MODEL-FREE ONLINE ROLLING OPTIMIZATION METHOD FOR BATCH PROCESS ON BASIS OF TIME PERIOD VARIABLE DECOMPOSITION

摘要

A model-free online rolling optimization method for a batch process on the basis of time period variable decomposition. By acquiring operation data of a variable closely related to product quality on the basis of performing time-domain variable division on a process by utilizing a data driving method, optimization action quantity of each subset is integrated and a global optimization strategy is formed, on the basis of which, an online rolling error correction optimization strategy is performed. The method forms the online optimization strategy on the basis of the operation data of the batch process completely without requiring prior knowledge or a model of a process mechanism. Also, an optimized operation path line can have better adaptability by using the online rolling correction strategy, and thus the anti-interference requirement of actual industrial production can be better met.