Multi-working-condition multi-stage batch process monitoring method based on density peak clustering and instant learning

The invention discloses a multi-working-condition multi-stage batch process monitoring method based on density peak clustering and instant learning, and the method is used for solving the problem of multi-mode multi-stage batch process monitoring by combining algorithms such as density peak clustering, instant learning and the like. In order to solve the problems of inter-batch difference and non-Gaussian distribution in batch process data, firstly, working conditions and stages of the batch process data are classified and identified by using density peak clustering; due to the fact that quality variable tracks under the same working condition and stage are diversified, similar tracks are extracted through instant learning so as to obtain sub-data sets with the similar quality variable tracks. Therefore, for each quality variable track of each sub-stage in a certain sub-working condition, a sub-model is established so as to realize an accurate modeling and monitoring scheme. Finally, aBayesian fusion method i