Abstract: The status of research and major achievements about the multivariate statistical methods in the fault diagnosis of chemical process are reviewed. Some primary multivariate statistical methods are introduced，including principal component analysis (PCA)，partial least square (PLS)，independent component analysis (ICA) and Fisher discriminant analysis (FDA). The application prospect, and existing problems are analyzed.

摘要： 综述了多元统计方法在化工过程故障诊断领域的理论进展和应用现状，介绍了一些主要的多元统计方法，其中包括主元分析、部分最小二乘、独立成分分析和Fisher判别分析，展望了多元统计方法在化工过程故障诊断领域的发展与应用前景。