Uncertain process fault monitoring method based on local and global interval embedding

The invention relates to the field of industrial process fault monitoring, in particular to an uncertain process fault monitoring method based on local and global interval embedding. Comprising the following steps: S1, acquiring an inaccurate single-valued data set of equipment under a normal condition; s2, establishing a local interval embedding model and a global interval embedding model; s3, calculating statistics associated with the normal interval value data set according to the interval principal component; s4, collecting a new data sample; s5, calculating the statistical magnitude of new interval value data by utilizing the interval principal components determined by the local and global interval embedding models; s6, monitoring whether the newly obtained four statistical magnitudes exceed the control limit or not, if the four statistical magnitudes exceed the control limit, determining that the system breaks down, and executing the step S7, otherwise, returning to the step S4 and monitoring the next s