引用本文: | 李荣雨,荣 冈.基于故障映射向量和结构化残差的主元分析(PCA)故障隔离[J].控制理论与应用,2008,25(6):1099~1104.[点击复制] |
LI Rong-yu,RONG Gang.Principal component analysis(PCA) of fault isolation based on fault mapping vector and structured residual[J].Control Theory and Technology,2008,25(6):1099~1104.[点击复制] |
|
基于故障映射向量和结构化残差的主元分析(PCA)故障隔离 |
Principal component analysis(PCA) of fault isolation based on fault mapping vector and structured residual |
摘要点击 1684 全文点击 1475 投稿时间:2007-02-06 修订日期:2008-04-17 |
查看全文 查看/发表评论 下载PDF阅读器 |
DOI编号 10.7641/j.issn.1000-8152.2008.6.025 |
2008,25(6):1099-1104 |
中文关键词 主元分析 故障隔离 过程故障 结构化残差 映射向量 |
英文关键词 principal component analysis fault isolation process fault structured residual mapping vector |
基金项目 国家自然科学基金资助项目(60421002); 国家高科技发展计划项目(2007AA04Z191). |
|
中文摘要 |
在基于主元分析(PCA)的多变量统计过程监控中, 现有方法很难直观有效地完全实现故障的隔离与诊断. 本文通过分析各类故障的数学模型, 提出一种基于结构化残差和故障映射向量的隔离方法, 并推导出PCA模型下过程故障映射向量方向的提取算法, 进而实现了传感器/执行器故障和过程故障的故障隔离, 在CSTR仿真上的研究进一步验证了该法的有效性. |
英文摘要 |
Few approaches in multivariate statistical process-monitoring based on principal component analysis(PCA) can implement fault isolation completely and effectively. After analyzing the mathematical models of all kinds of faults, this paper proposes an isolation method based on structured residuals, which can isolate sensor/actuator faults and process faults. Furthermore, the algorithms for obtaining the fault mapping-vector direction are also deduced. Simulation results show the effectiveness of this method. |