Online learning neural architectures and cross-correlation analysis for actuator failure detection and identification

Napolitano, Marcello R.; Casdorph, Van; Neppach, Charles; Naylor, Steve; Innocenti, Mario; Silvestri, Giovanni

doi:10.1080/00207179608921851

Cited by 39 publications

(20 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Stability of voting in learning is investigated by Turney (1994aTurney ( , 1994b. In learning, using small data sets, one or more elements can be left out for testing, the bulk of the data being used for training (Napolitano et al 1996;Opper and Winther 1996). The process of cross-validation can be performed leaving out one element or many, as in kfold cross-validation (Turney 1994a(Turney , 1994b.…”

Section: Discussionmentioning

confidence: 99%

Voting as Validation in Robot Programming

Utete

Barshan²,

Ayrulu

1999

The International Journal of Robotics Research

View full text Add to dashboard Cite

show abstract

Section: Discussionmentioning

confidence: 99%

Voting as Validation in Robot Programming

Utete

Barshan²,

Ayrulu

1999

The International Journal of Robotics Research

View full text Add to dashboard Cite

show abstract

“…The final size of the allocated MRAN and EMRAN networks were well within the above limit; in fact, only 48 and 49 basis functions were allocated, respectively. For the MLP-NN the design guidelines in [33] were followed to select a suitable architecture to provide the required mapping accuracy while minimizing its complexity. As in the previous case, the EBPA training algorithm was applied 100 times to the flight data.…”

Section: Off-line Training Phasementioning

confidence: 99%

On‐line learning neural networks for sensor validation for the flight control system of a B777 research scale model

Campa

Fravolini

Seanor

et al. 2002

Intl J Robust & Nonlinear

Self Cite

View full text Add to dashboard Cite

SUMMARYThis paper focuses on the analysis of a scheme for sensor failure, detection, identification and accommodation (SFDIA) using experimental flight data of a research aircraft model. Recent technical literature has shown the advantages of time-varying estimators and/or approximators. Conventional approaches are based on different versions of observers and Kalman filters while more recent methods are based on different approximators based on neural networks (NNs). The approach proposed in the paper is based on the use of on-line learning nonlinear neural approximators.The characteristics of three different neural architectures were compared through different sensor failures. The first architecture is based on a multi layer perceptron (MLP) NN trained with the extended back propagation algorithm (EBPA). The second and third architectures are based on a radial basis function (RBF) NN trained with the minimal resource allocating network (MRAN) and extended-MRAN (EMRAN). The MRAN and EMRAN algorithms have recently been developed for RBF networks and have shown remarkable learning capabilities at a fraction of the memory requirements and computational effort typically associated with conventional RBF NNs.The experimental data for this study are flight data acquired from the flight-testing of a 1 24 th semi-scale B777 research model designed, built, and flown at West Virginia University (WVU).

show abstract

“…Therefore, these techniques are required in this problem type, where the detection and the isolation of the fault for the actuator and sensors of specify system will be guaranteed with desired performances. Many works are used neural network methods to adopt the detection sensor/actuator failures for various systems as presented in [7][8][9][10][11][12][13]. Nowadays, asynchronous motors are widely used in the industries.…”

Section: Introductionmentioning

confidence: 99%

Sensor and actuator fault detection and isolation based on artificial neural networks and fuzzy logic applicated on induction motor

Adouni

Hamed

Aymen

et al. 2013

2013 International Conference on Control, Decision and Information Technologies (CoDIT)

View full text Add to dashboard Cite

This paper presents a scheme for fault detection and isolation (FDI). It deals with sensors and actuator fault of an induction machine. This scheme is established with artificial intelligent techniques in order to resolve two big troubles. The first is the detection problem. It is resolved with the neural network and the second is the isolation difficulty, it solved using the fuzzy logic. The proposed FDI approach is implemented on Matlab/Simulink software and tested under three types of fault (current, speed sensor fault and inverter fault). The obtained results improving the importance of this method. Then, the actuator and sensor fault are detected and isolated successfully.

show abstract

Online learning neural architectures and cross-correlation analysis for actuator failure detection and identification

Cited by 39 publications

References 16 publications

Voting as Validation in Robot Programming

Voting as Validation in Robot Programming

On‐line learning neural networks for sensor validation for the flight control system of a B777 research scale model

Sensor and actuator fault detection and isolation based on artificial neural networks and fuzzy logic applicated on induction motor

Contact Info

Product

Resources

About