摘要
采用混合建模技术,通过集成人工神经网络模型与单元过程机理模型建立了超临界流体萃取过程的混合模型,较好地解决了超临界过程的模拟计算、经济评价、设计和运行优化问题.通过软件集成的方法实现了商业模拟软件Aspenplus与ANN的集成,大大降低了模型的实现难度和实施时间.
This paper describes the modeling of dynamics of SFE process using a hybrid model that combines both ANN and the rigorous process unit mechanism model. The accuracy of hybrid model is satisfying. The method is found to be practical to solve the problems of economical analyses, design and operation optimization of SFE process. The difficulty and time used for implementation of the hybrid model can be greatly reduced through the integration of commercial process simulation software and self-developed module by software integration technique.
出处
《河南大学学报(自然科学版)》
CAS
北大核心
2006年第3期42-46,共5页
Journal of Henan University:Natural Science
基金
河南省自然科学基金资助项目(0511020200)
关键词
过程模拟
人工神经网络
超临界流体萃取
混合模型
process simulation
artificial neural network
supercritical fluid extraction
hybrid model