Deep ensemble learning with non-equivalent costs of fault severities for rolling bearing diagnostics

Yang, Honggang; Li, W.D.; Hu, Kaixiong; Liang, Yuchen; Lv, Yaqiong

doi:10.1016/j.jmsy.2021.09.009

Cited by 25 publications

(9 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ensemble learning, also known as multi-classifier system or committee-based learning, improves the classification accuracy and generalization performance by combining outputs of multiple classifiers. To enhance the quality and stability of the individual classifiers, the classification results are pooled by using proper algorithms [41][42][43][44]. Figure 3 shows the architecture of the selected ensemble learning algorithm in this paper.…”

Section: Ensemble Learningmentioning

confidence: 99%

AI based Inspection of the Axes of Machine Tools Corresponding author: Ibrahim N. Tansel E-Mail: tanseli@fiu.edu

Demetgul,

Wang,

Fleischer

et al. 2023

Preprint

View full text Add to dashboard Cite

Linear stages are one of the most important components of machine tools, additive manufacturing equipment, and many machines which are needed to create precise linear motion. Monitoring the current of electric motor has been used for sensorless diagnostic of linear stages. This paper proposed a new Automated Machine Learning (AutoML) approach. The proposed AutoML used multiple methods for interpretation of the current signal to estimate the extent of the misalignment problems. Support Vector Machine (SVM), Gradient Boosting (GB) and Auto-multilayer perceptron (AutoMLP) methods were used for classification of the data. To enhance the performance of these methods Ensemble learning (EL) was used to obtain the final decision by using estimations of each method. Motor current signals in the horizontal and vertical direction were saved in the user interface's database. AutoML learned the proper classification through the user interface which holds data and user interpretations for training and started to make classifications. To improve the classification performance, each hyperparameter was optimized and compared with the initial results. Experimental studies showed that the ensemble method was superior compared to the considered classification methods in fault detection through the motor current signal. The findings indicated that the current features could be used successfully discriminate the signals in the horizontal and vertical directions and could detect linear stage defects. In addition, the results demonstrated that additional fault detection capabilities may be added to the system.

show abstract

Section: Ensemble Learningmentioning

confidence: 99%

AI based Inspection of the Axes of Machine Tools Corresponding author: Ibrahim N. Tansel E-Mail: tanseli@fiu.edu

Demetgul,

Wang,

Fleischer

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…In the second experiment, the diagnosis models of RF, support vector machine (SVM), gradient boosting decision tree (GBDT), eXtreme gradient boosting (XGBoost), light gradient boosting machine (LightGBM), and decision tree (DT) [35][36][37][38][39] are selected to better explain the performance of the proposed method. Figure 13 presents the comparison of fault diagnosis result under diferent diagnosis models.…”

Section: Fault Diagnosismentioning

confidence: 99%

Rolling Bearing Fault Diagnosis Method Based on MCMF and SAIMFE

Meng

Miao

et al. 2022

Shock and Vibration

View full text Add to dashboard Cite

Fault diagnosis of rolling bearing is important for ensuring the safe operation of industrial machinery. In order to improve diagnosis accuracy of bearing fault, a rolling bearing fault diagnosis method based on multiscale combined morphological filter (MCMF) and self-adaption improved multiscale fuzzy entropy (SAIMFE) is proposed in this paper. First, the MCMF is designed to eliminate noise and preserve fault information more effectively. Second, SAIMFE is proposed to extract bearing fault features, and the optimized scale factor of SAIMFE is determined based on the absolute skewness. Third, some experiments are completed to demonstrate the effectiveness and superiority of the proposed method. The experimental results show that the proposed method not only has high diagnosis accuracy but also less dependent on the diagnosis model.

show abstract

“…As a result, damage to the components of machinery equipment is usually unavoidable [1]. Unforeseen machinery failures can result in immeasurable financial losses and even potential casualties [2]. Therefore, predictive maintenance of machinery equipment based on fault diagnosis technology is necessary [3][4][5].…”

Section: Introductionmentioning

confidence: 99%

Compound fault diagnosis of rolling bearings with few-shot based on DCGAN-RepLKNet

Zhu,

Fang

2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

To guarantee the security of personnel on-site, diagnosing the malfunction of mechanical apparatus is imperative. The accomplishments within the domain of fault diagnosis have been partly attributed to the advancements in deep learning technology, which excels in feature extraction through extensive datasets. However, it is difficult to collect sufficient data to train high-precision fault diagnosis models in practice. To solve this problem, a novel method called DCGAN-RepLKNet is proposed to address the challenge of gathering enough data to train high-precision fault diagnosis models in practice. This technique involves transforming a one-dimensional time series vibration signal into a two-dimensional time-frequency map via wavelet transform technology. Subsequently, DCGAN expands the two-dimensional time-frequency map samples produced. Finally, RepLKNet is used to classify the fault samples. The proposed method has been verified in PU compound fault data set and bearing real damage data set. The results show that the accuracy of this method has been improved by 5.70%, 6.34%, 9.08%, and 16.35% compared to 2D-CNN under different sizes datasets in Case 1, and by 24.5% compared to 2D-CNN in Case 2.

show abstract

Deep ensemble learning with non-equivalent costs of fault severities for rolling bearing diagnostics

Cited by 25 publications

References 34 publications

AI based Inspection of the Axes of Machine Tools Corresponding author: Ibrahim N. Tansel E-Mail: tanseli@fiu.edu

AI based Inspection of the Axes of Machine Tools Corresponding author: Ibrahim N. Tansel E-Mail: tanseli@fiu.edu

Rolling Bearing Fault Diagnosis Method Based on MCMF and SAIMFE

Compound fault diagnosis of rolling bearings with few-shot based on DCGAN-RepLKNet

Contact Info

Product

Resources

About