地下铁矿工业品位优化决策研究
|
摘要
工业品位本质上是确定给定开采块段工业储量的指标,是一个与矿体地质属性密切关联的指标。实际应用中,工业品位是用于区分有工业开采价值和没有工业开采价值矿石的临界品位,而矿石工业开采价值的有、无主要取决于开采过程能否带来经济价值。因此,工业品位具有明显的经济含义。事实上,工业品位的变化必然影响采出品位和采出矿量,进而影响矿石贫化率、回采率和选矿金属回收率,这些指标都会影响原矿产量或精矿产量,而精矿产量是矿山企业获取经济利润的主要手段。因此,工业品位对矿山企业经济利润有一定影响。
在开采地质品位既定的圈定矿体下,假如矿山企业生产技术水平不变,则精矿产量的变动主要受工业品位变动影响。精矿价格是矿山企业利润的重要决定因素,精矿产量决策受到精矿价格因素制约。精矿价格、精矿产量与工业品位三者之间呈现复杂的动态关系,它们一起共同影响矿山企业的经济效益。精矿价格是不断变化的,圈定矿体的地质品位也会呈现一定的随机波动,这些因素的动态变化必然影响到精矿产量的决策,也需要与之相适应的工业品位优化决策,以实现矿山企业利润最大化。综上所述,工业品位是与矿山地质、企业生产技术以及经营业绩密切相关的技术指标,它与矿山企业经济技术及生产要素构成一个复杂的、高度非线性的系统。传统技术与方法对复杂的、高度非线性系统难以有效模拟其动态行为,因此,工业品位优化需要采用非传统技术与方法,包括智能算法或动态规划法等对其建模与优化。
从中长期来看,工业品位还受到生产技术、经营环境的影响。采选技术的进步,为矿山生产过程降低原矿品位提供了条件,即降低工业品位提供了条件。矿山企业生产成本也影响工业品位,成本高,则需要降低矿石采选过程中的直接成本,这就需要尽可能利用高品位矿石资源,即工业品位相应提高。反之,工业品位应相应降低。工业品位优化是基于矿石生产经营中获取合理利润或最大利润基本假设下进行的,铁精矿价格和矿石生产成本是影响矿山企业经济利润的最重要指标,工业品位对矿山企业精矿价格和管理成本有着一定的敏感性。工业品位的优化还要考虑生产技术、经营环境的综合影响。
工业品位具有明显的技术经济意义,因此该品位的核算应该采用一定的经济学原理,建立核算模型。该品位是表外矿和表内矿的临界品位,如果利用经济学“边际成本=边际收益”,则表外矿利用的边际品位就是工业品位,由此提出的模型主要基于静态时点边际成本与边际收益的比较,因此称为静态模型；如果利用“经济人追求经济利润最大化”的经济学原理,则优化模型是基于矿山企业一段时期内矿石生产的利润净现值最大化模型,此时,矿石生产受到诸多与工业品位有联系的生产技术指标约束。由于该模型参数的动态性和求解过程的动态性,可称为动态优化模型。该方法应用的重要前提是依据样本数据地质品位分布特征,确定工业品位与矿量的对应关系,并明确各种相关生产约束。
金山店铁矿是典型的地下矿山,本文根据地下矿山工业品位优化理论对该矿山进行案例研究。金山店铁矿的矿山地质属性和工业品位管理相关的生产管理状况以及矿山主要经济技术参数,包括它的成本结构和成本特征是工业品位优化所必需的基础资料,为此,本文详细分析了该企业的成本管理结构,并根据企业成本2009-2011年成本资料,分析计算了单位可变成本与固定成本及其年度增长率。铁精矿价格是企业经营环境的重要指标,在基于武钢集团价格形成机制基础上,分析、比较了武钢集团国内矿与进口矿的价格走势特征。这些基础分析,为工业品位优化打下了良好基础。
应用金山店铁矿实际生产经营数据和矿山企业行业相关数据进行了静态工业品位计算,计算结果包括：基本工业品位、经济品位和利润品位。其中,利润品位的概念是本文结合计算内涵提出。根据国家最新固体资源储量分类标准,经济品位是确保矿山企业回收投资的品味,与补偿投资意义下的工业品味内涵一致,因此,“经济品位”也是一种工业品位。同时,矿山企业的利润追求是矿山企业可持续发展的基础,因此,“利润品位”是更高层次意义下的工业品位。计算结果分析表明,静态工业品位优化结果合理,表明该方法具有一定的实际应用价值。但该方法具有一些明显的不足,例如优化结果受铁精矿价格波动影响较大,还受到相关投资贷款指标、预期利润指标和投资回收期指标难以准确核算诸多因素的影响,存在着高价开采高品位矿的倾向,没有考虑资源利用率。尽管如此,仍不可否认它的实际指导意义,因为该方法经济意义明显,其计算结果直观、对应的附加品位符合投入产出思维方式,且实际结果基本符合行业规范和生产实际。根据利润净现值最大化原理,可以建立动态模型。该模型充分考虑了资金的时间价值和生产经营的动态性,与静态模型相比具有较大的优势,这个结论已为本文中的比较研究所证实。因此,本文工业品位优化决策系统以动态优化模型为内核构建。
总之,地下采矿工业品位的优化是一个系统决策工程,该系统涉及包括地质、生产技术及经济效益等诸多因素。本文以地下矿山铁矿石开采的工业品位优化为研究对象,对品位优化的基本理论、方法和应用进行了全面、系统地归纳、总结,融合或集成了矿山指标优化的建模以及智能算法,在此基础上,以金山店铁矿地下开采为案例,全面分析该矿山企业生产经营品位管理相关资料,建立了优化模型,优化了工业品位,并以优化模型为中心,建立了工业品位优化决策支持系统。主要成果如下：
(1)提出了地下矿山工业品位优化的静态模型。根据国内外对露天矿山低品位矿石经济品位的研究模型,结合地下铁矿生产技术条件,应用经济学“边际收益=边际成本”基本定价原理,给出了地下矿山企业的静态工业品位模型,并分别计算了维持企业正常生产经营的基本工业品位、除正常经营外,还可补偿投资贷款的经济工业品位、以及既保证生产,又补偿投资贷款和合理利润预期的利润工业品位。其中,“利润品位”一词是笔者根据该品位对应的经济意义提出的。计算结果符合金山店铁矿目前生产经营实际。
(2)提出了地下矿山工业品位优化的动态模型。以利润净现值最大为优化目标,根据"Lane"法原理,提出来地下铁矿企业工业品位优化动态模型,并描述了该模型的算法及基本步骤。在案例研究中,系统全面分析了金山店铁矿工业品位系统管理过程中的各种约束条件,建立了适用于该企业工业品位系统优化的数学模型,结合遗传算法,给出了优化计算结果,分析了该优化结果的合理性。
(3)系统比较了静态模型和动态模型的优劣。为了科学评估工业品位优化模型的合理性,本文比较了静态和动态模型的优劣,并在假定条件下,比较了金山店铁矿静态、动态相关计算结果。分析表明,动态优化模型比静态优化模型更有优势。这个结论为工业品位优化决策管理系统提供了决策依据。
(4)研究了成本预测技术。工业品位优化必然涉及到矿山企业生产成本,其中对优化模型和优化结果有重要影响的成本指标是单位精矿可变成本。由于工业品位决策的超前性,以及成本核算的滞后性、以及成本变化的多样性和复杂性,使得单位精矿可变成本预测对矿山企业来讲,难以承担。为了工业品位优化决策的需要,本文提出了单位精矿可变成本的理论算法,为该指标的量化提供了理论依据。
(5)根据工业品位系统分析,应用遗传算法对动态优化模型进行了计算。在金山店铁矿2011年平均工业品位40.35%,对应地质品位41.23%及2011年计划储量总计294.8万吨指标下,按2011年平均回采率0.807,贫化率0.248,回收率0.8188,铁精矿品位65%核算,优化结果为：工业品位23.52%、铁精矿产量120.62万吨、表外矿利用比例58.74%。对比2011年实际铁精矿产量115.35,实际铁精矿产量增产5.37万吨,产生经济效益400余万元。该优化结果非常符合金山店铁矿表外矿储量实际和生产实际。
(6)应用Matlab平台软件,构建“金山店铁矿工业品位优化决策系统”。在合理选择了工业品位优化模型及算法后,本文基于工业品位长期管理的需要,建立了品位管理的支持系统,便于该矿山企业进行日常工业品位的动态管理。
Industrial grade is an index to calculate industrial reserves at a given mining block, so it is a concept closely associating with the geological attribute of mine. In addition, the industrial grade of ore mining is an essentially critical grade between the industrial ore with economic value and the low-grade ore without economic value, therefore, industrial grade has explicit economic meaning. In fact, the change of industrial grade inevitably makes the ore grade and ore quantity changing correspondingly, even with the rate of recovery, dilution rate, concentration recovery ratio of ore changing more or less. This shows, industrial grade is a directly influence factor for the ore grade and ore quantity, which influences the concentrate ore' quantity. What is more, the quantity of concentrate ore is an important means of mine enterprise to gain profit, and the price of concentrate ore is also an important factor for the economic benefit of mine enterprises. So, the dynamics of mineral products price determines the ore industrial grade should also take on dynamic.
