基于改进粒子群算法的柴油调合配方优化

为了解决扬子石化炼油厂柴油调合质量过剩的问题,介绍了炼油厂柴油调合的组分资源及调合现状,采用改进粒子群算法,在多智能体仿真平台——Repast上建模仿真,对炼油厂现有组分油资源进行合理的调配,相对于传统的现场人员经验调合,线性规划方法调合等方式有效地解决了现有柴油调合配方调出的柴油硫含量、十六烷值、闪点等质量指标过剩的问题,对调合人员起到了一定的辅导作用,有效地降低了现有柴油调合的经济成本。

Diesel Blending Formulation Optimization Based on Modified Particle Swarm Algorithm

In order to solve the problem of diesel blending quality surplus of the Yangzi petrochemical refinery,the diesel blending component resources and present situation of the refinery were introduced.Modeling simulation was made in multi-agent simulation platform—Repast by using improved particle swarm algorithm and the existing oil components resource of the refinery was allocated reasonably.Compared with traditional experience of field technicians blending,the problems of high sulfur content and others quality standards such as cetane number,flash point by existing formula were solved by linear programming blending methods.The results could also be helpful for blending field technicians to develop new formula and the existing diesel blending economic cost were reduce.