### Citations

307 | Damage identification and health monitoring of structural and mechanical systems from changes in their vibration characteristics: A literature review,
- Doebling, Farrar, et al.
- 1996
(Show Context)
Citation Context ...s using a parametric model, such as ARMA, insdiagnosing the changes in physical parameters ofsstructures. These researches can be grouped in timesdomain category of structural health monitorings(SHM) =-=[1]-=- where measuring the input of structures issoften necessary for damage detection, in other words,sthe exogenous part of ARMAX model must be usedsfor SHM [2-5]. Sohn et al. [2] have proposed a twostage... |

232 |
Genetic Fuzzy Systems: Evolutionary Tuning and Learning of Fuzzy Knowledge Bases, World Scientific: Singapore,
- Cordon, Herrera, et al.
- 2001
(Show Context)
Citation Context ...roduct operators(max operator), allowing for interaction between thespropositions in the antecedent. The degree ofsactivation of the ith rule is calculated as:s1 1, 2,..., n i ij c j D A i N = = = . =-=(12)-=-sThe output of the fuzzy classifier ( y ) is determinedsby the rule that has the highest degree of activation:s* 1 , * arg max c i i i N y class i D = = . (13) In this paper the inputs (features) ... |

63 | Structural Health Monitoring Using Statistical Process Control.
- Sohn, JJ, et al.
- 2000
(Show Context)
Citation Context ... health monitorings(SHM) [1] where measuring the input of structures issoften necessary for damage detection, in other words,sthe exogenous part of ARMAX model must be usedsfor SHM [2-5]. Sohn et al. =-=[2]-=- have proposed a twostage time series analysis combining ARX and ARsprediction models as pattern recognition techniquesfor SHM where main object was to extract featuressand to construct a statistical ... |

62 |
ANeuro-Fuzzy Method to learn Fuzzy Classification Rules from Data',
- Nauck, Kruse
- 1997
(Show Context)
Citation Context ...lement numbersand the degree of its severity. For every feature thesantecedent fuzzy sets (membership function) may besdefined as:s2 ( ) exp 0.5 j j j i i i j i f mfµ = × s! " , =-=(14)-=- where, ( )jifµ is ith membership function of jthsfeature, jif is the absolute value,sj im is the relatedsmean value (midpoint) and ji is the variance of thesjth feature in ith class. Features are ra... |

46 | Generating the knowledge base of a fuzzy rule-based system by the genetic learning of the data base,”
- Cordon, Herrera, et al.
- 2001
(Show Context)
Citation Context ...al feature space and thesrule consequent is a crisp (non-fuzzy) class labelsfrom { }1, 2, ..., cN [20]:s1 1 2 2 : .... 1,2,.. i i i n in i c R if x is A and x is A and x is A then outputis class i N= =-=(11)-=-sHere n denotes the number of features,s{ }1 2, ,..., T nX x x x= is the input vector (features),siclass is the output of the ith rule ands1 2, ,...,i i inA A A are the antecedent fuzzy sets. Thes“and... |

10 |
Structural damage detection and identification using fuzzy logic,"
- Sawyer, Rao
- 2000
(Show Context)
Citation Context ...sby white Gaussian noise and is measured on aspredefined point of structure for fault diagnosis. Ins[6] a genetic fuzzy system is proposed for cracksdetection in beams and helicopter rotor blades. In =-=[7]-=-sa generalized methodology for structural faultsdetection using FE and fuzzy logic is presented. Ins[8] a fuzzy rule-based system is used for damagesdetection of blade in a helicopter rotor which issm... |

9 |
ReducedDimensionality Geometric Approach to Fault Identification in Stochastic Structural Systems
- Sadeghi, Fassois
- 1998
(Show Context)
Citation Context ...at the prediction error of the neuralsnetwork will increase because of damage. Based onsthis premise, damage classifier is constructed using asnew damage detection method, proposed in [3]. In [4]sand =-=[5]-=- a geometric approach is proposed based onsARMAX modeling of a stochastic structure excitedsby white Gaussian noise and is measured on aspredefined point of structure for fault diagnosis. Ins[6] a gen... |

8 |
Structural Damage Classification, Using Extreme Value
- Sohn, Allen, et al.
- 2005
(Show Context)
Citation Context ...ere main object was to extract featuressand to construct a statistical model that distinguishessthe signals recorded under different structuralsconditions of a boat. In the recent work of Sohn et al.s=-=[3]-=-, the ARX model parameters are fed to an autosassociative neural network which is trained toscharacterize the underlying dependency of extractedsfeatures (parameter of ARX model) on thesunmeasured env... |

