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

2011 ◽  
Author(s):  
Igor Loboda ◽  
Yakov Feldshteyn ◽  
Volodymyr Ponomaryov
Keyword(s):  
Neural Networks ◽  
Gas Turbine ◽  
Multilayer Perceptron ◽  
Input Data ◽  
Correct Diagnosis ◽  
Computation Time ◽  
Fault Identification ◽  
Radial Basis Network ◽  
Comparison Results ◽  
Radial Basis

Efficiency of gas turbine condition monitoring systems depends on quality of diagnostic analysis at all its stages such as feature extraction (from raw input data), fault detection, fault identification, and prognosis. Fault identification algorithms based on the gas path analysis may be considered as an important and sophisticated component of these systems. These algorithms widely use pattern recognition techniques, mostly different artificial neural networks. In order to choose the best technique, the present paper compares two network types: a multilayer perceptron and a radial basis network. The first network is being commonly applied to recognize gas turbine faults. However, some studies note high recognition capabilities of the second network. For the purpose of the comparison, both networks were included into a special testing procedure that computes for each network the true positive rate that is the probability of a correct diagnosis. Networks were first tuned and then compared using this criterion. Same procedure input data were fed to both networks during the comparison. However, to draw firm conclusions on the networks’ applicability, comparative calculations were repeated with different variations of these data. In particular, two engines that differ in an application and gas path structure were chosen as a test case. By way of summing up comparison results, the conclusion is that the radial basis network is a little more accurate than the perceptron, however the former needs much more available computer memory and computation time.

scholarly journals Estudo sobre Métricas para Definir Reputação do Autor de Comentários em Sites de Vendas de Produtos

2018 ◽  
Author(s):  
Carlos Augusto De Sá ◽  
Raimundo Santos Moura
Keyword(s):  
Radial Basis Function ◽  
Basis Function ◽  
Multilayer Perceptron ◽  
Radial Basis

Conhecer a reputação do autor de textos opinativos é de suma importância para avaliação de comentários na Web. Este artigo apresenta um estudo sobre medidas usadas no processo de avaliação da reputação do autor em sites de vendas de produtos. Realizou-se dois experimentos com as redes neurais Multilayer Perceptron (MLP) e Radial Basis Function (RBF), sendo que a rede MLP obteve melhor desempenho. Comparou-se também a abordagem TOP(X) original, usada para inferir os melhores comentários, com um novo modelo que utiliza rede MLP na dimensão da reputação do autor. Considerando os comentários excelentes e bons, a nova abordagem apresentou resultados significativamente superiores.

Ionic liquids: Determination of their aqueous content using differential scanning calorimeter equipment, chaotic parameters and a radial basis network model

Talanta ◽  
2010 ◽  
Vol 81 (4-5) ◽  
pp. 1766-1771 ◽  
Author(s):  
José S. Torrecilla ◽  
Ester Rojo ◽  
Juan C. Domínguez ◽  
Francisco Rodríguez
Keyword(s):  
Ionic Liquids ◽  
Network Model ◽  
Differential Scanning Calorimeter ◽  
Radial Basis Network ◽  
Radial Basis

Cancellable Biometrics for On-line Signature Recognition

2011 ◽  
pp. 290-315 ◽  
Author(s):  
Emanuele Maiorana ◽  
Patrizio Campisi ◽  
Alessandro Neri
Keyword(s):  
Recognition System ◽  
Design Of Algorithms ◽  
Biometric Data ◽  
The Public ◽  
Protection Scheme ◽  
Protection Method ◽  
Signature Database ◽  
On Line ◽  
Signature Recognition ◽  
Recognition Systems

With the widespread diffusion of biometrics-based recognition systems, there is an increasing awareness of the risks associated with the use of biometric data. Significant efforts are therefore being dedicated to the design of algorithms and architectures able to secure the biometric characteristics, and to guarantee the necessary privacy to their owners. In this work we discuss a protected on-line signature-based biometric recognition system, where the considered biometrics are secured by applying a set of non-invertible transformations, thus generating modified templates from which retrieving the original information is computationally as hard as random guessing it. The advantages of using a protection method based on non-invertible transforms are exploited by presenting three different strategies for the matching of the transformed templates, and by proposing a multi-biometrics approach based on score-level fusion to improve the performances of the considered system. The reported experimental results, evaluated on the public MCYT signature database, show that the achievable recognition rates are only slightly affected by the proposed protection scheme, which is able to guarantee the desired security and renewability for the considered biometrics.

Sign in / Sign up

Export Citation Format

Share Document