核电厂观察指导数据趋势分析方法研究

Research and Application of Trend Analysis Method of Observation and Couch Data in Nuclear Power Plant

借鉴美国核动力运行研究院出版物INPO_07_007GP和国家标准GB/T4091—2001中推荐的趋势分析方法,笔者尝试采用常规控制图对核电厂观察指导数据开展趋势分析,以有效判定人员行为异常趋势、分析人员行为共性偏差、定位人员行为弱项。本文验证该方法的可行性并确定合适的控制图种类、分析指标、分析周期、分析对象和分析范围,选定适合观察指导数据的判异准则,阐明异常趋势分析方法。通过数据实验笔者发现:采用单值X图、以观察子项低于期望比率为分析指标、以月度为分析周期、以领域或部门为分析对象,对核电厂观察指导数据开展趋势分析是可行且合适的。本文提出的核电厂观察指导数据趋势分析方法已在核电厂年度人员行为分析报告中使用,并嵌入观察指导信息化系统。

Drawing on the trend analysis methods recommended in INPO_07_007GP published by Institute of Nuclear Power Operations and the national standard GB/T4091-2001, the author tried to use Shewhart Control Charts to conduct trend analysis on the observation and couch data of nuclear power plants to effectively determine the abnormal trend of personnel behavior and analyze personnel common deviations in behavior,positioning personnel behavior weaknesses. This paper verifies its feasibility and determines the appropriate control chart types, analysis indicators, cycles, objects and ranges, and select trend judgment criteria for the observation and couch data, and clarify abnormal trend analysis method. After data experiments, it is found that it is feasible and appropriate to use single-value X-chart to carry out trend analysis on the observation and couch data of nuclear power plants by taking the ratio of observation sub-items lower than expected as analysis index, the monthly as analysis period, and the field or department as analysis object. The trend analysis method of observation and couch data for nuclear power plants proposed in this paper has been used in the annual personnel behavior analysis report of nuclear power plants and has been embedded in the observation and couch information system.