内蒙古科技大学硕士学位论文 -1-摘 要摘 要 铁精粉等高炉原料的采购成本在生铁成本中的比例可达 70％左右，因此，确立最佳的原料结构是降低产品（生铁）成本的有效手段另外，随着我国钢铁行业的高速发展，铁精粉的需求量和进口量大幅增加，铁精粉的价格不断上升，高炉含铁原料的采购决策模型也逐步完善和被广泛使用 本文对 B 钢铁公司物料结构进行优化分析，建立了面向产品成本的高炉物料结构优化模型数学模型分为6 个子模型，其中工艺模型有烧结机优化模型、竖炉优化模型、高炉优化模型，物料配置模型有铁精粉优化配置模型，烧结矿优化配置模型和球团矿优化配置模型，模型之间交互的数据流为原料或产品的物料流量、物料性能、价格及原料性能的影子价格工艺模型采用模糊规划方法建模，物料配置模型采用网络流优化方法建模，模型采用 MATLAB 软件编写程序计算结果以含铁原料使用数量的边际贡献的方式提交，以支持决策者调节原料结构 本论文的特色如下：首先，本论文模型将以产品成本最低作为原料结构优化的目标，而不以采购成本最低、烧结矿成本最低球团矿成本最低等作为目标；其次，模型运用模糊规划，从基准期的工艺参数和原料性能中校准出效率参数，从而精确描述出优化期内数量（产品产量或原料用量）、性能（产品性能和原料性能）和成本三者之间的数值关系；最后，计算含铁物料用量的边际贡献。
本模型不计算最优的原料结构，而是计算原料用量的边际贡献（原料用量发生单位变化时，铁水成本的变化值），这种提交方式允许决策者综合考虑社会人文等模型无法考虑的因素，另外，由于在计算边际贡献时，模型仅仅在基准值的一个邻域内求解，消解了插值和外推带来的误差，因而计算结果更为准确 本文得出以下结论：第一，模糊线性规划方法可有效地校准高炉过程的效率参数，较回归分析法简便，且计算精度高；第二，使用输运矩阵描述烧结矿、球团矿和铁精粉的运输网络，可将非线性规划转化为线性规划，简化了求解过程 关键词：冶金企业，生产计划，原料结构，模糊规划关键词：冶金企业，生产计划，原料结构，模糊规划 内蒙古科技大学硕士学位论文 -2-Abstract The raw material mixture structure is very important for reducing manufacturing cost and attaining maximum profits for the Iron and Steel enterprises. The proportion of the cost of raw materials in pig iron reaches nearly 70 percent. So it is an effective ways to decide a reliable raw materials mix reducing the cost of product. Based on this demand, a model of raw material mixture optimization for the cost of the pig iron was established to calculate the marginal contribution of all kinds of raw materials (focusing on the changes of iron raw materials). And it provides decision support for the enterprises to make a reasonable raw material mixture. A integrated mathematical model was established in accordance with the process of metallurgical production and it was composed of the process models and the raw materials distribution models. The process models include the sintering process model, the pelleting process model and the iron making process model. The raw materials distribute models include the concentrate ores distribution model, sinter and pellets distribution model. The dates transferred among the models were the weight of raw materials or product, compositions, the prices and the shadow price of compositions of raw materials. The special features of the production planning model are as follows: (1) The goal of the model was the cost of product, rather than that of the raw material or work in process, such as the cost of sinter, or the cost of pellets .(2) By making use of the fuzzy program, many important parameters were calibrated from the fuzzy process model and the compositions of raw materials in the period of optimization. Then the relations among the quantity (producing output or consumption of raw materials), the compositions (product output or consumption of raw materials) and the cost or profit were accurately formulated. (3) The marginal contribution of the raw materials was presented to the decision maker. This model doesn’t calculate the optimal value, but the marginal contribution of the raw material. (What we got is the change of the product profits when the amount of raw materials has a unit change). Such a result is more suited to the needs of decision-making of the current raw materials mix. In addition, the model was solved in the range of reference value of the period of optimization, removing the error. So the result was more accurate. Following conclusions can be drawn from the above research: (1) the method of fuzzy linear programming is able to calibrate the efficiency parameters of the process of BF. It is more easy and precise than the method of linear regression analysis. (2) By using distribution matrix to describe the distribution net of sinter, pellet and concentrate, can change the nonlinear programming into linear programming. It provides a reliable basis for calculating the marginal contribution of materials mixture. Key Words：：Iron and Steel Enterprise, Production Plan, Raw Material Mix, Fuzzy Program 内蒙古科技大学硕士学位论文 -3-图表清单图表清单 表 1.1 成本动因...................................................................................................................10 表 1.2 国内各钢铁企业优化原料结构和产品结构的实践..............................................17 表 2.1 工艺模型的数据结构...............................................................................................22 表 2.2 工艺模块内部主要状态数据...................................................................................27 表 2.3 烧结过程优化模型的映射逻辑..............................................................................27 表 2.4 烧结过程校准模块的映射逻辑..............................................................................30 表 2.5 球团优化模型的映射逻辑.......................................................................................32 表 2.6 球团校准模型映射逻辑..........................................................................................35 表 2.7 高炉冶炼的优化方程...............................................................................................39 表 2.8 影子价格的计算........................................................................................。