Sensor Validation: A Method to Enhance the Quality of the Man/Machine Interface in Nuclear Power Stations

Abstract

A major thrust in the current efforts to improve the safety of nuclear plants is to assure that the plant operator has accurate and timely information on which to base his actions. An important system capability, which is necessary to achieve this goal, is a reliable means to determine the validity of sensor data. Sensor signal validation is considered to be a prerequisite for successful use of disturbance analysis and surveillance systems (DASS), which are currently under study by EPRI and DOE. At the heart of the DASS concept of providing information to aid nuclear plant operating crews in making correct decisions, is the creation of a single, validated, data base; upon which DASS functional modules operate. Examples of DASS functions are monitoring of critical safety functions (reactivity control, core heat removal, primary coolant system integrity, radioactivity and containment integrity), disturbance detection, system configuration monitoring, verification of control actuation and technical specification margins monitoring. Signal validation is essential for the flow of correct information from the data base to the plant operating crew. The benefits from use of a DASS would quickly be negated by the possibility of situations where questionable or erroneous information could be generated by DASS. Furthermore, the validation methodology is likely to be applicable not only to signals, but also to reactor plant system and sub-system functional operation, i.e., diagnostic analysis of reactor control loops and sub-loops.