A systematic review on failure modes and proposed methodology to artificially seed faults for promoting PHM studies in laboratory environment for an industrial gearbox

Goswami, Priyom; Nandan, Rajiv

doi:10.1016/j.engfailanal.2023.107076

Cited by 25 publications

(11 citation statements)

References 96 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The process of graph convolution for two-layer structured GCN is as follows [14]: where the input of GCN, H (0) , is just the node feature matrix E. W (0) ∈ R d×d ′ and W (1) ∈ R d ′ ×D are the learnable feature transformation matrices of the two layers, respectively. After the first layer, the dimension of each label node's feature vector changes from d = 300 (i.e.…”

Section: Multi-label Classifier Learning Modulementioning

confidence: 99%

See 1 more Smart Citation

A compound fault diagnosis model for gearboxes using correlation information between single faults

Zeng,

Wang,

Cheng

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

Gearboxes are key components of rotating machinery. Performing intelligent fault diagnosis of gearboxes with condition-based monitoring information helps to make reliable decisions on equipment operation and maintenance. Besides single faults, compound faults also are common failure forms of gearboxes. Conventional intelligent diagnosis models (known as single-label models) generally treat a compound fault as a new fault type, ignoring the correlations between the compound fault and the corresponding single faults. To overcome this problem, multi-label learning has been introduced and developed into multi-label models. It is also possible that different single faults are not independent but correlated with each other. Existing multi-label models, however, usually ignore this aspect. Therefore, exploiting the correlation information between single faults can further improve multi-label models. To this end, every single fault is treated as a label node , resulting in a label graph. The feature vector of each label node is initialized by the word embedding of the corresponding single-fault label. All the word embeddings are mapped using graph convolutional networks (GCN) into the parameter vectors of a set of interdependent binary linear classifiers that can directly perform multi-label classification on health categories. Meanwhile, the adjacency matrix of the label graph is adaptively learned by self-attention (SA) from node feature vectors. In this way, a novel multi-label model based on SA and GCN (referred to as SA-GCN) is proposed for compound fault diagnosis of gearboxes. SA-GCN mainly consists of a ResNet-based fault feature learning module, an SA-based adjacency matrix learning module, and a GCN-based multi-label classifier learning module. The application results on two gearbox cases show that SA-GCN outperforms conventional single-label models as well as state-of-the-art multi-label models in terms of both the diagnostic accuracy of compound faults and the overall diagnostic accuracy. Moreover, the effects of internal modules and hyperparameters on SA-GCN are also investigated.

show abstract

Section: Multi-label Classifier Learning Modulementioning

confidence: 99%

“…working environments, operating conditions, and running time), however, the components inside gearboxes (e.g. gears, bearings, and shafts) may experience various fault types [1,2]. These faults will affect the operation safety and reliability of gearboxes and even the whole rotating machinery.…”

Section: Introductionmentioning

confidence: 99%

A compound fault diagnosis model for gearboxes using correlation information between single faults

Zeng,

Wang,

Cheng

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…In these complex systems, rolling bearings serve as critical components. Monitoring their health status ensures stable equipment operation, reduces maintenance costs, and safeguards personnel safety [3]. However, existing bearing remaining useful life (RUL) prediction models often perform poorly under variable operating conditions and high-noise environments.…”

Section: Introductionmentioning

confidence: 99%

A lifetime prediction model based on two-path convolution with attention mechanism and bidirectional long short-term memory network

Sun,

Dong,

Bai

et al. 2024

Meas. Sci. Technol.

View full text Add to dashboard Cite

With the continuous progress of technology, the structure of modern industrial equipment is becoming increasingly complex. Complex industrial processes often have multivariable, nonlinear, variable operating conditions and intense noise, making it a challenging research direction to establish accurate residual service life prediction models.This paper constructs a life prediction model based on two-way convolution, attention mechanisms, and a two-way short-term memory network. The front end of the model uses a two-way convolution scale and attention module to mine critical fault information of bearings, improve the anti-noise ability of the model, and use adaptive batch normalization (AdaBN) and Meta-Aconc activation function to adaptively adjust neurons to enhance the generalization ability of the model, At the back end of the model, the bidirectional long-term and short-term memory network is used to memorize the degradation information of the bearing and the residual service life of the bearing is predicted.Finally, it has a high prediction accuracy in noise interference and condition migration scenarios using the root mean square error (RMSE), average absolute error (MAE), and other prediction indicators.This model provides a method reference for predicting the lifespan of rotating machinery under intense noise and variable operating conditions.

show abstract

“…These failure modes can cause operation downtime and unsafe conditions in gear systems, and this can be mitigated by proper health monitoring [32]. Among the monitoring techniques, vibration monitoring techniques are versatile for both theoretical and experimental studies.…”

Section: Introductionmentioning

confidence: 99%

Mathematical Complexities in Modelling Damage in Spur Gears

Oreavbiere,

Khan

2024

Machines

View full text Add to dashboard Cite

Analytical modelling is an effective approach to obtaining a gear dynamic response or vibration pattern for health monitoring and useful life prediction. Many researchers have modelled this response with various fault conditions commonly observed in gears. The outcome of such models provides a good idea about the changes in the dynamic response available between different gear health states. Hence, a catalogue of the responses is currently available, which ought to aid predictions of the health of actual gears by their vibration patterns. However, these analytical models are limited in providing solutions to useful life prediction. This may be because a majority of these models used single fault conditions for modelling and are not valid to predict the remaining life of gears undergoing more than one fault condition. Existing reviews related to gear faults and dynamic modelling can provide an overview of fault modes, methods for modelling and health prediction. However, these reviews are unable to provide the critical similarities and differences in the single-fault dynamic models to ascertain the possibility of developing models under combined fault modes. In this paper, existing analytical models of spur gears are reviewed with their associated challenges to predict the gear health state. Recommendations for establishing more realistic models are made especially in the context of modelling combined faults and their possible impact on gear dynamic response and health prediction.

show abstract

A systematic review on failure modes and proposed methodology to artificially seed faults for promoting PHM studies in laboratory environment for an industrial gearbox

Cited by 25 publications

References 96 publications

A compound fault diagnosis model for gearboxes using correlation information between single faults

A compound fault diagnosis model for gearboxes using correlation information between single faults

A lifetime prediction model based on two-path convolution with attention mechanism and bidirectional long short-term memory network

Mathematical Complexities in Modelling Damage in Spur Gears

Contact Info

Product

Resources

About