Fusing neural networks, genetic algorithms and fuzzy logic for diagnosis of cracks in shafts

Saridakis, Kostas M.; Chasalevris, Athanasios; Dentsoras, A.J.; Papadopoulos, Chris

doi:10.1016/b978-008045157-2/50061-4

Cited by 7 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In literature there are suitable combinations of fuzzy, genetic and neural techniques for diff erent applications [9], [17], [16]. In [1] hybrid paradigms are successfully implemented to solve three prominent robot control issues.…”

Section: Introductionmentioning

confidence: 99%

Designing neural networks to improve timing performances of intelligent controllers

Pelusi

2013

Journal of Discrete Mathematical Sciences and Cryptography

View full text Add to dashboard Cite

Intelligent techniques are applied to improve the control methods of physical quantities. Many researchers employ the fuzzy logic combined with genetic procedures to achieve good control results. Such intelligent techniques can be enhanced using Neural Networks. In this paper, an optimization algorithm to define the best training set for suitable Neural Networks is designed. The network trained with the optimal sample is introduced into genetic-fuzzy controllers to improve the timing performances. The results show that the genetic-neuro-fuzzy controllers are better than genetic-fuzzy controllers in terms of settling time and rise time. © 2013 Taru Publications

show abstract

Section: Introductionmentioning

confidence: 99%

Designing neural networks to improve timing performances of intelligent controllers

Pelusi

2013

Journal of Discrete Mathematical Sciences and Cryptography

View full text Add to dashboard Cite

show abstract

“…Then, GA was used to find the characteristics of the cracks through ANNs that approximated the analytical model. Also, Saridakis et al [21], in another study, introduced a framework for implementing ANN, GA, and FL for identifying cracks in rotating shafts while diminishing the required computational time. The reduction in computational time was achieved by approximating the analytical model with an ANN and replacing the exhaustive search of the solution space with a GA whose objective function relies on an FL representation.…”

Section: Introductionmentioning

confidence: 99%

Analysis and optimization of a transcritical power cycle with regenerator using artificial neural networks and genetic algorithms

Rashidi

Bég

Parsa

et al. 2011

Proceedings of the Institution of Mechanical Engineers, Part A:

View full text Add to dashboard Cite

This article presents a parametric study for and optimization of a transcritical power cycle. First, thermal efficiency, exergy efficiency, and specific network are selected as objective functions for parametric optimization. In order to optimize these functions, a procedure based on artificial neural networks (ANNs) and genetic algorithms (GAs) is proposed. This procedure comprises three steps.Step 1 is to find thermal efficiency, exergy efficiency, and specific network for different values of inlet turbine pressure, inlet turbine temperature, and fraction of the maximum power using the robust numerical code, engineering equation solver. In step 2, three distinct multi-layer perceptron ANNs based on the data obtained from step 1 are trained. In step 3, three distinct GAs are used to optimize the thermal efficiency, exergy efficiency, and specific network. The variables and fitness functions in these algorithms constitute, respectively, the inputs and outputs of the corresponding trained neural networks. For the purpose of validation of this study, for a special case, the results were compared with a previously reported case and were found to be in good agreement. Also in this article, this optimization process is applied to four different working fluids. Several interesting features among optimal objective functions and decision variables involved in the transcritical power cycle are identified.

show abstract

“…Moreover, it is believed that the second super-harmonic component at half the first critical frequency is a good indicator of the crack, and observing the super-harmonic component during start-up/coast-won is more useful than during steady-state operation [6,7]. In addition, with the advance in signal processing methods, more and more research efforts have been made to extract the signatures of crack from the sampled vibration signals using the now so-called advanced signal processing methods, such as the wavelet transform [8], Wigner-Ville distribution [9] and the Hilbert-Huang transform [10] and et al Some other crack detection methods include the model-based methods [11], which are usually based on analytical or numerical models to simulate the behavior of cracked shafts during operation and to attempt correlate the observed vibration signature with the presence of a crack at discrete locations on the shaft, and the artificial methods [12,13], in which the problem of crack detection is normally regarded as a pattern reorganization problem and the fault features are often required to be prior extracted from the sampled vibration signals. A comprehensive review, which covers most literature about the crack detection for shaft up to 2003, was contributed by Sabnavis [14] and et al…”

Section: Introductionmentioning

confidence: 99%

The Effects of Crack on the Transmission Matrix of Rotor Systems

Peng

Lang

Meng

et al. 2011

Shock and Vibration

View full text Add to dashboard Cite

The dynamic behavior of rotor containing crack is a subject of particular interest and has been extensively investigated by researchers. The effects of crack on the natural frequencies and modal shapes and motion orbits of rotor systems have already been well explored by researchers. In the present study, the infl uence of crack on the transmission matrices of the rotor systems is investigated by using the FEM (finite element method) analysis and the HBM (harmonic balance method) technique. It is for the first time revealed that there are differences between the transmission matrices for the fundamental frequency components and the transmission matrices for the super-harmonic components, and the differences are mainly determined by the crack location. The results are validated by numerical experiments where the system responses of a rotor system are obtained usingRunge-Kuttamethod. The results are of significance for the development of effective crack detection methods in practice.

show abstract

Fusing neural networks, genetic algorithms and fuzzy logic for diagnosis of cracks in shafts

Cited by 7 publications

References 9 publications

Designing neural networks to improve timing performances of intelligent controllers

Designing neural networks to improve timing performances of intelligent controllers

Analysis and optimization of a transcritical power cycle with regenerator using artificial neural networks and genetic algorithms

The Effects of Crack on the Transmission Matrix of Rotor Systems

Contact Info

Product

Resources

About