To sum up, industrial grade is a dynamic concept, which is highly relvent with income, cost, geological grade, dilution rate, loss rate and so on, it is actually in the mining process associated with those indexs closely, so it is a highly nonlinear function, which would change with time and space complicatedly, in view of this, it is difficulty to try to build its fuction or expressions. Therefore, the grade optimization needs the help of intelligent algorithms or dynamic programming methods. Among them, genetic algorithm, neural network and particle swarm algorithm has been widely used in mining engineering.
From a long-term point of view, industrial grade is also affected by the production technology and opreation environment. With the development of mining and milling of ore technology, it is possible to reduce the ore grade in the process of ore mine, which means industrial grade should be adjusted accordingly. Mine production cost of iron ore enterprises also affects industrial grade too, if cost high, it is necessary to reduce direct cost in the process of mining and milling, which means that high grade ore rises esources should be used as much as possible, so industrial grade should be increased correspondingly. Conversely, industrial grade should be reduced accordingly. Optimization of industrial grade is based on the basic assumptions, which is that ore enterprises would gain reasonable profit or maximum profit from the ore productions, but the iron ore price and production cost of ore is the most important two indexes affecting the economic profit of mining enterprises, in fact, industrial grade has a certain sensitivity to concentrate prices and the cost of management of mining enterprises. Therefore, industrial grade is a complex system includes mining enterprises economic technology and geological properties. The research methods should adopte with system engineering method.
Industrial grade has the obvious technical and economic significance, so its grade value of accounting requires of the economics principles to establish the calculate model. This grade is the critical grade between the low-grade ore and the table of mine, if the economic principle of "incremental revenue equals with marginal cost" used, the marginal grade of low-grade ore mining is the industrial grade, so the calculating model based on this principle is called "static model", which is according to static compare between marginal cost and marginal revenue; if the economic principles of "economics people pursue the economic profit maximization" used, the optimal model of the industrial grade is to be establish based on the industrial grade system analysis, and this model is called dynamic model, just because of the dynamic of model parameters and its solving process. An important premise of the method is distribution of geological grade based on the sample data to be known well, which is very useful to determine the relation between the industrial grade and ore quantity, and all constraint conditions could be clear.
Iron mine of Jinshandian is a typical underground mine, and this iron ore enterprise is selected as a case study of industrial grade optimazation based on grade optimal theory in this paper. For this company, there are some essential information about production management, geological properties of Iron mine, the main technical parameters and its cost structure a required for optimal industrial grade, therefore, this paper gives a detailed analysis of the cost management structure of this enterprise, and calculate the unit variable cost and fixed cost and their annual growth rate based on2009-2011historic data, and analyzes those results. In addition, the price of iron ore is a very important indicator of business environment, so this paper analyzes the the mechanism of price of the Wuhan Iron and Steel Group, and compared the price trend characteristics between imported ore and domestic ore. All those basic analysis is to be a good foundation for optimize the industrial grade.
Accoring to the actual production data of Jinshandian iron mine and other related information, this paper gives the calculation results, such as the basic industrial grade, economic grade and profit grade. Among them, the concept of profit grade is put forward by the writer according to the calculating connotation. According to the latest national standard classification of solid resources reserves, economic grade can be used as industrial grade because of the consistency of practical significance. The results show that the industrial grade optimization result is reasonable, which means that this method has a certain practical value. But it is also some obvious shortcomings, such as iron ore price has large influence, and some indexs inclding the relevant investment loans index, expected profit index and the investment recovery period index are difficult to accurately calculate. Even so, this method still can not deny its practical significance, because its economic meaning is very clear, and the calculation results is ocular, as well as its additional grade corresponds with thinking mode of input and output, what is more, this result is consistent with the industry standard and actual production. Besides, this paper gives another dynamic model according to the profit of net present value maximizing principle. This model has its obvious advantages comparing with that static model, because it considers with the dynamic of the capital of time value and the enterprise business operation, this conclusion has been confirmed by this paper'research. Therefore, the industrial grade optimization decision system should be established by this paper based on the dynamic optimization model.
In short, industrial grade optimization in underground mining is a system engineering, which involves many factors including geological, production technology and economic benefit etc,. This paper studies on industrial grade of iron ore mining in uderground, and comprehensive, systematically summarizes the basic theories and method, meanwhile, this paper is also mixed together or integrated of intelligent algorithms for optimial modeling about the indexs of ore mine. According to those, this paper studies as a case of the iron ore of Jinshandian, and comprehensive analyzes of this mine enterprise grade management related data about production management and its operation, then, the optimization model is built, and the ndustrial grade is optimizated by this model. Lastly, a decision support system of industrial grade optimization is to be constructed based on this model. The main results are as follows:
(1)This paper put forwards the industrial grade optimization of static model in underground mine. The static industrial grade model is derivated by the model of economic grade about low-grade ore utilization in open pit mine at home and abroad, and it combines with the production technical conditions in the underground iron ore, what is more, it applies economics principle of " marginal revenue equals with marginal cost, then, calculated some grade indexs, such as the basic industrial grade to maintain the enterprise daily operation, and the economic grade, which can compensate for investment loans besides the normal operation, and the profit grade, which can ensures the normal production and investment loans compensating, and in particularly, a reasonable profit expectation can be gained. Among them,"profit grade" is a word putfowarded by the writer according to the economic significance of the grade. The calculation result shows that this result accords with the actual production and operation of this iron mine.
