Digital instrumentation and control failure events derivation and analysis for advanced boiling water reactor

摘要

[[abstract]]This research adopted Personal Computer Transient Analyzer- Advanced Boiling Water Reactor version (PCTran-ABWR) simulation computer code to analyze the software safety issue for a generic ABWR. A number of postulated instrumentation and control (I and C) system software failure events were derived to perform the dynamic analyses. The basis of event derivation includes the published classification for software anomalies, the digital I and C design data of ABWR, chapter 15 accident analysis of generic safety analysis report (SAR), and the reported nuclear power plant I and C software failure events. For the purpose of enhancing the ABWR major control systems simulation capability, this research incorporated MATLAB into PCTran-ABWR to improve the pressure control system, feedwater control system, recirculation control system, and automated power regulation control system. As a result, the software failure of these digital control systems can be properly simulated and analyzed. Moreover, via an internal tuning technique, the modified PCTran-ABWR can precisely reflect the characteristics of the power-core flow map. Hence, in addition to transient plots, the analysis results can then be demonstrated on the Power-Core Flow Map. The case study of this research includes (I) the software common mode failures analysis for the major digital control systems; and (2) postulated ABWR digital I and C software failure events derivation from the actual happening of non-ABWR digital I and C software failure events, which were reported to Licensee Event Report (LER) of US Nuclear Regulatory Commission (USNRC) or Incident Reporting System (1RS) of International Atomic Energy Agency (IAEA). These events were analyzed by PCTran-ABWR. Conflicts among plant status, computer status, and human cognitive status are successfully identified. The operator might not easily recognize the abnormal condition, because the computer status seems to progress normally. However, a well trained operator can become aware of the abnormal condition with the inconsistent physical parameters; and then can take early corrective actions to avoid the system hazard. This paper also discusses the advantage of Simulation-based method, which can investigate more in-depth dynamic behavior of digital I and C system than other approaches. Some unanticipated interactions can be observed by this method. © 2006 IEEE.