Fault Diagnosis Method for Imbalanced Data of Rotating Machinery Based on Time Domain Signal Prediction and SC-ResNeSt

Wang, Haitao; Guo, Yifan; Liu, Xiang; Yang, Jie; Xi-heng, Zhang; Shi, Lichen

doi:10.1109/access.2023.3264636

Cited by 5 publications

(5 citation statements)

References 48 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the equation, s dw and s db represent the weight update and bias update for first-order moment estimation, respectively, r dw and r db represent the weight update and bias update for first-order moment estimation, respectively, β 1 and β 2 are hyperparameters controlling the decay rate of the moving averages, and s 1 and r 1 are bias correction formulas to prevent very small gradients at the beginning of the optimization. Based on Equation (11), the gradient update algorithm for Adam can be summarized as follows:…”

Section: Optimize Network Weightsmentioning

confidence: 99%

“…For example, Wang et al proposed a fault diagnosis method based on the combination of a new dual-stage attention-based recurrent neural network (DA-RNN) and depth residual dispersion self-calibration convolution network (SC-ResNeSt). To address the issue of the traditional convolution layers lacking a dynamic receptive field, self-calibrated convolution modules were introduced on the basis of the distraction network (ResNeSt), and a new network model, SC-ResNeSt, was established [11]. Chen et al proposed an adaptive multi-channel residual shrinkage network (AMC-RSN), which extracts as many features as possible by constructing an adaptive multi-channel network.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Intelligent and Small Samples Gear Fault Detection Based on Wavelet Analysis and Improved CNN

Hu,

Zhao,

Huang

et al. 2023

Processes

View full text Add to dashboard Cite

Traditional methods for identifying gear faults typically require a substantial number of faulty samples, which in reality are challenging to obtain. To tackle this challenge, this paper introduces a sophisticated approach for intelligent gear fault identification, utilizing discrete wavelet decomposition and an enhanced convolutional neural network (CNN) optimized for scenarios with limited sample data. Initially, the features of the sample signal are extracted and enhanced using discrete wavelet decomposition. Subsequently, the refined signal is transformed into a two-dimensional image through a Markov transition field, preparing it for improved two-dimensional CNN training. Finally, the refined network model is applied to assess the gear fault dataset, achieving a training accuracy of 97% and a classification accuracy of 88.33%. This demonstrates the method’s feasibility and effectiveness in identifying gear faults with limited sample data.

show abstract

Section: Optimize Network Weightsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Intelligent and Small Samples Gear Fault Detection Based on Wavelet Analysis and Improved CNN

Hu,

Zhao,

Huang

et al. 2023

Processes

View full text Add to dashboard Cite

show abstract

“…The VS acquired from the CP is complex and non-stationary; therefore, extracting health-sensitive information from these signals can help AI algorithms identify if the health state of CPs requires preprocessing in the temporal domain (TD), time-frequency domain (TFD), and Frequency domain (FD) [13,14]. Wang et al [15] introduced a dualphase approach for SF classification in pump rotating parts. First, to process the VS, the dot products of each time-series pair of VS points are used to create a Gramian matrix.…”

Section: Related Research Studiesmentioning

confidence: 99%

A Technique for Centrifugal Pump Fault Detection and Identification Based on a Novel Fault-Specific Mann–Whitney Test

Ahmad,

Kim,

Kim

2023

Sensors

View full text Add to dashboard Cite

This work presents a technique for fault detection and identification in centrifugal pumps (CPs) using a novel fault-specific Mann–Whitney test (FSU Test) and K-nearest neighbor (KNN) classification algorithm. Traditional fault indicators, such as the mean, peak, root mean square, and impulse factor, lack sensitivity in detecting incipient faults. Furthermore, for defect identification, supervised models rely on pre-existing knowledge about pump defects for training purposes. To address these concerns, a new centrifugal pump fault indicator (CPFI) that does not rely on previous knowledge is developed based on a novel fault-specific Mann–Whitney test. The new fault indicator is obtained by decomposing the vibration signature (VS) of the centrifugal pump hierarchically into its respective time-frequency representation using the wavelet packet transform (WPT) in the first step. The node containing the fault-specific frequency band is selected, and the Mann–Whitney test statistic is calculated from it. The combination of hierarchical decomposition of the vibration signal for fault-specific frequency band selection and the Mann–Whitney test form the new fault-specific Mann–Whitney test. The test output statistic yields the centrifugal pump fault indicator, which shows sensitivity toward the health condition of the centrifugal pump. This indicator changes according to the working conditions of the centrifugal pump. To further enhance fault detection, a new effect ratio (ER) is introduced. The KNN algorithm is employed to classify the fault type, resulting in promising improvements in fault classification accuracy, particularly under variable operating conditions.

show abstract

“…The number of fault samples is expanded according to different principles to rebalance the dataset. Examples of these approaches are random multisampling, under-sampling, and data synthesis [34], [35]. The second approach is to optimize the models from the algorithm perspective of it.…”

Section: Introductionmentioning

confidence: 99%

Fault Diagnosis Using Imbalanced Data of Rolling Bearings Based on a Deep Migration Model

Wang,

Zhang

2024

IEEE Access

Self Cite

View full text Add to dashboard Cite

To address the problem that uneven sample distribution can affect the accuracy and stability of fault diagnosis outcomes, we propose a deep transfer learning-Res2Net-convolutional block attention mechanism model. Firstly, the deep migration technique is used to transfer weights of the imbalanced source domain samples to the balanced target domain, expanding the data samples. Secondly, in the feature extraction and detection phase, All eight residual blocks are embedded in convolution block attention to ensure that interference signals are suppressed and that the key fault features are retained. Multilayer feature fusion extracts faulty sample features from multiple residual blocks of the network, which are combined into the feature fusion layer by parallel splicing. Finally, this model is experimentally validated using two different bearing datasets, and the combined evaluation indexes of the new model under the severe imbalance condition are 95.80% and 95.85%, respectively, demonstrating the feasibility and excellent performance of the model.

show abstract

Fault Diagnosis Method for Imbalanced Data of Rotating Machinery Based on Time Domain Signal Prediction and SC-ResNeSt

Cited by 5 publications

References 48 publications

Intelligent and Small Samples Gear Fault Detection Based on Wavelet Analysis and Improved CNN

Intelligent and Small Samples Gear Fault Detection Based on Wavelet Analysis and Improved CNN

A Technique for Centrifugal Pump Fault Detection and Identification Based on a Novel Fault-Specific Mann–Whitney Test

Fault Diagnosis Using Imbalanced Data of Rolling Bearings Based on a Deep Migration Model

Contact Info

Product

Resources

About