(2) Proposes a dynamic model of industrial grade optimization in underground iron mine. This paper gives the dynamic model of industrial grade optimization based on the principle of "Lane", which can carry out the target of NPV of profit maximam, and describes the algorithm of this model and its basic steps. In the case study, after systematic and comprehensive analysis of various constraints about industrial grade in Jinshandian iron mine in the process of production management, the mathematical model is built to optimizate the industrial grade of this enterprise, and calculation results are also obtained by genetic algorithm, then the analysis of results shows the reasonablities of the model.
(3) Systematic comparison the static models with the dynamic models. In order to scientific evaluation optimal models of industrial grade, this paper compares the static with the dynamic models, and compares the calculation results from two types of models based on the data of Jinshandian Iron mine with a lot of assumption conditions. Analysis of comparison results shows that the dynamic model has more advantages than the static model. This conclusion provides the basis of construction the decision supporting system for the industrial grade optimzation.
(4) Studies the forcasting technology of Concentrate production cost. Optimization of industrial grade must involve the production cost indexs of mine enterprises, in which the unit concentrate variable cost has important influence on the optimization models and the optimization results. However, the mine enterprises could generally not forcast the cost, because of the leading of industrial grade decision-making, and the lag of the cost accounting, as well as the cost varieties diversity and complexity, so that the unit variable cost prediction of mining enterprises concentrate. In order to optimization of industrial grade, this paper presents the theoretical arithmetic about the unit concentrate variable cost, and provides the theory basis for quantifying this index.
(5) Carries out calculation of the dynamic optimization model by genetic algorithm, based on the analysis of industrial grade system. In Jinshandian Iron Mine, the average industrial grade is40.35%in2011, geological grade is correspondingly41.23%, while the industrial reserves is totally2948000tons of annual plan. In addition, the average recovery rate is0.807, the dilution rate is0.248, the concentration recovery ratio is0.8188, and the grade of iron concentrate is65%, after calculation, the results of Optimization:industrial grade is23.52%, and the output of iron concentrate is1206200tons, while the utilization ratio of low-grade ore is58.74%. Comparison of actual iron concentrate output of115.35in2011, the actual production is increased by53700tons, and economic benefits improving more than4000.00thousands yuan. The optimization results are accorded with the table of mine reserves and the practical production in Jinshandian Iron Mine.
(6) Constructs the industrial grade optimization decision system of Jinshandian iron mine by Matlab software platform. This paper constructs the industrial grade optimization decision system based on the dynamic model and its optimal algorithm, by doing so, the industrial grade of Jinshandian iron mine can carry out dynamic management of daily operation.
在开采地质品位既定的圈定矿体下,假如矿山企业生产技术水平不变,则精矿产量的变动主要受工业品位变动影响。精矿价格是矿山企业利润的重要决定因素,精矿产量决策受到精矿价格因素制约。精矿价格、精矿产量与工业品位三者之间呈现复杂的动态关系,它们一起共同影响矿山企业的经济效益。精矿价格是不断变化的,圈定矿体的地质品位也会呈现一定的随机波动,这些因素的动态变化必然影响到精矿产量的决策,也需要与之相适应的工业品位优化决策,以实现矿山企业利润最大化。综上所述,工业品位是与矿山地质、企业生产技术以及经营业绩密切相关的技术指标,它与矿山企业经济技术及生产要素构成一个复杂的、高度非线性的系统。传统技术与方法对复杂的、高度非线性系统难以有效模拟其动态行为,因此,工业品位优化需要采用非传统技术与方法,包括智能算法或动态规划法等对其建模与优化。
从中长期来看,工业品位还受到生产技术、经营环境的影响。采选技术的进步,为矿山生产过程降低原矿品位提供了条件,即降低工业品位提供了条件。矿山企业生产成本也影响工业品位,成本高,则需要降低矿石采选过程中的直接成本,这就需要尽可能利用高品位矿石资源,即工业品位相应提高。反之,工业品位应相应降低。工业品位优化是基于矿石生产经营中获取合理利润或最大利润基本假设下进行的,铁精矿价格和矿石生产成本是影响矿山企业经济利润的最重要指标,工业品位对矿山企业精矿价格和管理成本有着一定的敏感性。工业品位的优化还要考虑生产技术、经营环境的综合影响。
工业品位具有明显的技术经济意义,因此该品位的核算应该采用一定的经济学原理,建立核算模型。该品位是表外矿和表内矿的临界品位,如果利用经济学“边际成本=边际收益”,则表外矿利用的边际品位就是工业品位,由此提出的模型主要基于静态时点边际成本与边际收益的比较,因此称为静态模型；如果利用“经济人追求经济利润最大化”的经济学原理,则优化模型是基于矿山企业一段时期内矿石生产的利润净现值最大化模型,此时,矿石生产受到诸多与工业品位有联系的生产技术指标约束。由于该模型参数的动态性和求解过程的动态性,可称为动态优化模型。该方法应用的重要前提是依据样本数据地质品位分布特征,确定工业品位与矿量的对应关系,并明确各种相关生产约束。
金山店铁矿是典型的地下矿山,本文根据地下矿山工业品位优化理论对该矿山进行案例研究。金山店铁矿的矿山地质属性和工业品位管理相关的生产管理状况以及矿山主要经济技术参数,包括它的成本结构和成本特征是工业品位优化所必需的基础资料,为此,本文详细分析了该企业的成本管理结构,并根据企业成本2009-2011年成本资料,分析计算了单位可变成本与固定成本及其年度增长率。铁精矿价格是企业经营环境的重要指标,在基于武钢集团价格形成机制基础上,分析、比较了武钢集团国内矿与进口矿的价格走势特征。这些基础分析,为工业品位优化打下了良好基础。
应用金山店铁矿实际生产经营数据和矿山企业行业相关数据进行了静态工业品位计算,计算结果包括：基本工业品位、经济品位和利润品位。其中,利润品位的概念是本文结合计算内涵提出。根据国家最新固体资源储量分类标准,经济品位是确保矿山企业回收投资的品味,与补偿投资意义下的工业品味内涵一致,因此,“经济品位”也是一种工业品位。同时,矿山企业的利润追求是矿山企业可持续发展的基础,因此,“利润品位”是更高层次意义下的工业品位。计算结果分析表明,静态工业品位优化结果合理,表明该方法具有一定的实际应用价值。但该方法具有一些明显的不足,例如优化结果受铁精矿价格波动影响较大,还受到相关投资贷款指标、预期利润指标和投资回收期指标难以准确核算诸多因素的影响,存在着高价开采高品位矿的倾向,没有考虑资源利用率。尽管如此,仍不可否认它的实际指导意义,因为该方法经济意义明显,其计算结果直观、对应的附加品位符合投入产出思维方式,且实际结果基本符合行业规范和生产实际。根据利润净现值最大化原理,可以建立动态模型。该模型充分考虑了资金的时间价值和生产经营的动态性,与静态模型相比具有较大的优势,这个结论已为本文中的比较研究所证实。因此,本文工业品位优化决策系统以动态优化模型为内核构建。
总之,地下采矿工业品位的优化是一个系统决策工程,该系统涉及包括地质、生产技术及经济效益等诸多因素。本文以地下矿山铁矿石开采的工业品位优化为研究对象,对品位优化的基本理论、方法和应用进行了全面、系统地归纳、总结,融合或集成了矿山指标优化的建模以及智能算法,在此基础上,以金山店铁矿地下开采为案例,全面分析该矿山企业生产经营品位管理相关资料,建立了优化模型,优化了工业品位,并以优化模型为中心,建立了工业品位优化决策支持系统。主要成果如下：
(1)提出了地下矿山工业品位优化的静态模型。根据国内外对露天矿山低品位矿石经济品位的研究模型,结合地下铁矿生产技术条件,应用经济学“边际收益=边际成本”基本定价原理,给出了地下矿山企业的静态工业品位模型,并分别计算了维持企业正常生产经营的基本工业品位、除正常经营外,还可补偿投资贷款的经济工业品位、以及既保证生产,又补偿投资贷款和合理利润预期的利润工业品位。其中,“利润品位”一词是笔者根据该品位对应的经济意义提出的。计算结果符合金山店铁矿目前生产经营实际。
(2)提出了地下矿山工业品位优化的动态模型。以利润净现值最大为优化目标,根据"Lane"法原理,提出来地下铁矿企业工业品位优化动态模型,并描述了该模型的算法及基本步骤。在案例研究中,系统全面分析了金山店铁矿工业品位系统管理过程中的各种约束条件,建立了适用于该企业工业品位系统优化的数学模型,结合遗传算法,给出了优化计算结果,分析了该优化结果的合理性。
