| | |
Summary: TDA Progress Report 42-132 February 15, 1998
An Intelligent Fault Detection and Isolation
Architecture For Antenna Arrays
K. Rahnamai,1
P. Arabshahi,2
T.-Y. Yan,2
T. Pham,2
and S. G. Finley2
This article describes a general architecture for fault modeling, diagnosis, and
isolation of the DSN antenna array based on computationally intelligent techniques
(neural networks and fuzzy logic). It encompasses a suite of intelligent test and di-
agnosis algorithms in software. By continuously monitoring the health of the highly
complex and nonlinear array observables, the automated diagnosis software will be
able to identify and isolate the most likely causes of system failure in cases of faulty
operation. Furthermore, it will be able to recommend a series of corresponding
corrective actions and effectively act as an automated real-time and interactive sys-
tem supervisor. In so doing, it will enhance the array capability by reducing the
operational workload, increasing science information availability, reducing the over-
all cost of operation by reducing system downtimes, improving risk management,
and making mission planning much more reliable. Operation of this architecture
|
