An evaluation of the use of signal validation techniques as a defense against common-cause failures: Final report
The major objective of this project is to assess on-line signal validation techniques as a defense against common cause failures in instrumentation and control systems. The present study specifically addressed the BWR vessel level instrumentation, which has been a long-standing safety concern. The results of this study consist of (1) an assessment of current fault detection and isolation capabilities, (2) development of a generalized decision estimator for detection and isolation of common cause failures, and (3) quantification of safety impact of various signal validation design options. The major output of this study was the development of a generalized decision estimator with the ability to synthesize diverse information, identify plausible alternatives, and draw inferences from the available information or evidence and encoded knowledge. The basis of the proposed algorithm for the detection and identification of common cause failures is the use of functional diversity supplemented by additional plant condition information in situations where all direct redundant measurements may have failed or the specific evidence is not convincing. The inference process or decision estimator is based on a Bayesian formulation. The major conclusion of this study is that a properly implemented signal validation system using the proposed generalized decision estimator appears to provide successful protection against common cause failures in instrumentation and control systems. Specifically, it has been shown that implementation of signal validation using the generalized decision estimator could essentially eliminate a potentially important safety concern associated with the BWR vessel level instrumentation.
- Research Organization:
- Los Alamos Technical Associates, Inc., NM (USA); Electric Power Research Inst., Palo Alto, CA (USA)
- OSTI ID:
- 6670631
- Report Number(s):
- EPRI-NP-5081; ON: TI87920311
- Country of Publication:
- United States
- Language:
- English
Similar Records
Considerations regarding the Use of Computer Vision Machine Learning in Safety-Related or Risk-Significant Applications in Nuclear Power Plants
Risk Analysis of Various Design Architectures for High Safety-significant Safety-related Digital Instrumentation and Control Systems of Nuclear Power Plants during Accident Scenarios
Related Subjects
21 SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLANTS
BWR TYPE REACTORS
REACTOR MONITORING SYSTEMS
INFORMATION VALIDATION
EXPERT SYSTEMS
FAULT TREE ANALYSIS
LOSS OF FLOW
REACTOR CONTROL SYSTEMS
REACTOR INSTRUMENTATION
REACTOR SAFETY
RELIABILITY
SIGNALS
ACCIDENTS
CONTROL SYSTEMS
REACTOR ACCIDENTS
REACTORS
SAFETY
SYSTEM FAILURE ANALYSIS
SYSTEMS ANALYSIS
TESTING
VALIDATION
WATER COOLED REACTORS
WATER MODERATED REACTORS
220400* - Nuclear Reactor Technology- Control Systems
220900 - Nuclear Reactor Technology- Reactor Safety
210100 - Power Reactors
Nonbreeding
Light-Water Moderated
Boiling Water Cooled