Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Sensor failure detection and estimation

Journal Article · · Nucl. Saf.; (United States)
OSTI ID:5042018

In a large dynamic system, such as a commercial nuclear plant, process signals are used for plant control; safety systems; monitoring devices, such as the safety parameter display system and the Technical Support Center; postaccident diagnostics; event prediction; and transient recovery. It is necessary to address the problem of signal validation and sensor verification to improve the reliability of such systems. Several sensor failure detection and estimation methods are outlined, including analytical redundancy, generalized likelihood ratio approach, parity-space representation, data-driven representation, and verification by using correlation from independent instruments. Two of the applications are related to the loss-of-fluid test reactor, and this illustrates the flexibility of implementing failure detection methods in a well-instrumented facility. In this context, the aspects of sensor location, measurement models, and reactor subsystem description are discussed.

Research Organization:
Univ. of Tennessee, Knoxville
OSTI ID:
5042018
Journal Information:
Nucl. Saf.; (United States), Journal Name: Nucl. Saf.; (United States) Vol. 26:1; ISSN NUSAA
Country of Publication:
United States
Language:
English