Aiming at the problems of difficulty in extracting feature information generated by belt conveyor roller bearing faults, as well as low accuracy and poor robustness of fault diagnosis and identification, NSST4, SVD and DBN methods were combined to propose a suitable method for belt conveyor roller bearing acoustic signal fault diagnosis. Firstly, sequential variational mode decomposition (SVMD) was used to process the acoustic signals to enhance the recognizability of fault features. Second, the processed one-dimensional signals were converted to a two-dimensional time-frequency matrix by NSST4, which was used as the inputs of the feature matrix. Subsequently, the feature matrix was downsized using SVD technique to extract the key singular value vectors that might characterize the status of the roll bearings. These singular value vectors were then input into DBN, and the DBN core parameters were optimized by the improved sparrow search algorithm (ISSA) to improve the recognition performance of the model. Finally, in order to further validate the effectiveness of the proposed method, it was tested by simulated fault experiments and field experiments. In the simulated fault experiments of the roller bearings, the accuracy rate of the proposed method reaches 97.91%. Compared with other 5 methods, the accuracy of the proposed method is the highest, and the mean absolute error (MAE) is the lowest. In the field experiments, the recognition accuracy reaches 96.57%.
为此,本文提出了一种融合高阶同步压缩变换(SSTM)和多重同步压缩(NSST)的方法,构建多重高阶同步压缩变换(NSSTM)。在该框架下采用NSST4提高时频分析(TFA)分辨率,减少能量泄漏和模态混叠,更好地适应非平稳信号变化。但二重四阶同步压缩变换(NSST4)产生的时频矩阵维数较高、冗余信息较多,为简化后续处理过程,本文采用奇异值分解(singular value decomposition,SVD)对其降维,通过保留最大奇异值对应的特征向量实现特征矩阵的有效简化[11]。
基于式(18)可知,矩阵 H 可表示为若干个由奇异向量构成的秩为1的分量矩阵的线性叠加形式,各分量对原矩阵的重要程度由对应奇异值的大小决定。因此,可通过舍弃较小奇异值对应分量,实现矩阵的低秩近似表示。奇异值由快速衰减到平缓的位置点可作为矩阵有用成分与无用成分的分界点。定义分界点的序号为l,则托辊故障信号时频矩阵经SVD降维后的特征矩阵为
ZHOUPing, MAGuoqing, ZHOUGongbo, et al. Health Monitoring Technology for the Intelligent Belt Conveyor: a Review[J]. Chinese Journal of Scientific Instrument, 2023, 44(12): 1-21.
WANGHaijun, WANGHonglei. Status and Prospect of Intelligent Key Technologies of Belt Conveyor[J]. Coal Science and Technology, 2022, 50(12): 225-239.
[5]
LINNanying, ZHANGXiangzhen, YANShengming,et al. The Structure and Performance Analysis of Ejector Dust-removing Nozzle[J]. Fluid Machinery,1998,26(12):28-30.
FENGSiqian, WANGJiaxu, ZHANGXin, et al. Rolling Bearing Fault Diagnosis Using a New Resonance Demodulation Method[J]. China Mechanical Engineering, 2025, 36(9): 2022-2031.
GUOYanqiu, MIAOChangyun, LIUYi. Research on Fault Detection of Belt Conveyor Roller Based on Thermal Infrared Image[J]. Industry and Mine Automation, 2023, 49(10): 52-60.
[10]
PENGChen, LIZhipeng, YANGMinjing, et al. An Audio-based Intelligent Fault Diagnosis Method for Belt Conveyor Rollers in Sand Carrier[J]. Control Engineering Practice, 2020, 105: 104650.
CHENJian, YANMinghui, CHENPin. Acoustic Signal Fault Diagnosis Method of Centrifugal Pumps Based on Bayesian Optimization Multiscale DenseNet[J]. China Mechanical Engineering, 2025, 36(9): 2032-2038.
HAOWangshen, LIJikang, DONGXinmin, et al. Research on Bearing Fault Diagnosis Based on Acoustic Signal Recursive Hilbert Transform[J]. Machine Tool & Hydraulics, 2024, 52(4): 195-199.
