Run-Ahead Predictive Simulator Implementation in Nuclear Power Plant Control Rooms

摘要

The Run-Ahead Predictive Simulation (RAPS) is a new methodology (illustrated in Fig. 1) being designed by researchers at Oregon State University to assist the senior members of a nuclear power plant operating crew (i.e., the plant's shift or unit supervisor and the shift technical advisor). This tool intends to generate a set of scenarios to predict what could happen in the near future (with associated probabilities) by continuously performing a faster-than-real-time probabilistic risk assessment including outcome and consequence analyses. The RAPS utilizes current plant data to generate an input for a parallel-running systems simulation code, such as RELAP (RELAP5 Code Development Team, 2001), or other system simulators. As opposed to the slower running dynamic probabilistic risk analysis codes (Hakobyan et al., 2008), these parallel "potential futures" calculations are determined utilizing a small number of streamlined, risk-informed algorithms that repeatedly initiate, identify, analyze and disposition possible near future scenarios in a probabilistic manner as plant conditions continuously evolve. Results are presented to senior operating staff who can make use of these risk informed projections of future events to help guide decisions concerning the direction of plant operations.