(3)系统比较了静态模型和动态模型的优劣。为了科学评估工业品位优化模型的合理性,本文比较了静态和动态模型的优劣,并在假定条件下,比较了金山店铁矿静态、动态相关计算结果。分析表明,动态优化模型比静态优化模型更有优势。这个结论为工业品位优化决策管理系统提供了决策依据。
(4)研究了成本预测技术。工业品位优化必然涉及到矿山企业生产成本,其中对优化模型和优化结果有重要影响的成本指标是单位精矿可变成本。由于工业品位决策的超前性,以及成本核算的滞后性、以及成本变化的多样性和复杂性,使得单位精矿可变成本预测对矿山企业来讲,难以承担。为了工业品位优化决策的需要,本文提出了单位精矿可变成本的理论算法,为该指标的量化提供了理论依据。
(5)根据工业品位系统分析,应用遗传算法对动态优化模型进行了计算。在金山店铁矿2011年平均工业品位40.35%,对应地质品位41.23%及2011年计划储量总计294.8万吨指标下,按2011年平均回采率0.807,贫化率0.248,回收率0.8188,铁精矿品位65%核算,优化结果为：工业品位23.52%、铁精矿产量120.62万吨、表外矿利用比例58.74%。对比2011年实际铁精矿产量115.35,实际铁精矿产量增产5.37万吨,产生经济效益400余万元。该优化结果非常符合金山店铁矿表外矿储量实际和生产实际。
(6)应用Matlab平台软件,构建“金山店铁矿工业品位优化决策系统”。在合理选择了工业品位优化模型及算法后,本文基于工业品位长期管理的需要,建立了品位管理的支持系统,便于该矿山企业进行日常工业品位的动态管理。
Industrial grade is an index to calculate industrial reserves at a given mining block, so it is a concept closely associating with the geological attribute of mine. In addition, the industrial grade of ore mining is an essentially critical grade between the industrial ore with economic value and the low-grade ore without economic value, therefore, industrial grade has explicit economic meaning. In fact, the change of industrial grade inevitably makes the ore grade and ore quantity changing correspondingly, even with the rate of recovery, dilution rate, concentration recovery ratio of ore changing more or less. This shows, industrial grade is a directly influence factor for the ore grade and ore quantity, which influences the concentrate ore' quantity. What is more, the quantity of concentrate ore is an important means of mine enterprise to gain profit, and the price of concentrate ore is also an important factor for the economic benefit of mine enterprises. So, the dynamics of mineral products price determines the ore industrial grade should also take on dynamic.
To sum up, industrial grade is a dynamic concept, which is highly relvent with income, cost, geological grade, dilution rate, loss rate and so on, it is actually in the mining process associated with those indexs closely, so it is a highly nonlinear function, which would change with time and space complicatedly, in view of this, it is difficulty to try to build its fuction or expressions. Therefore, the grade optimization needs the help of intelligent algorithms or dynamic programming methods. Among them, genetic algorithm, neural network and particle swarm algorithm has been widely used in mining engineering.
From a long-term point of view, industrial grade is also affected by the production technology and opreation environment. With the development of mining and milling of ore technology, it is possible to reduce the ore grade in the process of ore mine, which means industrial grade should be adjusted accordingly. Mine production cost of iron ore enterprises also affects industrial grade too, if cost high, it is necessary to reduce direct cost in the process of mining and milling, which means that high grade ore rises esources should be used as much as possible, so industrial grade should be increased correspondingly. Conversely, industrial grade should be reduced accordingly. Optimization of industrial grade is based on the basic assumptions, which is that ore enterprises would gain reasonable profit or maximum profit from the ore productions, but the iron ore price and production cost of ore is the most important two indexes affecting the economic profit of mining enterprises, in fact, industrial grade has a certain sensitivity to concentrate prices and the cost of management of mining enterprises. Therefore, industrial grade is a complex system includes mining enterprises economic technology and geological properties. The research methods should adopte with system engineering method.
Industrial grade has the obvious technical and economic significance, so its grade value of accounting requires of the economics principles to establish the calculate model. This grade is the critical grade between the low-grade ore and the table of mine, if the economic principle of "incremental revenue equals with marginal cost" used, the marginal grade of low-grade ore mining is the industrial grade, so the calculating model based on this principle is called "static model", which is according to static compare between marginal cost and marginal revenue; if the economic principles of "economics people pursue the economic profit maximization" used, the optimal model of the industrial grade is to be establish based on the industrial grade system analysis, and this model is called dynamic model, just because of the dynamic of model parameters and its solving process. An important premise of the method is distribution of geological grade based on the sample data to be known well, which is very useful to determine the relation between the industrial grade and ore quantity, and all constraint conditions could be clear.
Iron mine of Jinshandian is a typical underground mine, and this iron ore enterprise is selected as a case study of industrial grade optimazation based on grade optimal theory in this paper. For this company, there are some essential information about production management, geological properties of Iron mine, the main technical parameters and its cost structure a required for optimal industrial grade, therefore, this paper gives a detailed analysis of the cost management structure of this enterprise, and calculate the unit variable cost and fixed cost and their annual growth rate based on2009-2011historic data, and analyzes those results. In addition, the price of iron ore is a very important indicator of business environment, so this paper analyzes the the mechanism of price of the Wuhan Iron and Steel Group, and compared the price trend characteristics between imported ore and domestic ore. All those basic analysis is to be a good foundation for optimize the industrial grade.
