基于指标异常诊断的核电化学知识库的研究及创建

Research & Construction of Chemistry Knowledgebase in Nuclear Power Plant Based on Abnormal Index

通过对某压水堆核电站二回路CEX–λ+、CEX–O2、APG–Na+及APG–p H四大关键化学指标进行层次分析和特点挖掘,根据指标异常状态建立核电化学指标知识库;实现化学指标异常知识的存储与调用,增加核电运行的数字化程度与远程监控能力;该知识库基于三维矩阵数学模型来设计和创建,通过结合模拟诊断的实施,对模拟结果进行分析得出知识库功能实现全面、运行稳定;该知识库不仅可以实现化学指标的异常诊断和智能分析,还可以包含或随时扩充大量知识信息项目,实现高效的调用和诊断分析。

By study and analysis of four key chemical index（ CEX – λ＋,CEX – O2,APG – Na＋and APG– p H） of PWR nuclear power plant,chemistry index knowledgebase was built for nuclear power plants. Its functions are to storage and recall the abnormalities of chemical indices knowledge,Increase the level of digital and capabilities of remote monitoring in operation of nuclear power plant. The knowledgebase was based on mathematical model of three dimension matrix. Through diagnosis flow of one chemistry index and the analysis of simulation results,it is concluded that the knowledgebase is comprehensive and can operate stablly. The knowledgebase was proved to be useful for exception analysis and diagnosis,and knowledge item information can be called and expanded easily.