【作者】 唐玉龙； 廖敦明； 陈涛；

【Author】 TANG Yu-long;LIAO Dun-ming;CHEN Tao;State Key Lab.of Material Processing and Die & Mould Technology, Huazhong University of Science & Technology;Wuhan Chenxi Yunfeng Technology Co., Ltd.;

【机构】 华中科技大学材料成形与模具技术国家重点实验室； 武汉晨曦芸峰科技有限公司；

【摘要】 在铸造生产中,物料的配比不仅要满足最终钢液的质量达标,还要考虑降低成本。现有优化方法大多是局部最优的算法,并且对搜索的初始值非常敏感。本文以元素平衡和成本最低为目标,采用遗传算法实现炉料配比全局最优模型的构建,分别利用SGA算法和DE算法进行最优配比参数的优化求解。结果表明,遗传算法能够正确求解最优配比方案,实现降本增效的目的。相较于SGA算法,DE算法的收敛特性更佳,具有很好的推广应用价值。更多还原

【Abstract】 In the steelmaking process of casting production, the ratio of raw materials needs not only meet the quality of the end-point molten steel, but also minimum the cost. Most of the existing optimization algorithms are locally optimal, which are very sensitive to the initial value. In this paper, two subsets of the genetic algorithms – Simple Genetic Algorithms(SGA) and Differential Evolution Algorithm(DE) – are used to optimize the ratio of raw material, focusing on the element balance and cost minimization. The results show that while both of the genetic algorithms solve the optimal ratio correctly, the DE algorithm convergences much faster, which has a higher value of practical application.更多还原