Title :
A Hopfield-Based Neuro-Diagnostic System
Author :
Chu, S.R. ; Shoureshi, Rahmat ; Healey, A.J.
Author_Institution :
School of Mechanical Engineering, Purdue University, W. Lafayette, IN 47907
Abstract :
A high potential application area for neural networks in the dynamic systems is the area of failure detection and identification. An innovation based failure diagnostic is considered in this paper. This scheme requires a fast on-line system identification technique. Formulation and development of a recurrent Hopfield network for system identification is presented. The general case of a combined parameter idenfification and state observation is considered.
Keywords :
Circuits; Equations; Intelligent networks; Mechanical engineering; Neural networks; Neurons; Real time systems; Recurrent neural networks; System identification; Technological innovation;
Conference_Titel :
American Control Conference, 1992
Conference_Location :
Chicago, IL, USA
Print_ISBN :
0-7803-0210-9
