Model Predictive Control (MPC)-Based Supervisory Control and Design of Off-Gas Recovery Plant with Periodic Disturbances from Parallel Batch Reactors

详细信息    查看全文

• 作者：Hyuncheol Ryu ; Jong Min Lee
• 刊名：Industrial & Engineering Chemistry Research
• 出版年：2016
• 出版时间：March 23, 2016
• 年：2016
• 卷：55
• 期：11
• 页码：3013-3025
• 全文大小：1044K
• ISSN：1520-5045

文摘

This paper proposes a plantwide control structure and a model predictive control (MPC) scheme for the effective separation of off-gas from a polysilicon plant with periodic disturbances in the feed. The target process studied in this work produces 10 000 MT of polysilicon per year with 28 chemical vapor deposition (CVD) reactors. CVD reaction kinetic models were constructed, and their parameters were identified using a genetic algorithm. Aspen Dynamics was used to model the off-gas recovery process and implement proportional–integral (PI) controllers for regulating the inventory levels. A centralized supervisory layer based on MPC, to minimize the fluctuation of the recovered concentration, was linked with regulatory controllers through a MATLAB/SIMULINK environment. Using a rigorous plantwide dynamic model, four cases were studied with different control schemes with and without buffer tanks. The proposed control strategy showed better performance, compared with fully decentralized classical proportional–integral–derivative (PID) schemes. It is thus expected that the developed control strategy will be able to achieve the design of batch-continuous processes with periodic disturbances. The benefits of using the multivariable MPC approach as a supervisory layer are shown and analyzed using Aspen Plus, Aspen Dynamics, and MATLAB/SIMULINK.