8 | A Fuzzy Logic System for Ground Based Structural Health Monitoring of a Helicopter Rotor using Modal Data
- Ganguli
(Show Context)
Citation Context ... a genetic fuzzy system is proposed for cracksdetection in beams and helicopter rotor blades. In [7]sa generalized methodology for structural faultsdetection using FE and fuzzy logic is presented. Ins=-=[8]-=- a fuzzy rule-based system is used for damagesdetection of blade in a helicopter rotor which issmodeled as a cantilever beam.sIn this work, a new method of damage detectionsand locating is proposed ba... |

4 |
Geometric Approach to the Non-destructive Identification of Fault in
- Sadeghi, Fassois
- 1997
(Show Context)
Citation Context ...ected that the prediction error of the neuralsnetwork will increase because of damage. Based onsthis premise, damage classifier is constructed using asnew damage detection method, proposed in [3]. In =-=[4]-=-sand [5] a geometric approach is proposed based onsARMAX modeling of a stochastic structure excitedsby white Gaussian noise and is measured on aspredefined point of structure for fault diagnosis. Ins[... |

4 |
Tuning of a Fuzzy Classifier Derived from Data
- Abe, Lan, et al.
- 1996
(Show Context)
Citation Context ..., 2,..., n i ij c j D A i N = = = . (12)sThe output of the fuzzy classifier ( y ) is determinedsby the rule that has the highest degree of activation:s* 1 , * arg max c i i i N y class i D = = . =-=(13)-=- In this paper the inputs (features) of fuzzysclassification system are ARMA model’s parameterssand its outputs of it is the damaged element numbersand the degree of its severity. For every feature th... |

4 |
System Identification
- Soderstorm, Stoica
- 1989
(Show Context)
Citation Context ...parameter in ith run (nosoverlapping window) and N is the number ofswindows. It is noted that the assumption of Gaussiansdistribution of parameters is admissible in PEsestimation method of ARMA model =-=[15, 16]-=-. In nextssection by using the mean value of î ( ) andsvariance of î derived from covariance matrix ( p ),sthe parameters of membership function of eachsARMA model’s parameters (used as features ... |

2 |
System Identification: Theory for the User
- Lijung
- 1987
(Show Context)
Citation Context ...ion is that thescovariance matrix of parameters are obtained duringsthe estimation process, in other words it is assumedsthat the parameters are random variables asymptoticswith Gaussian distribution =-=[15]-=-. By this way the mostsdifficult stage of fuzzy classification method, whichsis finding the uncertainty bound, can be eliminated orswww.SID.ir Ar ch ivesofsSID Mech. & Aerospace Eng. J. Vol. 3, No. 2,... |

1 |
Genetic Fuzzy for Damage Detection in Beams and Helicopter Rotor
- Pawar, Ganguli
- 2003
(Show Context)
Citation Context ...]sand [5] a geometric approach is proposed based onsARMAX modeling of a stochastic structure excitedsby white Gaussian noise and is measured on aspredefined point of structure for fault diagnosis. Ins=-=[6]-=- a genetic fuzzy system is proposed for cracksdetection in beams and helicopter rotor blades. In [7]sa generalized methodology for structural faultsdetection using FE and fuzzy logic is presented. Ins... |

1 |
A Polynomial Algebraic Method for Non-Stationary TARMA Signal Analysis- part
- Mrad, Fassois, et al.
- 1998
(Show Context)
Citation Context ...aspect of this paper is usingsonly the output data (response) of structure excitedsby white Gaussian noise and modeling the responsesdata by ARMAsmodel and system identificationsalgorithm proposed in =-=[9]-=- for damage diagnosing.sAnother very important aspect of this work isseliminating the optimization or tuning process whichsis a necessary stage in fuzzy rule based systems. Thesoptimization stage is v... |

1 |
Automatic Design of a Fuzzy Systems by Genetic Algorithms”, The Fuzzy Logic and
- Heider, Tryba, et al.
- 1995
(Show Context)
Citation Context ...atrix ofsparameters with Gaussian pdf:swww.SID.ir Ar ch ivesofsSID 88sMech. & Aerospace Eng. J. Vol. 3, No. 2, September 2007s1 1 ˆ ˆ ˆ( )( ) N i i N T i i x i p = = = = , =-=(10)-=-swhere, î is the estimated parameter in ith run (nosoverlapping window) and N is the number ofswindows. It is noted that the assumption of Gaussiansdistribution of parameters is admissible in PEsest... |