[17]
HEMMATIF, ORFALIW, GADALAM S. Roller Bearing Acoustic Signature Extraction by Wavelet Packet Transform, Applications in Fault Detection and Size Estimation[J]. Applied Acoustics, 2016, 104: 101-118.
ZHAOXuezhi, YEBangyan. Multi-resolution SVD Packet Theory and Its Application to Signal Processing[J]. Acta Electronica Sinica, 2012, 40(10): 2039-2046.
LIUJunfeng, DONGBaoying, YUXiang, et al. Rolling Bearing Fault Diagnosis Method Based on FSC-MPE and BP Neural Network[J]. Chinese Journal of Ship Research, 2021, 16(6): 183-190.
LIUYunhang, SONGYubo, ZHUDapeng. A Rolling Bearing Fault Classification Method Based on IGWO-SVM Combined with Center Correction Projection[J]. Journal of Vibration and Shock, 2023, 42(24): 267-275.
[24]
JIAOJian, ZHENGXuejiao. Fault Diagnosis Method for Industrial Robots Based on DBN Joint Information Fusion Technology[J]. Computational Intelligence and Neuroscience, 2022, 2022(1): 4340817.
[25]
NASKATHJ, SIVAKAMASUNDARIG, BEGUMA A S. A Study on Different Deep Learning Algorithms Used in Deep Neural Nets: MLP SOM and DBN[J]. Wireless Personal Communications, 2023, 128(4): 2913-2936.
[26]
XUEJiankai, SHENBo. A Survey on Sparrow Search Algorithms and Their Applications[J]. International Journal of Systems Science, 2024, 55(4): 814-832.
LIUGuojin, LIUDaming, MIAOJianhua, et al. Fault Identification of Automatic Transfer Switching Equipment Based on VMD-WPE and IGWO Optimized DBN[J]. Transactions of China Electrotechnical Society, 2024, 39(4): 1221-1233.
LIUJie, FUXuejiao, SUNXingwei. Operating State Identification of Wind Turbine Gearbox Based on PSO-DBN[J]. Chinese Journal of Sensors and Actuators, 2023, 36(3): 434-440.
WEIDong, LIUKan, DINGRongjun, et al. A Multi-synchrosqueezing Transformation Based Early Stage Detection of Inter-turn Short Circuit Fault in Permanent Magnet Synchronous Machine[J]. Transactions of China Electrotechnical Society, 2022, 37(18): 4651-4663.
GUOFengyi, GAOHongxin, WANGZhiyong, et al. Feature Extraction Method of Series Fault Arc Based on ST-SVD-PCA[J]. Journal of China Coal Society, 2018, 43(3): 888-896.
ZHANGZhengwu, FENGZhipeng, CHENXiaowang. Acoustic Signal Analysis of the Resonance Frequency Region for Planetary Gearbox Fault Diagnosis Based on High-order Synchrosqueezing Transform[J]. Chinese Journal of Engineering, 2020, 42(8): 1048-1054.
[37]
DEHGHANIM, TROJOVSKÝP. Osprey Optimization Algorithm: a New Bio-inspired Metaheuristic Algorithm for Solving Engineering Optimization Problems[J]. Frontiers in Mechanical Engineering, 2023, 8: 1126450.
RENLiang, ZHENLongxin, ZHAOYun, et al. Fault Diagnosis of Rolling Bearing under Strong Background Noise Based on SSA-VMD-MCKD[J]. Journal of Vibration and Shock, 2023, 42(3): 217-226.
[44]
邵思羽. 基于深度学习的旋转机械故障诊断方法研究[D]. 南京: 东南大学, 2019.
[45]
SHAOSiyu. Methodologies for Fault Diagnosis of Rotary Machine Based on Deep Learning[D]. Nanjing: Southeast University, 2019.
CHENChuang, LIXianfeng, SHIJiantao, et al. Intelligent Fault Diagnosis Method for Rolling Bearings Based on Flexible Residual Neural Network[J]. Chinese Journal of Engineering, 2025, 47(3): 480-488.