, et al. Deep learning algorithms for bearing fault diagnostics—A comprehensive review. IEEE Access2020; 8: 29857–29881.
30.
LiuYGuoLWangQ, et al. Application to induction motor faults diagnosis of the amplitude recovery method combined with FFT. Mech Syst Signal Process2010; 24(8): 2961–2971. https://doi.org/10.1016/j.ymssp.2010.03.008
31.
DaiYYuWLengM. A hybrid ensemble optimized BiGRU method for short-term photovoltaic generation forecasting. Energy2024; 299: 131458. https://doi.org/10.1016/j.energy.2024.131458
32.
ShaoHLaiYLiuH, et al. LSFConvformer: a lightweight method for mechanical fault diagnosis under small samples and variable speeds with time-frequency fusion. Mech Syst Signal Process2025; 236: 113016. https://doi.org/10.1016/j.ymssp.2025.113016
JiangZTanQLiN, et al. A novel BiGRU multi-step wind power forecasting approach based on multi-label integration random forest feature selection and neural network clustering. Energy Convers Manag2024; 319: 118904. https://doi.org/10.1016/j.enconman.2024.118904
35.
ZhaoJGuoSMuD. DouBiGRU-A: software defect detection algorithm based on attention mechanism and double BiGRU. Comput Secur2021; 111: 102459. https://doi.org/10.1016/j.cose.2021.102459
36.
DengLWangGZhaoC, et al. MARNet: multi-head attention residual network for rolling bearing fault diagnosis under noisy condition. Proc Inst Mech Eng C J Mech Eng Sci2024; 238(19): 9726–9747. https://doi.org/10.1177/09544062241259614
JinZHeDWeiZ. Intelligent fault diagnosis of train axle box bearing based on parameter optimization VMD and improved DBN. Eng Appl Artif Intell2022; 110: 104713. https://doi.org/10.1016/j.engappai.2022.104713
39.
SmithWARandallRB. Rolling element bearing diagnostics using the Case Western Reserve University data: a benchmark study. Mech Syst Signal Process2015; 64–65: 100–131. https://doi.org/10.1016/j.ymssp.2015.04.021
40.
XuHPanHZhengJ, et al. Dynamic penalty adaptive matrix machine for the intelligent detection of unbalanced faults in roller bearing. Knowl Syst2022; 247: 108779. https://doi.org/10.1016/j.knosys.2022.108779
41.
CuiLJinZHuangJ, et al. Fault severity classification and size estimation for ball bearings based on vibration mechanism. IEEE Access2019; 7: 56107–56116. https://doi.org/10.1109/access.2019.2911323
42.
HuCLiYChenZ, et al. A novel rolling bearing fault diagnosis method based on parameter optimization variational mode decomposition with feature weighted reconstruction and multi-target attention convolutional neural networks under small samples. Rev Sci Instrum2023; 94(7): 075109. https://doi.org/10.1063/5.0158412
43.
MenZChenZLiY, et al. Railway wagon bearing fault diagnosis method based on improved sparrow search algorithm optimizing variational mode decomposition and multi-level convolutional neural network. Rev Sci Instrum2024; 95(4): 045104. https://doi.org/10.1063/5.0191209
44.
LiXZhangWDingQ, et al. Intelligent rotating machinery fault diagnosis based on deep learning using data augmentation. J Intell Manuf2020; 31(2): 433–452. https://doi.org/10.1007/s10845-018-1456-1
45.
SunTGaoJ. New fault diagnosis method for rolling bearings based on improved residual shrinkage network combined with transfer learning. Sensors2024; 24(17): 5700. https://doi.org/10.3390/s24175700
