文献类型：专利

英文题名：Industrial ethylene cracking furnace based proxy model full cycle optimizing method, involves obtaining hydrocarbon feed flow rate, constructing continuous non-linear dynamic optimization model, and obtaining optimization setting input

作者：DU W;LI J;QIAN F;YE Z;WANG Z

机构：[1]UNIV EAST CHINA SCI & TECHNOLOGY

申请号：CN103605325-A

申请日：2013-09-16

公开日：2014-02-26

语种：英文

收录：DERWENT

摘要：NOVELTY - The method involves constructing a mechanism model. An input variable and an output variable are selected. Cracking depth of propylene/ethylene ratio is obtained. A neural network agent model structure is determined. A multi-input neural network model is selected to form a neural network proxy model. Key product yield is obtained. Hydrocarbon feed flow rate is obtained. A continuous non-linear dynamic optimization model is constructed. An optimization setting input is obtained. An input variable range is set. Coking thickness of a furnace tube is obtained. USE - Industrial ethylene cracking furnace based proxy model full cycle optimizing method. ADVANTAGE - The method enables simplifying deriving design and ensuring simple operation steps, better engineering applicability and wide adaptability. DESCRIPTION OF DRAWING(S) - The drawing shows a schematic view illustrating an industrial ethylene cracking furnace based proxy model full cycle optimizing method. '(Drawing includes non-English language text)'