Abstract: A chemical process has a large number of measured variables, which are usually driven by fewer latent variables. The latent variables are extracted by using independent component analysis (ICA) and monitored to detect the fault.According to the situation of alarm limit violations of independent components, the diagnosis of faults is reduced to a string matching problem. The proposed method is data driven and any training data or a process model is not required. It was evaluated by the application to the Tennessee Eastman challenge process and its effectiveness was demonstrated.

Key words: 独立成分分析, 字符串匹配, 最长公共子字符串, 统计过程监控, 故障诊断

摘要： 化工过程中存在大量测量变量，这些变量一般不是相互独立的，而是由少数必要的潜隐变量驱动，这些潜隐变量通过独立成分分析方法（ICA）抽取出来，可用于故障检测和监控.在基于ICA的故障检测和监控基础上提出一种故障诊断方法——字符串匹配方法，此种方法仅由数据驱动，既不需要训练数据也不需要建立过程模型，实施起来简单方便.通过在TE仿真模型上的应用，表明了此方法的有效性.

关键词: 独立成分分析, 字符串匹配, 最长公共子字符串, 统计过程监控, 故障诊断