An improved ensemble fusion autoencoder model for fault diagnosis from imbalanced and incomplete data

Yang, Jing; Xie, Guo; Yang, Yanxi

doi:10.1016/j.conengprac.2020.104358

Cited by 76 publications

(22 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data enhancement is applied in the field of fault diagnosis by some scholars. A variable-scale resampling strategy, which employed the large shift and less overlap resampling to obtain the sample sets, is designed to solve the problem of fault diagnosis from imbalanced data [ 36 ]. The resampling strategy is more suitable for data without obvious phases, such as vibration signal, rather than the power of RTS.…”

Section: Introductionmentioning

confidence: 99%

A Fault-Diagnosis Method for Railway Turnout Systems Based on Improved Autoencoder and Data Augmentation

Hei

et al. 2022

Sensors

View full text Add to dashboard Cite

In recent years, with the rapid increase in coverage and lines, security maintenance has become one of the top concerns with regard to railway transportation in China. As the key transportation infrastructure, the railway turnout system (RTS) plays a vital role in transportation, which will cause incalculable losses when accidents occur. The traditional fault-diagnosis and maintenance methods of the RTS are no longer applicable to the growing amount of data, so intelligent fault diagnosis has become a research hotspot. However, the key challenge of RTS intelligent fault diagnosis is to effectively extract the deep features in the signal and accurately identify failure modes in the face of unbalanced datasets. To solve the above two problems, this paper focuses on unbalanced data and proposes a fault-diagnosis method based on an improved autoencoder and data augmentation, which realizes deep feature extraction and fault identification of unbalanced data. An improved autoencoder is proposed to smooth the noise and extract the deep features to overcome the noise fluctuation caused by the physical characteristics of the data. Then, synthetic minority oversampling technology (SMOTE) is utilized to effectively expand the fault types and solve the problem of unbalanced datasets. Furthermore, the health state is identified by the Softmax regression model that is trained with the balanced characteristics data, which improves the diagnosis precision and generalization ability. Finally, different experiments are conducted on a real dataset based on a railway station in China, and the average diagnostic accuracy reaches 99.13% superior to other methods, which indicates the effectiveness and feasibility of the proposed method.

show abstract

Section: Introductionmentioning

confidence: 99%

A Fault-Diagnosis Method for Railway Turnout Systems Based on Improved Autoencoder and Data Augmentation

Hei

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…Compared to single prediction algorithms, ensemble methods trains several base models and com-bines them resulting in higher accuracy and lower variance. The ensemble learning approach has been used to diagnose faults in a variety of fields with good performance [15]- [17]. However, ensemble learning, on the other hand, has not been tested for fault diagnosis in coal power plant induced fans.…”

Section: Introductionmentioning

confidence: 99%

Power Plant Induced-Draft Fan Fault Prediction using Machine Learning Stacking Ensemble

Emmanuel¹,

Mpoeleng²,

Maupong³

2022

Preprint

View full text Add to dashboard Cite

<p>The data used our analysis is from an ID fan which is part of a coal powered plant that produces a target output of 117W. The fan is placed at the outlet of the boiler. The fan runs at an air flow rate of 5365 Am3/min , with a density of 0.85 kg/m3, diameter of 2440 mm , with a total pressure of 505 mmAq and a speed of 989 rpm. The fan impeller is driven by a 650 kW electric induction motor (6600 V 50 Hz 3 Phase) with a speed of 1000 rpm.</p>

show abstract

“…Since positive instances are under-represented, they are likely guessed as noise or outliers, or assigned to the majority class regardless the value of their features (García et al, 2019b;Haixiang et al, 2017). The problem of class imbalance is common to many real-world application domains such as fraud detection (Hassan & Abraham, 2016;Zhu et al, 2020b), medical diagnosis (Bach et al, 2017;Wang et al, 2020), credit risk and bankruptcy prediction (García et al, 2019a;Kim et al, 2015;Marqués et al, 2013), fault detection (Codetta-Raiteri & Portinale, 2015;Yang et al, 2020) and document categorization (Bruni & Bianchi, 2020;Jiang et al, 2019), to cite just a few examples.…”

Section: Introductionmentioning

confidence: 99%

DBIG-US: A two-stage under-sampling algorithm to face the class imbalance problem

Guzmán-Ponce

Sánchez

Valdovinos

et al. 2021

Expert Systems with Applications

View full text Add to dashboard Cite

An improved ensemble fusion autoencoder model for fault diagnosis from imbalanced and incomplete data

Cited by 76 publications

References 33 publications

A Fault-Diagnosis Method for Railway Turnout Systems Based on Improved Autoencoder and Data Augmentation

A Fault-Diagnosis Method for Railway Turnout Systems Based on Improved Autoencoder and Data Augmentation

Power Plant Induced-Draft Fan Fault Prediction using Machine Learning Stacking Ensemble

DBIG-US: A two-stage under-sampling algorithm to face the class imbalance problem

Contact Info

Product

Resources

About