A novel approach to gas turbine fault diagnosis based on learning of fault characteristic maps using hybrid residual compensation extreme learning machine-growing neural gas model

Montazeri-Gh, Morteza; Nekoonam, Ali; Yazdani, Shabnam

doi:10.1007/s40430-021-03136-9

Cited by 6 publications

(4 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared with multilayer perception (MLP), the classification accuracy of this method is more than 10% higher, and the classification accuracy for three component faults is 99.4%. Montazeri Gh M et al [28] combined a growth neural network (GNN) and residual compensation limit learning machine (RCELM) to learn the fault characteristic map (FCM) of components to diagnose fouling and erosion faults in compressors, gas generators, and power turbines. The user inputs known measurement parameters into an extreme learning machine (ELM) to train and output health status bias.…”

Section: Introductionmentioning

confidence: 99%

Fast Prediction Method of Combustion Chamber Parameters Based on Artificial Neural Network

Shao,

Liu,

Zhang

et al. 2023

Electronics

View full text Add to dashboard Cite

Gas turbines are widely used in industry, and the combustion chamber, compressor, and turbine are known as their three important components. In the design process of the combustion chamber, computational fluid dynamics simulation takes up a lot of time. In order to accelerate the design speed of the combustion chamber, this article proposes a combustion chamber design method that combines an artificial neural network (ANN) and computational fluid dynamics (CFD). CFD results are used as raw data to establish a fast prediction model using ANN and eXtreme Gradient Boosting (XGBoost). The results show that the mean squared error (MSE) of the ANN is 0.0019, and the MSE of XGBoost is 0.0021, so the ANN’s prediction performance is slightly better. This fast prediction method combines CFD and the ANN, which can greatly shorten CFD calculation time, improve the efficiency of gas turbine combustion chamber design, and provide the possibility of achieving digital twins of gas turbine combustion chambers.

show abstract

Section: Introductionmentioning

confidence: 99%

Fast Prediction Method of Combustion Chamber Parameters Based on Artificial Neural Network

Shao,

Liu,

Zhang

et al. 2023

Electronics

View full text Add to dashboard Cite

show abstract

“…Traditional fault diagnosis methods rely on signal processing techniques and expert knowledge, while deep learning can extract data features automatically, making it gain widespread attention and surpass inherent limitations [1][2][3][4][5][6][7][8][9][10][11].…”

Section: Introductionmentioning

confidence: 99%

Fault diagnosis method for unbalance data based on Gramian angular field

Yu¹,

Cao

et al. 2023

Preprint

View full text Add to dashboard Cite

The commonplace practice in deep learning is to make use of a balanced dataset, whereby the number of fault data is equivalent to that of normal data. However, in the context of real-world applications, the number of fault data is generally insufficient compared to that of normal data. In this study, a new approach for diagnosing faults in unbalanced data sets is proposed using the Gramian Angular Field(GAF) method. Firstly, the Gramian Angular Field (GAF) method is employed to convert one-dimensional data into two-dimensional data, which enhances the feature extraction process. Secondly, to balance the sample distribution, fault data is generated using Generative Adversarial Networks (GANs).Finally, the residual neural network (ResNet) with an attention mechanism is utilized to improve the accuracy of fault diagnosis. The proposed method was experimentally validated using open-source bearing datasets that were published by Case Western Reserve University and the University of Ottawa. The experimental results show that the proposed method has greatly improved fault diagnosis performance in cases of data distribution imbalance, surpassing that of the compared methods.

show abstract

“…[9][10][11][12] Montazeri-Gh et al proposed a novel approach based on learning the fault characteristic maps of gas turbine components using an ELM. 13 Fentaye presented a gas turbine fault detection and isolation method using modular convolutional neural networks preceded by a physics-driven performance-trendmonitoring system. 14 Montazeri-Gh and Nekoonam 15 applied a component fault diagnostic system based on a bank of online sequential extreme learning machines (OSELMs) that can be used for gas path fault diagnosis.…”

Section: Introductionmentioning

confidence: 99%

Method of aeroengine fault diagnosis based on a Fingerprint map

Gou

Pan³

et al. 2023

Proceedings of the Institution of Mechanical Engineers, Part G:

View full text Add to dashboard Cite

The electrification and intelligentization trends of aeroengine systems present new challenges for the diagnostics of traditionally centralized control architectures. Scientific and practical fault diagnosis is of great significance for reducing aeroengine failures, improving operation efficiency, and ensuring safety. To realize accurate fault diagnosis, an aeroengine fault diagnosis method based on an improved fingerprint map is proposed in this paper. Based on aeroengine data analysis, ISTOA-SVM diagnostic model that embeds improved fingerprint map data as a priori knowledge is adopted to identify the fault sets. Data based on the statistics of the fingerprint map are processed with a ratio coefficient, and the feature parameters are appended through Floyd-principal component analysis (Floyd-PCA) to increase the identification data dimension. On the other hand, to improve the diagnostic performance of the classifier, the sooty terns optimization algorithm algorithm combined with chaotic maps, a variable helix mechanism and a separating operator is designed, and the ISTOA-SVM is constructed. The fault diagnosis method was validated by using a real-world dataset of aeroengines. Comparison experiments show that the proposed method is superior in performance over the state-of-the-art approaches in diagnostic precision under the constraint of limited data. The ISTOA-SVM addresses the issue that similar features may be confused in traditional fingerprint maps. The results are presented through 5 practical cases. The study demonstrates that the proposed method achieves superior performance and can be applied to aeroengine fault diagnosis.

show abstract

A novel approach to gas turbine fault diagnosis based on learning of fault characteristic maps using hybrid residual compensation extreme learning machine-growing neural gas model

Cited by 6 publications

References 65 publications

Fast Prediction Method of Combustion Chamber Parameters Based on Artificial Neural Network

Fast Prediction Method of Combustion Chamber Parameters Based on Artificial Neural Network

Fault diagnosis method for unbalance data based on Gramian angular field

Method of aeroengine fault diagnosis based on a Fingerprint map

Contact Info

Product

Resources

About