Accoring to the actual production data of Jinshandian iron mine and other related information, this paper gives the calculation results, such as the basic industrial grade, economic grade and profit grade. Among them, the concept of profit grade is put forward by the writer according to the calculating connotation. According to the latest national standard classification of solid resources reserves, economic grade can be used as industrial grade because of the consistency of practical significance. The results show that the industrial grade optimization result is reasonable, which means that this method has a certain practical value. But it is also some obvious shortcomings, such as iron ore price has large influence, and some indexs inclding the relevant investment loans index, expected profit index and the investment recovery period index are difficult to accurately calculate. Even so, this method still can not deny its practical significance, because its economic meaning is very clear, and the calculation results is ocular, as well as its additional grade corresponds with thinking mode of input and output, what is more, this result is consistent with the industry standard and actual production. Besides, this paper gives another dynamic model according to the profit of net present value maximizing principle. This model has its obvious advantages comparing with that static model, because it considers with the dynamic of the capital of time value and the enterprise business operation, this conclusion has been confirmed by this paper'research. Therefore, the industrial grade optimization decision system should be established by this paper based on the dynamic optimization model.
In short, industrial grade optimization in underground mining is a system engineering, which involves many factors including geological, production technology and economic benefit etc,. This paper studies on industrial grade of iron ore mining in uderground, and comprehensive, systematically summarizes the basic theories and method, meanwhile, this paper is also mixed together or integrated of intelligent algorithms for optimial modeling about the indexs of ore mine. According to those, this paper studies as a case of the iron ore of Jinshandian, and comprehensive analyzes of this mine enterprise grade management related data about production management and its operation, then, the optimization model is built, and the ndustrial grade is optimizated by this model. Lastly, a decision support system of industrial grade optimization is to be constructed based on this model. The main results are as follows:
(1)This paper put forwards the industrial grade optimization of static model in underground mine. The static industrial grade model is derivated by the model of economic grade about low-grade ore utilization in open pit mine at home and abroad, and it combines with the production technical conditions in the underground iron ore, what is more, it applies economics principle of " marginal revenue equals with marginal cost, then, calculated some grade indexs, such as the basic industrial grade to maintain the enterprise daily operation, and the economic grade, which can compensate for investment loans besides the normal operation, and the profit grade, which can ensures the normal production and investment loans compensating, and in particularly, a reasonable profit expectation can be gained. Among them,"profit grade" is a word putfowarded by the writer according to the economic significance of the grade. The calculation result shows that this result accords with the actual production and operation of this iron mine.
(2) Proposes a dynamic model of industrial grade optimization in underground iron mine. This paper gives the dynamic model of industrial grade optimization based on the principle of "Lane", which can carry out the target of NPV of profit maximam, and describes the algorithm of this model and its basic steps. In the case study, after systematic and comprehensive analysis of various constraints about industrial grade in Jinshandian iron mine in the process of production management, the mathematical model is built to optimizate the industrial grade of this enterprise, and calculation results are also obtained by genetic algorithm, then the analysis of results shows the reasonablities of the model.
(3) Systematic comparison the static models with the dynamic models. In order to scientific evaluation optimal models of industrial grade, this paper compares the static with the dynamic models, and compares the calculation results from two types of models based on the data of Jinshandian Iron mine with a lot of assumption conditions. Analysis of comparison results shows that the dynamic model has more advantages than the static model. This conclusion provides the basis of construction the decision supporting system for the industrial grade optimzation.
(4) Studies the forcasting technology of Concentrate production cost. Optimization of industrial grade must involve the production cost indexs of mine enterprises, in which the unit concentrate variable cost has important influence on the optimization models and the optimization results. However, the mine enterprises could generally not forcast the cost, because of the leading of industrial grade decision-making, and the lag of the cost accounting, as well as the cost varieties diversity and complexity, so that the unit variable cost prediction of mining enterprises concentrate. In order to optimization of industrial grade, this paper presents the theoretical arithmetic about the unit concentrate variable cost, and provides the theory basis for quantifying this index.
(5) Carries out calculation of the dynamic optimization model by genetic algorithm, based on the analysis of industrial grade system. In Jinshandian Iron Mine, the average industrial grade is40.35%in2011, geological grade is correspondingly41.23%, while the industrial reserves is totally2948000tons of annual plan. In addition, the average recovery rate is0.807, the dilution rate is0.248, the concentration recovery ratio is0.8188, and the grade of iron concentrate is65%, after calculation, the results of Optimization:industrial grade is23.52%, and the output of iron concentrate is1206200tons, while the utilization ratio of low-grade ore is58.74%. Comparison of actual iron concentrate output of115.35in2011, the actual production is increased by53700tons, and economic benefits improving more than4000.00thousands yuan. The optimization results are accorded with the table of mine reserves and the practical production in Jinshandian Iron Mine.
(6) Constructs the industrial grade optimization decision system of Jinshandian iron mine by Matlab software platform. This paper constructs the industrial grade optimization decision system based on the dynamic model and its optimal algorithm, by doing so, the industrial grade of Jinshandian iron mine can carry out dynamic management of daily operation.
引文
[1]Montimer G.J., Grade Control[J], Trans. Inst. Min. and Metal,1950(3),357-399.
[2]Erickson J.D., Long-time Open Pit Planning[J], Mining Engineering,1968(2),75-88。
[3 J Taylor H.K., General background theory of cutoff grade. Trans. Instn Min. Metall(Seet. A:Min. industry),81,1972, A160-179.
[4]彭兆德、云庆夏等,采矿手册第四卷[M],北京：冶金工业出版社,1991,24-28.
[5]陈希廉等,矿业经济学[M],北京：中国国际广播出版社,1992,27-33.
[6]郑学敏,矿床工业品位分析与经济品位应用[J],2005, Tsinghua Tongfang Optical Disc. Co., Ltd.10-13.
[7]胡铁流,市场经济条件下矿产工业指标的变革[J],金属矿山,1998(2),18-20.
[8]苑香民、王会文等.黄金矿山采矿规模设计的不确定性分析[J],黄金,2002(12),18-21.
[9]李德斌.利用盈亏平衡品位进行黑斑矿的采选决策[J],有色金属,2003(11),173-176.
[10]向鹏成.应用盈亏平衡分析法确定低品位矿利用的临界品位[J],冶金矿山,设计与建设,2001(11),24-27.
[11]李泽临,边界品位小议[J],有的金属,1951(6),39-40.
[12]王惠民,浅谈矿石工业指标[J],化工矿山技术,1952(3),47-50.
[13]任启明,确定矿山工业指标中几个问题的探讨[J1,冶金经济分析,1983(4),36-41.
[14]王介甫,许楚彬,关于矿床工业指标的研究[J],化工矿山技术,1984(6),9-14.
[15]侯景儒,黄竟先等,对数正态克立格法的理论及应用[J],北京科技大学学报,1998(11),391-398.
[16]侯景儒,潘汉军等,多元地质统计学的基本理论与方法[J],北京科技大学学报,1992(14),95-103.
[17]侯景儒,王志民等,湖南桃江锰矿地质统计学及数学地质研究[J],中国南方锰矿地质,1992,33(3)：34-40.
[18]刘新.论最佳边界品位及其确定.地质经济动态,1985,(1)：25-34.
