Corpus ID: 15390747

FAULT DETECTION AND CLASSIFICATION OF AN ELECTROHYDROSTATIC ACTUATOR USING A NEURAL NETWORK TRAINED BY THE SMOOTH VARIABLE STRUCTURE FILTER

@inproceedings{Ahmed2011FAULTDA,
  title={FAULT DETECTION AND CLASSIFICATION OF AN ELECTROHYDROSTATIC ACTUATOR USING A NEURAL NETWORK TRAINED BY THE SMOOTH VARIABLE STRUCTURE FILTER},
  author={R. Ahmed and S. A. Gadsden and M. E. Sayed and S. Habibi},
  year={2011}
}
A multilayered neural network is a multi-input, multioutput (MIMO) nonlinear system in which training can be regarded as a nonlinear parameter estimation problem by estimating the network weights. In this paper, the relatively new smooth variable structure filter (SVSF) is used for the training of a nonlinear multilayered feed forward network. The SVSF is a recursive sliding mode parameter and state estimator that has a predictor-corrector form. Using a switching gain, a corrective term is… Expand

Figures from this paper

SMOOTH VARIABLE STRUCTURE FILTERING: THEORY AND APPLICATIONS
Filtering strategies play an important role in estimation theory, and are used to extract knowledge of the true states typically from noisy measurements or observations made of the system. The nameExpand
A Review of Smooth Variable Structure Filters: Recent Advances in Theory and Applications
The smooth variable structure filter (SVSF) is a relatively new state and parameter estimation technique. Introduced in 2007, it is based on the sliding mode concept, and is formulated in aExpand

References

SHOWING 1-10 OF 12 REFERENCES
A real-time learning algorithm for a multilayered neural network based on the extended Kalman filter
TLDR
A new real-time learning algorithm for a mul- tilayered neural network that approximately gives the mini- mum variance estimate of the linkweights and the convergence performance is improved in comparison with the backwards error propagation algorithm using the steepest descent tech- niques. Expand
The Smooth Variable Structure Filter
  • S. Habibi
  • Mathematics, Computer Science
  • Proceedings of the IEEE
  • 2007
TLDR
The SVSF method is model based and applies to smooth nonlinear dynamic systems and allows for the explicit definition of the source of uncertainty and can guarantee stability given an upper bound for uncertainties and noise levels. Expand
Decoupled extended Kalman filter training of feedforward layered networks
TLDR
These studies demonstrate that the judicious grouping of weights along with the use of artificial process noise in DEKF result in input-output mapping performance that is comparable to the global extended Kalman algorithm, and is often superior to SBP, while requiring significantly fewer presentations of training data than SBP and less overall training time than either of these procedures. Expand
Parameter Identification for a High-Performance Hydrostatic Actuation System Using the Variable Structure Filter Concept
Parameter estimation is an important concept that can be used for health and condition monitoring. Estimation or measurement of physically meaningful parameters and their evaluation againstExpand
A Smooth Variable Structure filter for State estimation
TLDR
The reaching stability of the SVSF (closely related to that of Variable Structure Control (VSC)), is verified by a mathematical proof and the mathematical foundation behind, and the methodology of SVSf are presented. Expand
A new form of the smooth variable structure filter with a covariance derivation
  • S. Gadsden, S. Habibi
  • Mathematics, Computer Science
  • 49th IEEE Conference on Decision and Control (CDC)
  • 2010
TLDR
The smooth variable structure filter (SVSF) is introduced in a new form without affecting its original proof of stability, and the derivation of a covariance matrix is outlined that can be used for comparative purposes as well as other applications. Expand
Training recurrent networks using the extended Kalman filter
  • Ronald J. Williams
  • Computer Science
  • [Proceedings 1992] IJCNN International Joint Conference on Neural Networks
  • 1992
TLDR
The author describes some relationships between the extended Kalman filter (EKF) as applied to recurrent net learning and some simpler techniques that are more widely used, and gives rise to an algorithm essentially identical to the real-time recurrent learning (RTRL) algorithm. Expand
Training Multilayer Perceptrons with the Extende Kalman Algorithm
TLDR
It is shown that training multilayer perceptrons is an identification problem for a nonlinear dynamic system which can be solved using the Extended Kalman Algorithm. Expand
Kalman Filtering and Neural Networks
From the Publisher: Kalman filtering is a well-established topic in the field of control and signal processing and represents by far the most refined method for the design of neural networks. ThisExpand
The Variable Structure Filter
This paper presents a new strategy for estimation of state variables. The strategy may be applied to linear systems and is referred to as the Variable Structure Filter. The filter is considered forExpand
...
1
2
...