不确定性环境下面向生物柴油的厨余油脂回收网络鲁棒优化研究  

Robust Optimization of Waste Cooking Oil Recovery Network for Biodiesel Production under Uncertainties

作　　者：徐鑫 杨东[1] Xu Xin;Yang Dong(School of Business and Management,Donghua University,Shanghai 200051,China)

机构地区：[1]东华大学管理学院,上海200051

出　　处：《工业工程与管理》2024年第3期137-148,共12页Industrial Engineering and Management

基　　金：国家自然科学基金资助项目(71971053,71832001);教育部人文社会科学研究项目(18YJA630129)。

摘　　要：作为化石类柴油的良好替代品,生物柴油的全球产量逐年增长。而中国由于对粮食和食用油需求巨大,无法形成以生物质(大豆、棕榈和蓖麻等)为原料生产第一代生物柴油产业链。考虑到国内每年产生大量的废弃厨余油脂,本文构建了以总成本最小化为目标的厨余油脂回收网络优化模型,从而实现废弃油脂的回收利用以及提供合格的生物柴油。针对厨余油脂回收问题中回收量和生物柴油需求的不确定性,采用了预算鲁棒优化方法来刻画不确定参数,进一步考虑了0号柴油价格的不确定性,并引入一个与0号柴油价格相挂钩的政府补贴分段函数。通过线性化处理和对偶理论,将非线性混合整数规划模型转化为易处理的线性鲁棒等价模型,从而采用CPLEX进行求解。通过不同算例进行仿真验证,证明了鲁棒优化模型在不确定环境下更为稳健,同时以实际案例验证了所提鲁棒优化模型的有效性。As a good substitute for fossil diesel,the global production amount of biodiesel is increasing year by year.However,due to huge demands for grain and edible oil,the first-generation biodiesel could not be produced from biomass in China.Considering that a large amount of waste cooking oil was produced in China every year,an optimization model of cooking oil recovery network was constructed aiming at minimizing the total cost,so as to realize the recycling of waste oil and provide qualified biodiesel.In view of the uncertainties in daily recovery volumes and demands for biodiesel,a budget-based robust optimization model with government subsidies was formulated.By using linearization and dual transformation,the nonlinear mixed integer programming model was transformed into a tractable linear robust equivalent model,which could be solved by CPLEX.The effectiveness of the robust optimization model was demonstrated with an actual case.Simulation experiments demonstrate that the robust optimization model is effective when uncertainties are dealt with.

关 键 词：生物柴油回收网络 厨余油脂 鲁棒优化 

分 类 号：F252[经济管理—国民经济]