Using neural networks to establish the structural controllability and observability

Authors:
R. Pereira;J. C. Quadrado
Affiliations:
DEEA, ISEL, Lisboa, Portugal;CAULT, Lisboa, Portugal
Venue:
NN'05 Proceedings of the 6th WSEAS international conference on Neural networks
Year:
2005

Citing 2
Cited 0

Fundamentals of neural networks: architectures, algorithms, and applications

Fundamentals of neural networks: architectures, algorithms, and applications
Controllability and observability of linear systems: somenoninvariant aspects

IEEE Transactions on Education

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper the usage of neural networks together with bondgraphs allowing the determination of the structural controllability and observability, is presented. The usage of bondgraphs allows a preliminary interpretation that enables the conclusion drawing about the structural characteristics, regardless of the numerical parameters of the system. The artificial neural networks are used to incorporate this knowledge again without the need for a formal mathematical model of the system. A mechanical suspension system is used as a case study.