A comparative feature analysis for gear pitting level classification by using acoustic emission, vibration and current signals

Sánchez, René–Vinicio; Lucero, Pablo; Vásquez, Rafael E.; Cerrada, Mariela; Cabrera, Diego

doi:10.1016/j.ifacol.2018.09.600

Cited by 22 publications

(11 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The previous section shows the proper performance of the original time-domain condition indicators for fault severity classification with the model obtained by the AutoML systems. The results are compared to alternative ones obtained by the authors when the best set of features and classification model is obtained by manually adjusting the classification parameters and taking the individual ranking provided by the feature ranking algorithm [80,81].…”

Section: Discussionmentioning

confidence: 99%

AutoML for Feature Selection and Model Tuning Applied to Fault Severity Diagnosis in Spur Gearboxes

Cerrada

Trujillo

Hernández

et al. 2022

MCA

Self Cite

View full text Add to dashboard Cite

Gearboxes are widely used in industrial processes as mechanical power transmission systems. Then, gearbox failures can affect other parts of the system and produce economic loss. The early detection of the possible failure modes and their severity assessment in such devices is an important field of research. Data-driven approaches usually require an exhaustive development of pipelines including models’ parameter optimization and feature selection. This paper takes advantage of the recent Auto Machine Learning (AutoML) tools to propose proper feature and model selection for three failure modes under different severity levels: broken tooth, pitting and crack. The performance of 64 statistical condition indicators (SCI) extracted from vibration signals under the three failure modes were analyzed by two AutoML systems, namely the H2O Driverless AI platform and TPOT, both of which include feature engineering and feature selection mechanisms. In both cases, the systems converged to different types of decision tree methods, with ensembles of XGBoost models preferred by H2O while TPOT generated different types of stacked models. The models produced by both systems achieved very high, and practically equivalent, performances on all problems. Both AutoML systems converged to pipelines that focus on very similar subsets of features across all problems, indicating that several problems in this domain can be solved by a rather small set of 10 common features, with accuracy up to 90%. This latter result is important in the research of useful feature selection for gearbox fault diagnosis.

show abstract

Section: Discussionmentioning

confidence: 99%

AutoML for Feature Selection and Model Tuning Applied to Fault Severity Diagnosis in Spur Gearboxes

Cerrada

Trujillo

Hernández

et al. 2022

MCA

Self Cite

View full text Add to dashboard Cite

show abstract

“…One commonly used approach is to fix Enc by computing statistical condition indicators from different domains of s c such as time, frequency, or time-frequency. Among the most common statistical indicators are RMS, standard deviation, kurtosis, and Shannon entropy [40]. The resulting vectors in each domain are concatenated to obtain a unique feature vector.…”

Section: Bigan-based Modelingmentioning

confidence: 99%

From fault detection to one-class severity discrimination of 3D printers with one-class support vector machine

Cabrera

Sancho

et al. 2021

ISA Transactions

Self Cite

View full text Add to dashboard Cite

The lack of faulty condition data reduces the feasibility of supervised learning for fault detection or fault severity discrimination in new manufacturing technologies. To deal with this issue, one-class learning arises for building binary discriminative models using only healthy condition data. However, these models have not been extrapolated to severity discrimination. This paper proposes to extend OCSVM, which is typically used for fault detection, to 3D printer fault severity discrimination. First, a set of features is extracted from a set of normal signals. An optimized OCSVM model is obtained by tuning the kernel and model hyperparameters. The resulting models are evaluated for fault detection and fault severity discrimination using a proposed performance evaluation approach. Experimental comparisons for belt-based faults in 3D printers show that the distance to the hyperplane has the information to discriminate the severity level, and its use is feasible. The proposed hyperparameter optimization technique improves the OCSVM for fault detection and severity discrimination compared to some other methods.

show abstract

“…This pave way for producing superior quality products. [8]The work performed the feature comparison that was extracted from current signals, acoustic emission, and vibration in the time domain for the determination of 8 level pit severity in gearbox. The acoustic emission, vibration and current signals was obtained with the use of gearbox test bed [9].…”

Section: Related Workmentioning

confidence: 99%

Fault Monitoring System for the Multistage Gear Box Based on MSTFT and GFLA Algorithm

Ghulanavar*,

Dama

2020

IJITEE

View full text Add to dashboard Cite

Gearbox functions as a significant transmission module in the mechanical devices in spite of its failure prone nature and hence there exists a need for the diagnosing the gearbox faults with an optimized solution and new methods should be engaged for improving the effectiveness, accuracy, reliability and few other such parameters. These attempts could meet the growing requirements for the condition monitoring in the detection of gear faults. The feature selection process is a notable process in the machine learning to achieve good performance in the diagnostic process. This framework builds up an innovative structure for fault diagnosis in the gearbox system. Six vibrating signals which are healthy gear signal, fault gear signal, healthy gear signal affected by noise, faulty gear signal affected by noise, distributed fault signal, and local fault signal have been generated with the prescribed dataset and the feature extraction was done by employing the Modified Short-Time Fourier Transform on the basis of Blackman window. This step was followed by selection of the features by using the combination of Genetic algorithm and Frog Leaping algorithm that is termed as GFLA.This novelty approach enhances the initialization process thereby enabling for the calculation of accurate best score calculation. The classification method is performed Support points on the basis of neural network. Finally the performance analysis with respect to the existing methodologies proved the efficiency in detecting the fault of the proposed framework.

show abstract

A comparative feature analysis for gear pitting level classification by using acoustic emission, vibration and current signals

Cited by 22 publications

References 18 publications

AutoML for Feature Selection and Model Tuning Applied to Fault Severity Diagnosis in Spur Gearboxes

AutoML for Feature Selection and Model Tuning Applied to Fault Severity Diagnosis in Spur Gearboxes

From fault detection to one-class severity discrimination of 3D printers with one-class support vector machine

Fault Monitoring System for the Multistage Gear Box Based on MSTFT and GFLA Algorithm

Contact Info

Product

Resources

About