[19]徐熹,矿石级差品位指标体系与兰纳的边界品位理论[J],数量经济技术经济研究,1985(14),41-47
[20]任毅,市场经济下金属露天矿边界品位的优化与控制[J],金属矿山,1997(2),1-6.
[21]任毅,露天矿次边界品位和次品位堆储的优化决策[J],有色金属,1997(3),10-17.
[22]黄光球,露天开采矿体边界品位动态优化新方法[J],中国锰业,1995(5),18-22.
[23]黄光球,地下开采矿体边界品位动态优化法[J],化工矿山技术,1997(4),1-6.
[24]谢英亮,露天矿边际品位最优化的经济分析[D],长沙：中南大学,2004.
[25]谢英亮,基于控制论的露天矿边际品位最优化[J],金属矿山,2000(11),14-16.
[26]谢英亮,基于期权定价的矿业投资决策[J],冶金矿山设计与建设,2001(4),22-25.
[27]谢英亮,露天矿边际品位优化及其动态调整模型[Jl,金属矿山,1996(7), 11-17.
[28]谢英亮,半投入产出模型及求解[J],系统工程理论与实践,1996(7),7-11.
[29]黄光球,露天矿边界品位的动态最优化及人工神经网络的应用[J],中国矿业,1996(4),32—38.
[30]赵德孝,柳小波等,金山店铁矿入选品位动态优化数学模型建立[J],金属矿山,2004(4),8-10.
[31]姜谙男,赵德孝等,矿石入选品位优化决策支持系统的开发[J],矿业研究与开发,2003(4),43-45.
[32]云庆夏,矿石品位优化的遗传算法[J],金属矿山,1997(5),13-16.
[33]李克庆,刘保顺,确定合理入选品位的通用系统1J],北京科技大学学报,1997(10),28-31.
[34]张玉清.用动态法确定开采工业品位的探讨.矿山地质,1986,(1)：12-15.
[35]叶胜德.矿床经济评价中的贴现方法探讨.矿产地质动态,1985,(11)：28-31.
[36]王成发.关于经济地质学的几个问题.矿产地质动态,1984,(3)：6-9.
[37]钱抗生.开创冶金地质工作新局面.矿产地质动态,1984,(10)：18-20.
[38]扬久柱.关于合理确定黑钨脉状矿开采价值问题研究.有色金属(矿山)1986,(3)：9-13.
[39]王青,史维祥主编,采矿学[M],北京：冶金工业出版社,2001,327,377-387.
[40]Vikers E., Marginal analysis, its application in determining cutoff grade. Min. Eng.N.Y.13. 1961,578-582.
[41]王青、王之静.露天开采整体优化-理论模型和算法[M]。北京：冶金工业出版社,2006,56-60.
[42]Lane, K.F., Choosing Cutoff Grade[J], Colorado School of Mines quarterly 1964, (59-4), 811-829.
[43]Lane, K.F., Commercial Aspects of Choosing Cutoff Grade[J],16th Apcom Symposium, 1979,280-285.
[44][28] Taylor H. K., The northern Porphyries:Local features and their effects on Valuation and Planning. Min. Eng. N.Y.1. Nov.1980,41-67.
[45]Taylor, H.K, Cutoff Grades-Some Further Reflections, Trans. Inst. Min. Metal-(Section A, Mining Industry),1985(5), A04-16.
[46]Musgrove P., Lead grade-tonnage relation. Mining Magzine, London 112,1965,36-44.
[47]Omer Mol, Cutoff grade determination for mines producing direct-shipping Iron ore. Proe. Australas. Inst. Min.Metall.No.289,1984,283-286.
[48]Henning U, Calculation of cutoff grade, Can.Min. J.,84 March 1963,54-57.
[49]Taylor H.K. Cutoff grades-some further reflections. Trans. Inst. Min. Metall.(Seet.A:Min, industry),94,1985, A2O4-216.
[50]贾文龙,陈甲斌.低品位铁矿资源开发经济性分析[J].金属矿山,2009(3)：1-3.
[51]谢英亮,李玉.表外矿利用的临界品位分析模型[J].矿产保护与利用,2000(2)：1-2.
[52]李敏强,范纪淞.遗传算法的基本理论与应用(M].北京：科学出版社,2002.
[53]王凌.智能优化方法及其应用[M].北京：清华大学出版社,2001.
[54]席裕庚,柴天佑,-阵为民,遗传算法综述[J].控制理论与应用,1996,13(6)：697-705.
[55]任平.遗传算法综述[J].工程数学学报,1999,16(1)：1-8.
[56]M.Berry and G. Linoff, Data Mining Techniques[J]. John Wiley,1997.
[57]C.Westphal and T.Blaxton, Data Mining Solutions[J]. John Wiley,1998.
[58]R.Groth. Data Mining:Building Competitive Advantage[J]. Prentiee-Hall,1997.
[59]A.Berson, J.Smith. Data Mingehousing and OLAP[J]. McGraw-Hill,1997.
[60]肖故安,李腊元.数据挖掘与知识发现的理论方法及技术分析.交通与计算机,2002,(1)：57-61.
[61]吴少敏,冯建生.宝钢数据挖掘系统.宝钢技术,2001,(1)：43-47.
[62]王小平,曹立明.遗传算法一理论、应用与软件实现[M].西安：西安交通大学出版社,2002.
[63]贺勇,诸克军,徐思新,刘婷,李明.一种Pso和ANN集成的品位优化模型[J],运筹与管理,2009,18(2)：91-94.
[64]刘鹏飞等.基于多目标粒子群的对接函数综合[J].微计算机信息,2009,25(3)：197-199.
[65]Rumelhart, D E, G. E. Hinton, R. J. Williams, Learning representations by back-Propagating errors [J]. Nature,1986,323(9):533-536.
[66]Moody J, Darken C. Learning with Localized receptive fields [C]. Proeeedings of the 1988 Connectionist Models Summer School, San Mateo, CA,1988, Carnegie Mellon University, Morgan Kaufmann.
[67]Moody J, Darken C. Fast learning in networks of locally-tuned processing, units [J]. Neural Computation,1989,1(2):281-294.
[68]Ma Xiaomin. Inverse Identification and Closed-Loop Control of Dynamic Systems Using Neural Networks [J].控制理论与应用,1997,14(6)：829-836.
[69]Rovithakis. G.A, Christodoulou. M.A. Adaptive Control of Unknown Plants Using Dynamical Neural Networks. IEEE Trans Syst Man Cyber,1994,24(3):400-412.
[70]Hopfield. J.J. Neural Networks and Physical Systems with Emergent Collective Computational Abilities [C]. In:Proc. Natl. Acad. Sci.,1982,79:2554-2558.
[71]Hopfield. J. J. Neurons with graded response have collective computational properties like those of two state neuron[C]. In:Proc. Natl. Acad. Sci.,1984,81:3088-3092.
[72]Hopfield. J. J., Tank. D. W., Neural computation of decision in optimization problem [J]. Biological, Cybernetics,1985,52:141-152.
[73]Roman R.J., The Use of Dynamic Programming for Determining Mine-mill Production Schedules[J], Application of Computer Methods in the Mineral Industry (Johannesburg: South African), IMM 1973,65-70.
[74]Jorgen Elbrond etc., Use of an Interactive Dynamic Programming System as an Aid to Mine Evaluation[J], Proeeeding,17th Apcom,463-474.
