Scholarlink

The impact of Performance Influence Factors (PIFs) on human error in nuclear power plants (NPPs) operations remains unclear. This study focuses on estimating PIF impact using a plant simulator experiment. Using the Macro-Physics Simulator, scenarios, including Total Loss of Feedwater (LOFW) and Steam Generator Tube Rupture (SGTR), were simulated under varied PIF states. The observational data was collected from procedural actions. The data was classified based on cognitive functions, and the failure probability for each cognitive function was calculated using data assimilation based on Bayesian estimation to calculate the Human Error Probability (HEP). Unlike previous approaches using final outcome data, this study highlights the advantages of using process data. The results show that to allow to estimate of PIF impacts on any cognitive function and the HEP calculated by using process data were more consistent with the observational than that by using conventional method.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Data Assimilation with Process Information from Plant Simulator Experiment for Estimation of PIF Impact

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Tomonori Sakamaki,
  • Satoshi Takeda,
  • Takanori Kitada

摘要

The impact of Performance Influence Factors (PIFs) on human error in nuclear power plants (NPPs) operations remains unclear. This study focuses on estimating PIF impact using a plant simulator experiment. Using the Macro-Physics Simulator, scenarios, including Total Loss of Feedwater (LOFW) and Steam Generator Tube Rupture (SGTR), were simulated under varied PIF states. The observational data was collected from procedural actions. The data was classified based on cognitive functions, and the failure probability for each cognitive function was calculated using data assimilation based on Bayesian estimation to calculate the Human Error Probability (HEP). Unlike previous approaches using final outcome data, this study highlights the advantages of using process data. The results show that to allow to estimate of PIF impacts on any cognitive function and the HEP calculated by using process data were more consistent with the observational than that by using conventional method.