文献类型：专利

英文题名：Pareto optimization criterion-based multi-target non-linear continuous tank type diesel blending method, comprises e.g. establishing a multi-target non-linear mathematic model, performing algorithm-based evaluation, and blending diesel

作者：CAO C;GAO M;GU X;LI X;YU T;CHEN M;HAN M;KANG Y

机构：[1]UNIV EAST CHINA SCI & TECHNOLOGY

申请号：CN103497789-A

申请日：2013-10-11

公开日：2014-01-08

语种：英文

收录：DERWENT

摘要：NOVELTY - Pareto optimization criterion-based multi-target non-linear continuous tank type diesel blending method, comprises e.g. (i) establishing a multi-target non-linear mathematic model based on e.g. a related linear and non-linear relationship between the real-time flow and the flow of each component, (ii) inputting the real-time flow data of each lateral line and the real-time quality data of each component respectively, and performing evolutionary algorithm-based evaluation by adopting Pareto optimization criterion, (iii) blending the finished diesel product according to the real-time flow data. USE - The method is useful for Pareto optimization criterion-based multi-target non-linear continuous tank type diesel blending (claimed). DETAILED DESCRIPTION - Pareto optimization criterion-based multi-target non-linear continuous tank type diesel blending method, comprises (i) establishing a multi-target non-linear mathematic model based on a related linear and non-linear relationship between the real-time flow and the flow of each component which forms the finished diesel product, and a linear and nonlinear relationship between each component and the real-time quality attribute of the finished diesel product, (ii) inputting the oil refining rate of crude oil, the real-time flow data of each lateral line and the real-time quality data of each component respectively into the multi-target non-linear mathematic model established in the step (i), and performing evolutionary algorithm-based evaluation by adopting the Pareto optimization criterion to obtain the real-time flow data of each component which forms the finished diesel product, (iii) blending the finished diesel product according to the real-time flow data obtained in the step (ii), where the target of the multi-target non-linear mathematic model is a maximum final diesel product (target A) total yield, maximum total sales revenue of diesel product (target B), maximum total yield of final blended diesel product with the highest price in coordinated production (target C), minimum total quantity of remaining intermediate product produced by a production device capable of producing diesel components (target D) and/or minimal sulfur content in final diesel product (target E).