Reliability analysis of an HTGR SCRAM system including human interfaces. Final report
N.J. Becar, G.B. Curtis, David K. Wood
Abstract
N.J. Becar, G.B. Curtis, David K. Wood
Abstract
The SCRAM Protective System for Fort St. Vrain HTGR power plant has been reviewed in depth using the ''GO'' reliability analysis technique developed by Kaman Sciences Corporation. A GO logic model of the SCRAM Protective System was created by referencing an intermediate set of logic schematics drawn from numerous detailed system drawings and procedures. Computer analysis of the model provided success probabilities for the various events. Time lines were constructed from the operational and surveillance procedures to define significant points of human interface. The results of the GO analysis indicate that the probability of SCRAM failure is small. Signals from individual SCRAM parameters were evaluated to determine those parameters which might be most effectively improved, if desired, in future reactors. A sensitivity analysis was performed to define the major contributors to system unreliability. Accident scenarios, as submitted by RRD, AEC (ERDA), were evaluated to show the improvement in SCRAM reliability resulting from functional diversity, and to rank the probabilities of successful SCRAM when initiated by the postulated accidents. In the most sensitive areas, results are presented parametrically for ease of extrapolation to newer HTGR designs. (auth)
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The SCRAM Protective System for Fort St. Vrain HTGR power plant has been reviewed in depth using the ''GO'' reliability analysis technique developed by Kaman Sciences Corporation. A GO logic model of the SCRAM Protective System was created by referencing an intermediate set of logic schematics drawn from numerous detailed system drawings and procedures. Computer analysis of the model provided success probabilities for the various events. Time lines were constructed from the operational and surveillance procedures to define significant points of human interface. The results of the GO analysis indicate that the probability of SCRAM failure is small. Signals from individual SCRAM parameters were evaluated to determine those parameters which might be most effectively improved, if desired, in future reactors. A sensitivity analysis was performed to define the major contributors to system unreliability. Accident scenarios, as submitted by RRD, AEC (ERDA), were evaluated to show the improvement in SCRAM reliability resulting from functional diversity, and to rank the probabilities of successful SCRAM when initiated by the postulated accidents. In the most sensitive areas, results are presented parametrically for ease of extrapolation to newer HTGR designs. (auth)
Key concepts: Scram, Reliability (semiconductor), Reliability engineering, Computer science, Schematic, Nuclear engineering, Engineering, Power (physics)