基于改进SPEA2的原油短期调度问题研究  

Research on Short-term Crude Oil Schedule Based on Improved SPEA2

作　　者：王书娟 侯艳[1] 滕少华[1] 朱清华[1] WANG Shujuan;HOU Yan;TENG Shaohua;ZHU Qinghua(School of Computers,Guangdong University of Technology,Guangzhou 510006,China)

机构地区：[1]广东工业大学计算机学院,广东广州510006

出　　处：《工业工程》2023年第3期124-133,共10页Industrial Engineering Journal

基　　金：国家自然科学基金资助项目(61603100);广东省重点领域研发计划资助项目(2020B010166006)。

摘　　要：针对原油短期调度多目标优化问题,在分析已有多目标模型对原油调度过程中的供油罐个数、供油罐切换次数、原油在管道中的混合成本和供油罐罐底混合成本这4个目标优化的基础上,本文建立的模型增加了原油在管道转运过程中的能耗成本这一优化目标,使模型更吻合生产实际。在SPEA2算法中引入极值归档集,结合MOGWO算法指导极值归档集更新来提高算法的全局搜索能力;利用余弦相似度对归档集进行裁剪操作,以保证归档集中个体的多样性。将改进算法与多个具有代表性的进化多目标优化算法进行对比实验,结果表明,本文所提出算法在求解原油短期调度问题时性能较优。To solve the multi-objective optimization problem of short-term crude oil scheduling,on the basis of four objectives in existing optimization models including the number of charging tanks,the times of charging tank switching,the mixing cost of crude oil in pipelines and the mixing cost of charging tank bottoms in a crude oil scheduling process,the model established in this paper adds an optimization objective of the energy consumption cost of crude oil transferring through pipelines,which is more coincident to real production.The extreme value archive set is introduced into Strength Pareto Evolutionary Algorithm 2(SPEA2),while Multi-Objective Grey Wolf Optimizer(MOGWO)algorithm is combined to guide the update of the extreme value archive set to enhance the global search ability of the algorithm;the cosine similarity is adopted to prune the extreme value archive set to improve the diversity of individuals in the set.The modified algorithm is compared with several representative evolutionary multi-objective optimization algorithms,and the experimental results show that the proposed one has better performance in solving the short-term scheduling problem of crude oil.

关 键 词：原油调度 多目标优化 SPEA2算法 极值归档集 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]