[75]Yves Lizotte, Jorgen Elbrond. Choice of Mine-mill Capacities and Production Schedules Using Open-ended Dynamic Programming[J], CIM Bulletin,1982(3),154-163.
[76]Dowd P. A., The Use of Dynamic Programming in Cutoff Grade Determination[J], Tech. Rep. DeP. Genie Miner,1975(4),11-29.
[77]Dowd P. A., Application of Dynamic and Stochastic Programming to Optimize Cutoff Grades and production rates[J], Trans. Inst. Mining,1976(1), Sect. A, VOL.85, A22-31.
[78]Noren N.E., Long-Range Decision Models in Mining[J], Economic Research Institute, 1969(12),240-253.
[79]N.M. Satybaldiev, A.M. Freidin, Optimization of Mineral mining Geometry and Mine Production Capacity[J], Journal of Ming Science, VOL.42,2006(5),27-34.
[80]Qing Wang, Long-term Open-pit Production Scheduling Through Dynamic Phrase-bench Scheduling[J], Trans. Inst. Min. Metal. (Section A:Mine Industry),105, May-August, 1996, A99-103.
[81]初道忠,王青,郑怀昌.边界品位的动态规划优化方法及其应用[J].中国矿业,2008,17(9)：68.
[82]梁中扬,李克庆.湖北三鑫金铜股份有限公司生产技术指标的优化研究[J].黄金,2006,27(12)：31
[83]黄晓乃,曹鼎阶等.应用数理统计估值法测算320铀矿床的边界品位[J].有色金属,2005,57(3)：86—90.
[84]解世俊,矿床地下开采理论与实践[M],北京：冶金工业出版社,1990,27-36.
[85]张玉清,矿山技术经济学[M],北京：冶金工业出版社,1987,62-67.
[86]Robert D. Cairns, Takayoshi shinkuma, The choice of the cutoff grade in mining[J], Resource Policy,2003(29),75-81.
[87]解世俊,金属矿床地下开采[M],北京：冶金工业出版社,1986,89-101.
[88]王青,史维祥主编,采矿学[M],北京：冶金工业出版社,2001,327,377-387.
[89]李祥仪、李仲学,矿业经济学[M],北京：冶金工业出版社,2001,89-90.
[90]彭兆德、云庆夏等,采矿手册第七卷[M],北京：冶金工业出版社,1991,38-39.
[91]谢英亮,露天矿边际品位最优化的经济分析[D],长沙：中南大学,2004.
[92]任毅、孙英健,市场价格波动与选别式地下开采边界品位控制[J],有色金属,1998(8),21-27.
[93]谢英亮、高阳,论矿产品价格波动与采矿边际品位的调整[J],中国矿业,2003(6),70—72.
[94]Taylor, H.K, Cutoff grade-some further reflections[J], Trans. Inst. Mining Metallurgy,1985, Vol.94, A204-216.
[95]高福德、吴双生等,矿山开发地质经济概论[M],北京：冶金工业出版社,1991,10-15.
[96]曹乐农等.攀枝花铁矿露采境界内表外矿合理利用[J].矿山地质,1966,(6)
[97]袁怀雨,舒航.矿产经济理论与实践[M].北京：冶金工业出版社,1993.152-158.
[98]赵文才.我国重点铁矿平衡表外矿量和顺便开采[J].矿山地质,1993,14(1)：60-61.
[99]Lane, K.F., Economic Defintion of ore-Cutoff grade in Theory and Practice[J], Mining Journal books, London 1988,211-232.
[100]Lane,K.F., Commercial Aspeets of Choosing Cutoff grade[J],16th APCOM Sylnposium,1979,280-285.
[101]谢英亮.露天矿边际品位最优化的经济分析[D].长沙：中南大学,2004：54-56.
[102]初道忠.地下开采边界品位动态优化研究及其应用[D].沈阳：东北大学,2008：62-64.
[103]赵增辉.地下采矿边界品位优化研究[D].沈阳：东北大学,2008：2.
[104]赵德孝,柳小波等.金山店铁矿入选品位动态优化数学模型建立[J].金属矿山,2004(4)：8.
[105]陈清运,杨从兵等.金山店铁矿崩落法过渡分段放矿指标优化研究[J].金属矿山,2010(4)：24-25.
[106]李讯.金山店铁矿出矿优化模型及应用研究[D].武汉：武汉科技大学,2010：20.
[107]程爱平,许梦国等.金山店铁矿低贫化放矿数值模拟[J].金属矿山,2011(1)：31.
[108]秦茂林.金山店铁矿采矿运输系统优化研究[D].武汉：武汉科技大学,2009.
[109]罗文斌.德瑞克高频振动细筛在金山店铁矿选矿厂的应用[J].现代矿业,2009(9)：4-5.
[110]黄晓东.多金属矿开采的经济性评价[J].南方冶金学院学报,2003,24(4)：5-8.
[111]李柏林.采矿权价值评估研究[D].昆明：昆明理工大学采矿工程,2001.
[112]李菊平.矿石可采品位及可采厚度的技术经济分析[J].冶金经济分析,1987,(4)：23-29.
[113]樊春辉.低品位矿资源量的归类、圈定和估算问题的探讨[J].中国矿业,2010,19(7)：47-48.
[114]江四海,李德斌.运用盈亏平衡品位进行低品位矿采选决策[J].江西有色金属,2006,20(1)：22-25.
[115]向鹏程,谢英亮.应用盈亏平衡分析法确定低品位矿利用的临界品位[J].冶金矿山设计与建设,2001,23(6)：24-25.
[116]Taylor, H.K, Mine valuation and Feasibility studies[J], Mineral Industry Cost, Northwest Mining Association,1977(2),1-17.
[117]Halls. K. F., Determination of optimum ore reserves and plant size by Incremental financial analysis. CIM.72. August.1969.
[118]谢英亮,高阳.露天矿边际品位优化模型—边际分析和机会成本分析方法的应用[J].江西有色金属,2003,7(3)：13-17.
[119]Lane, K.F., Economic Defintion of ore-Cutoff grade in Theory and Practice[J], Mining Journal books, London 1988,211-232.
[120]魏一鸣、童光煦,矿床技术经济评价现状及未来发展[J],黄金,1995(8),10-12.
[121]Robert D.Cairns, Takayoshi Shinkuma, The Choice of Cutoff Grade in Mining[J], Resource policy[J],2003(29),75-81.
[122]Erickson J.D. Long-range Openpit Planning. Min. Eng., N.Y 20. April,1968,5-78.
[123]张晓,戴冠中,徐乃平.一种新的优化搜索算法一遗传算法[J].控制理论与应用,1995,12(3)：265-273.
[124]韩力群.人工神经网络教程[M].北京：北京邮电大学出版社,2006.
[125]黄光球.复杂信息条件下矿石品位推断的神经网络方法[J].西部探矿工程,1996,8(1)：28-32.
[126]王淑红.基于神经网络的选矿厂预测模型及应用研究D],昆明：昆明理工大学,2002.
[127]邓衍义.Matlab在露天矿品位参数优化中的应用[J].江西理工大学学报,2008,29(3)：13-16.
