基于改进蚁群算法的油品调和配方优化研究
首发时间:2011-10-31
摘要:油品调和是生产成品油的关键环节,调和配方的优劣决定了炼油厂最终利润的大小。针对油品调和复杂非线性约束配方优化问题,提出了一种改进的具有交叉算子的蚁群算法。该算法对蚂蚁的募集行为进行了改进,同时将遗传算法中的交叉算子引入蚁群算法中,在局部搜索中,应用了Hooke-Jeeves方法。仿真结果表明,新算法能够获得非常理想的配方模型、保证对质量指标的卡边要求、获得理想的调和利润。
关键词: 蚁群算法 非线性约束 约束处理 配方优化 油品调和
For information in English, please click here
Application of improved ant colony algorithm in gasoline blending
Abstract:Gasoline blending is a key step of producing refined oil, the blending recipe determined profits in refineries. The blending process involves many components mixing, and the character of mixing is nonlinear. Thus gasoline blending could be abstracted as a complex nonlinear optimization problem. It is difficult to obtain satisfying optimum solution by traditional methods. According to this problem, an improved ant colony algorithm, called Ant Colony Optimization algorithm with Crossover operator (ACOC), was presented. The proposed algorithm introduced crossover operator into the ant colony algorithm, and improved the global search ability. In the process of local searching, the ACOC applied the Hooke-Jeeves algorithm to improve the performance of optimization algorithms and improve the convergence speed. The simulation results present that the ideal blending recipes can be found, and get the maximum profit with a little margin of quality index.
Keywords: ACO non-linear constraints constraint dispose recipe optimization oil blending
基金:
论文图表:
引用
No.****
同行评议
共计0人参与
勘误表
基于改进蚁群算法的油品调和配方优化研究
评论
全部评论0/1000