基于群智能的间歇过程分阶段融合建模与协同优化

项目名称： 基于群智能的间歇过程分阶段融合建模与协同优化

项目编号： No.21206053

项目类型： 青年科学基金项目

立项/批准年度： 2013

项目学科： 化学工程及工业化学

项目作者： 熊伟丽

作者单位： 江南大学

项目金额： 26万元

中文摘要： 间歇过程由于其反应机理复杂、高度非线性和模型的不确定性等特点，相对于其它模型一定的工业过程，其自动化程度还很低，已经成为当今国际过程控制领域研究的热点。首先，本项目针对间歇过程"有限时间"表现出来的批次轨迹不同步和存在离群点而导致的实际间歇过程机理模型容易老化失效的问题，在充分研究各种基于数据驱动的建模方法基础上，综合考虑间歇过程反应时段建模数据特征以及关键变量的变化规律，判断运行阶段，同步处理过程变量，建立相应的子阶段模型；然后针对间歇过程"无现实稳态性"表现出的动态非线性本质特性，提出基于群智能优化的多模型融合建模策略，以最终建立适合间歇过程特点的软测量模型；最后基于所建立的软测量模型，运用群智能优化算法实现间歇过程的多目标协同优化控制，并设计最优控制器。本项目属于应用基础研究，研究成果在生物发酵、精细化工等高附加值的间歇过程工业领域有广泛的应用前景。

中文关键词： 建模；多模型；融合；优化；间歇过程

英文摘要： The degree of automation for batch process is relatively low compared to other model-oriented industrial processes, due to its complex mechanism of the reaction as well as the high nonlinearity and the uncertainty of its model. Many researchers have been devoted to developing control strategies to improve the performance of batch process. A batch process has a certain period of operation time, and there are a number of batches in a typical operation. Consequently, the mechanism models always become invalid for the batch process. As such, the data-driven modeling methods are investigated in detail in this project firstly. Combined with the data characteristics and the changing rules of the main variables, the methods can determine the operation stages and construct the corresponding multi-stage model. Then based on swarm intelligence optimization, a novel modeling strategy is proposed to construct the final fusion model. This software sensor model can reflect the nonlinearity better for the batch process. Finally, by using this model, the multi-objective and collaborative optimization control is realized and the optimal controller is designed for the batch process. It can be expected that the innovative research of this project on the software sensor modeling and optimization control will be of great significance

英文关键词： modeling；multiple model；fution；optimization；batch process