Title :
Hyperfast algorithms for model-based diagnosis
Author :
Larsson, Jan Eric
Author_Institution :
Knowledge Syst. Lab., Stanford Univ., Palo Alto, CA, USA
Abstract :
This paper describes small, efficient implementations of three diagnostic methods for use with industrial processes. They are measurement validation, alarm analysis, and fault diagnosis. All three methods use multilevel flow models, to describe the target process. They have been implemented in G2 and C, and provide fast, hard real-time diagnosis
Keywords :
fault location; manufacturing data processing; monitoring; process control; real-time systems; C toolbox; alarm analysis; fault diagnosis; hyperfast algorithms; industrial process; measurement validation; model based diagnosis; multilevel flow models; real time diagnosis; target process; Algorithm design and analysis; Diagnostic expert systems; Fault diagnosis; Fluid flow measurement; Formal languages; Gold; Knowledge based systems; Laboratories; Magnetic force microscopy; Real time systems;
Conference_Titel :
Computer-Aided Control System Design, 1994. Proceedings., IEEE/IFAC Joint Symposium on
Conference_Location :
Tucson, AZ
Print_ISBN :
0-7803-1800-5
DOI :
10.1109/CACSD.1994.288881
