Neural Networks for Gas Turbine Fault Identification: Multilayer Perceptron or Radial Basis Network?

Loboda, Igor; Feldshteyn, Yakov; Ponomaryov, Volodymyr

doi:10.1515/tjj-2012-0005

Cited by 31 publications

(18 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this Section, four models are compared to our proposed models, which are the fuzzy logic method (FL) [21], extended Kalman filtering (EKF) [43], support vector machine (SVM) [14] and back propagation network (BPN) [12]. Fuzzy set theory and fuzzy logic provide the framework for nonlinear mapping.…”

Section: Comparison Of the Different Modelsmentioning

confidence: 99%

“…Model-based methods typically includes linear gas path analysis [4][5][6], nonlinear gas path analysis [7,8], Kalman filters [9,10] and expert systems [11]. Data driven-based methods typically include artificial neural networks [12,13], support vector machine [14,15], Bayesian approaches [16,17], genetic algorithms [18] and fuzzy reasoning [19][20][21].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Anomaly Detection in Gas Turbine Fuel Systems Using a Sequential Symbolic Method

Wang

Zhou

et al. 2017

Energies

View full text Add to dashboard Cite

Abstract:Anomaly detection plays a significant role in helping gas turbines run reliably and economically. Considering the collective anomalous data and both sensitivity and robustness of the anomaly detection model, a sequential symbolic anomaly detection method is proposed and applied to the gas turbine fuel system. A structural Finite State Machine is used to evaluate posterior probabilities of observing symbolic sequences and the most probable state sequences they may locate. Hence an estimation-based model and a decoding-based model are used to identify anomalies in two different ways. Experimental results indicate that both models have both ideal performance overall, but the estimation-based model has a strong robustness ability, whereas the decoding-based model has a strong accuracy ability, particularly in a certain range of sequence lengths. Therefore, the proposed method can facilitate well existing symbolic dynamic analysis-based anomaly detection methods, especially in the gas turbine domain.

show abstract

Section: Comparison Of the Different Modelsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Anomaly Detection in Gas Turbine Fuel Systems Using a Sequential Symbolic Method

Wang

Zhou

et al. 2017

Energies

View full text Add to dashboard Cite

show abstract

“…Our previous work [15] compares the Bayesian Approach (BA) and two artificial neural net works, Multilayer Perceptron (MLP) and Radial Basis Net work (RBN). The comparison has yielded an interesting result that all three techniques are equally accurate.…”

Section: Common Problem Of Diagnostic Algorithm Optimizationmentioning

confidence: 99%

“…The application of these tech niques for gas turbine fault isolation is described in [10,15]. In the next three subsections, we give only their brief description.…”

Section: Classification Techniquesmentioning

confidence: 99%

Gas Turbine Fault Diagnosis Using Probabilistic Neural Networks

Loboda

Olivares-Robles

2015

International Journal of Turbo &Amp; Jet-Engines

Self Cite

View full text Add to dashboard Cite

Efficiency of gas turbine monitoring systems pri marily depends on the accuracy of employed algorithms, in particular, pattern classification techniques for diag nosing gas path faults. In recent investigations many tech niques have been applied to classify gas path faults, but recommendations for selecting the best technique for real monitoring systems are still insufficient and often con tradictory. In our previous work, three classification tech niques were compared under different conditions of gas turbine diagnosis. The comparative analysis has shown that all these techniques yield practically the same accu racy for each comparison case. The present contribution considers a new classification technique, Probabilistic Neural Network (PNN), and we compare it with the tech niques previously examined. The results for all compari son cases show that the PNN is not inferior to the other techniques. We recommend choosing the PNN for real monitoring systems because it has an important advan tage of providing confidence estimation for every diagnos tic decision made.

show abstract

“…Model based methods are typically includes linear gas path analysis [4][5][6], nonlinear gas path analysis [7,8], Kalman filters [9,10] and expert systems [11]. Data driven based methods typically include artificial neural networks [12,13], support vector machine [14,15], Bayesian approaches [16,17], genetic algorithms [18] and fuzzy reasoning [19][20][21].…”

Section: Introductionmentioning

confidence: 99%

Anomaly Detection on Gas Turbine Fuel System Using a Sequential Symbolic Method

Li¹,

Wang²,

Zhou³

et al. 2017

Preprint

View full text Add to dashboard Cite

Anomaly detection plays a significant role in helping gas turbines run reliably and economically. Considering collective anomalous data and both sensitivity and robustness of the anomaly detection model, a sequential symbolic anomaly detection method is proposed and applied to the gas turbine fuel system. A structural Finite State Machine is to evaluate posterior probabilities of observing symbolic sequences and most probable state sequences they may locate.Hence an estimating based model and a decoding based model are used to identify anomalies in two different ways. Experimental results indicates that these two models have both ideal performance overall, and estimating based model has a strong ability in robustness, while decoding based model has a strong ability in accuracy, particularly in a certain range of length of sequence. Therefore, the proposed method can well facilitate existing symbolic dynamic analysis based anomaly detection methods especially in gas turbine domain.

show abstract

Neural Networks for Gas Turbine Fault Identification: Multilayer Perceptron or Radial Basis Network?

Cited by 31 publications

References 0 publications

Anomaly Detection in Gas Turbine Fuel Systems Using a Sequential Symbolic Method

Anomaly Detection in Gas Turbine Fuel Systems Using a Sequential Symbolic Method

Gas Turbine Fault Diagnosis Using Probabilistic Neural Networks

Anomaly Detection on Gas Turbine Fuel System Using a Sequential Symbolic Method

Contact Info

Product

Resources

About