期刊
INDUSTRIAL & ENGINEERING CHEMISTRY RESEARCH
卷 58, 期 30, 页码 13533-13543出版社
AMER CHEMICAL SOC
DOI: 10.1021/acs.iecr.9b00900
关键词
-
资金
- Natural Sciences and Engineering Research Council of Canada (NSERC) through the Engage Grant
- McMaster Advanced Control Consortium (MACC)
This paper addresses the problem of synergizing first-principles models with data-driven models. This is achieved by building a hybrid model where the subspace model identification algorithm is used to create a model for the residuals (mismatch in the outputs generated by the first principles model and the plant output) rather than being used to create a dynamic model for the process outputs. A continuous stirred tank reactor (CSTR) setup is used to illustrate the proposed approach on a continuous system. To further evaluate its efficacy, the proposed methodology is applied on a batch poly(methyl methacrylate) (PMMA) polymerization reactor and the predictions are compared with that of first principles modeling and the data-driven approach alone. The paper demonstrates the improved modeling capability of the hybrid model over either of its components.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据