The Fault Diagnosis and Monitoring of Rotating Machines by Thermography

Lim, Gang-Min; Ali, Younus; Yang, Bo‐Suk

doi:10.1007/978-0-85729-493-7_43

Cited by 8 publications

(5 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second step can consist of enhancing the image or ROI [10]. From this (enhanced) ROI, statistical features are derived such as the standard deviation, mean, skewness, kurtosis, variance, entropy, energy, central moments, maximum and minimum [7,10,11] or the components of the discrete wavelet decomposition of the thermal image [12,13]. In the penultimate step, undiscriminating features are sometimes removed or combined to create better features.…”

Section: Related Literaturementioning

confidence: 99%

“…In the penultimate step, undiscriminating features are sometimes removed or combined to create better features. Algorithms used for this step include Principal Component Analysis [9], independent component analysis [11], discriminant analysis [10] or relief algorithm [12,13]. The resulting features are subsequently used to determine the condition of the rotating machine using a classification algorithm.…”

Section: Related Literaturementioning

confidence: 99%

“…The resulting features are subsequently used to determine the condition of the rotating machine using a classification algorithm. The classification algorithms which have been employed are support vector machine [11,8], relevance vector machine [10], self-organising map [7] and linear discriminant analysis [12,13]. Most of these approaches will result in a system that can accurately detect the latent machine condition with an accuracy of 74% up to 100 % [7]- [13].…”

Section: Related Literaturementioning

confidence: 99%

See 2 more Smart Citations

Data-driven imbalance and hard particle detection in rotating machinery using infrared thermal imaging

Janssens

Loccufier

Walle

et al. 2017

Infrared Physics & Technology

View full text Add to dashboard Cite

Currently, temperature-based condition monitoring cannot be used to accurately identify potential faults early in a rotating machines' lifetime since temperature changes are only detectable when the fault escalates. However, currently only point measurements, i.e. thermocouples, are used. In this article, infrared thermal imaging is used which -as opposed to simple thermocouples-provides spatial temperature information. This information proves crucial for the identification of several machine conditions and faults. In this paper the conditions considered are outer-raceway damage in bearings, hard-particle contamination in lubricant and several gradations of shaft imbalance. The fault detection is done using an image processing and machine learning solution which can accurately detect the majority of the faults and conditions in our data set.

show abstract

Section: Related Literaturementioning

confidence: 99%

Section: Related Literaturementioning

confidence: 99%

Section: Related Literaturementioning

confidence: 99%

See 1 more Smart Citation

Data-driven imbalance and hard particle detection in rotating machinery using infrared thermal imaging

Janssens

Loccufier

Walle

et al. 2017

Infrared Physics & Technology

View full text Add to dashboard Cite

show abstract

“…Interpreting the phenomena in the infrared thermal (IRT) images requires however substantial insights into the mechanics and thermodynamics of the systems. As complete physics-based modelling of a machine is difficult and requires a lot of knowledge, effort and time [6], mainly data-driven feature engineering has been applied for IRT-based condition monitoring [7], [8], [9], [5].…”

Section: Introductionmentioning

confidence: 99%

Deep Learning for Infrared Thermal Image Based Machine Health Monitoring

Janssens

Walle

Loccufier

et al. 2018

IEEE/ASME Trans. Mechatron.

174

View full text Add to dashboard Cite

The condition of a machine can automatically be identified by creating and classifying features that summarize characteristics of measured signals. Currently, experts, in their respective fields, devise these features based on their knowledge. Hence, the performance and usefulness depends on the expert's knowledge of the underlying physics, or statistics. Furthermore, if new and additional conditions should be detectable, experts have to implement new feature extraction methods. To mitigate the drawbacks of feature engineering, a method from the sub-field of feature learning, i.e. deep learning, more specifically convolutional neural networks, is researched in this article. The objective of this article is to investigate if and how deep learning can be applied to infrared thermal video to automatically determine the condition of the machine. By applying this method on infrared thermal data in two use cases, i.e. machine fault detection and oil level prediction, we show that the proposed system is able to detect many conditions in rotating machinery very accurately (i.e. 95 % and 91.67 % accuracy for the respective use cases) without requiring any detailed knowledge about the underlying physics, and thus having the potential to significantly simplify condition monitoring using complex sensor data. Furthermore, we show that by using the trained neural networks, important regions in the infrared thermal images can be identified related to specific conditions which can potentially lead to new physical insights.

show abstract

“…Major benefit of this approach is the independence of background noise and machine operating conditions, however its drawbacks include processing complexity and classification. Thermal analysis based condition monitoring approaches [16,17] provide similar advantages and drawbacks. For improved fault diagnostics combinations of different approaches and analysis techniques can be applied [18,19].…”

Section: Introductionmentioning

confidence: 99%

A semi-analytical bearing model considering outer race flexibility for model based bearing load monitoring

Kerst

Shyrokau

Holweg

2018

Mechanical Systems and Signal Processing

View full text Add to dashboard Cite

This paper proposes a novel semi-analytical bearing model addressing flexibility of the bearing outer race structure. It furthermore presents the application of this model in a bearing load condition monitoring approach. The bearing model is developed as current computational low cost bearing models fail to provide an accurate description of the more and more common flexible size and weight optimized bearing designs due to their assumptions of rigidity. In the proposed bearing model raceway flexibility is described by the use of static deformation shapes. The excitation of the deformation shapes is calculated based on the modelled rolling element loads and a Fourier series based compliance approximation. The resulting model is computational low cost and provides an accurate description of the rolling element loads for flexible outer raceway structures. The latter is validated by a simulation-based comparison study with a well-established bearing simulation software tool. An experimental study finally shows the potential of the proposed model in a bearing load monitoring approach.

show abstract

The Fault Diagnosis and Monitoring of Rotating Machines by Thermography

Cited by 8 publications

References 3 publications

Data-driven imbalance and hard particle detection in rotating machinery using infrared thermal imaging

Data-driven imbalance and hard particle detection in rotating machinery using infrared thermal imaging

Deep Learning for Infrared Thermal Image Based Machine Health Monitoring

A semi-analytical bearing model considering outer race flexibility for model based bearing load monitoring

Contact Info

Product

Resources

About