[2]Erickson J.D., Long-time Open Pit Planning[J], Mining Engineering,1968(2),75-88。
[3 J Taylor H.K., General background theory of cutoff grade. Trans. Instn Min. Metall(Seet. A:Min. industry),81,1972, A160-179.
[4]彭兆德、云庆夏等,采矿手册第四卷[M],北京：冶金工业出版社,1991,24-28.
[5]陈希廉等,矿业经济学[M],北京：中国国际广播出版社,1992,27-33.
[6]郑学敏,矿床工业品位分析与经济品位应用[J],2005, Tsinghua Tongfang Optical Disc. Co., Ltd.10-13.
[7]胡铁流,市场经济条件下矿产工业指标的变革[J],金属矿山,1998(2),18-20.
[8]苑香民、王会文等.黄金矿山采矿规模设计的不确定性分析[J],黄金,2002(12),18-21.
[9]李德斌.利用盈亏平衡品位进行黑斑矿的采选决策[J],有色金属,2003(11),173-176.
[10]向鹏成.应用盈亏平衡分析法确定低品位矿利用的临界品位[J],冶金矿山,设计与建设,2001(11),24-27.
[11]李泽临,边界品位小议[J],有的金属,1951(6),39-40.
[12]王惠民,浅谈矿石工业指标[J],化工矿山技术,1952(3),47-50.
[13]任启明,确定矿山工业指标中几个问题的探讨[J1,冶金经济分析,1983(4),36-41.
[14]王介甫,许楚彬,关于矿床工业指标的研究[J],化工矿山技术,1984(6),9-14.
[15]侯景儒,黄竟先等,对数正态克立格法的理论及应用[J],北京科技大学学报,1998(11),391-398.
[16]侯景儒,潘汉军等,多元地质统计学的基本理论与方法[J],北京科技大学学报,1992(14),95-103.
[17]侯景儒,王志民等,湖南桃江锰矿地质统计学及数学地质研究[J],中国南方锰矿地质,1992,33(3)：34-40.
[18]刘新.论最佳边界品位及其确定.地质经济动态,1985,(1)：25-34.
[19]徐熹,矿石级差品位指标体系与兰纳的边界品位理论[J],数量经济技术经济研究,1985(14),41-47
[20]任毅,市场经济下金属露天矿边界品位的优化与控制[J],金属矿山,1997(2),1-6.
[21]任毅,露天矿次边界品位和次品位堆储的优化决策[J],有色金属,1997(3),10-17.
[22]黄光球,露天开采矿体边界品位动态优化新方法[J],中国锰业,1995(5),18-22.
[23]黄光球,地下开采矿体边界品位动态优化法[J],化工矿山技术,1997(4),1-6.
[24]谢英亮,露天矿边际品位最优化的经济分析[D],长沙：中南大学,2004.
[25]谢英亮,基于控制论的露天矿边际品位最优化[J],金属矿山,2000(11),14-16.
[26]谢英亮,基于期权定价的矿业投资决策[J],冶金矿山设计与建设,2001(4),22-25.
[27]谢英亮,露天矿边际品位优化及其动态调整模型[Jl,金属矿山,1996(7), 11-17.
[28]谢英亮,半投入产出模型及求解[J],系统工程理论与实践,1996(7),7-11.
[29]黄光球,露天矿边界品位的动态最优化及人工神经网络的应用[J],中国矿业,1996(4),32—38.
[30]赵德孝,柳小波等,金山店铁矿入选品位动态优化数学模型建立[J],金属矿山,2004(4),8-10.
[31]姜谙男,赵德孝等,矿石入选品位优化决策支持系统的开发[J],矿业研究与开发,2003(4),43-45.
[32]云庆夏,矿石品位优化的遗传算法[J],金属矿山,1997(5),13-16.
[33]李克庆,刘保顺,确定合理入选品位的通用系统1J],北京科技大学学报,1997(10),28-31.
[34]张玉清.用动态法确定开采工业品位的探讨.矿山地质,1986,(1)：12-15.
[35]叶胜德.矿床经济评价中的贴现方法探讨.矿产地质动态,1985,(11)：28-31.
[36]王成发.关于经济地质学的几个问题.矿产地质动态,1984,(3)：6-9.
[37]钱抗生.开创冶金地质工作新局面.矿产地质动态,1984,(10)：18-20.
[38]扬久柱.关于合理确定黑钨脉状矿开采价值问题研究.有色金属(矿山)1986,(3)：9-13.
[39]王青,史维祥主编,采矿学[M],北京：冶金工业出版社,2001,327,377-387.
[40]Vikers E., Marginal analysis, its application in determining cutoff grade. Min. Eng.N.Y.13. 1961,578-582.
[41]王青、王之静.露天开采整体优化-理论模型和算法[M]。北京：冶金工业出版社,2006,56-60.
[42]Lane, K.F., Choosing Cutoff Grade[J], Colorado School of Mines quarterly 1964, (59-4), 811-829.
[43]Lane, K.F., Commercial Aspects of Choosing Cutoff Grade[J],16th Apcom Symposium, 1979,280-285.
[44][28] Taylor H. K., The northern Porphyries:Local features and their effects on Valuation and Planning. Min. Eng. N.Y.1. Nov.1980,41-67.
[45]Taylor, H.K, Cutoff Grades-Some Further Reflections, Trans. Inst. Min. Metal-(Section A, Mining Industry),1985(5), A04-16.
[46]Musgrove P., Lead grade-tonnage relation. Mining Magzine, London 112,1965,36-44.
[47]Omer Mol, Cutoff grade determination for mines producing direct-shipping Iron ore. Proe. Australas. Inst. Min.Metall.No.289,1984,283-286.
[48]Henning U, Calculation of cutoff grade, Can.Min. J.,84 March 1963,54-57.
[49]Taylor H.K. Cutoff grades-some further reflections. Trans. Inst. Min. Metall.(Seet.A:Min, industry),94,1985, A2O4-216.
[50]贾文龙,陈甲斌.低品位铁矿资源开发经济性分析[J].金属矿山,2009(3)：1-3.
[51]谢英亮,李玉.表外矿利用的临界品位分析模型[J].矿产保护与利用,2000(2)：1-2.
[52]李敏强,范纪淞.遗传算法的基本理论与应用(M].北京：科学出版社,2002.
[53]王凌.智能优化方法及其应用[M].北京：清华大学出版社,2001.
[54]席裕庚,柴天佑,-阵为民,遗传算法综述[J].控制理论与应用,1996,13(6)：697-705.
[55]任平.遗传算法综述[J].工程数学学报,1999,16(1)：1-8.
[56]M.Berry and G. Linoff, Data Mining Techniques[J]. John Wiley,1997.
[57]C.Westphal and T.Blaxton, Data Mining Solutions[J]. John Wiley,1998.
[58]R.Groth. Data Mining:Building Competitive Advantage[J]. Prentiee-Hall,1997.
[59]A.Berson, J.Smith. Data Mingehousing and OLAP[J]. McGraw-Hill,1997.
[60]肖故安,李腊元.数据挖掘与知识发现的理论方法及技术分析.交通与计算机,2002,(1)：57-61.
[61]吴少敏,冯建生.宝钢数据挖掘系统.宝钢技术,2001,(1)：43-47.
[62]王小平,曹立明.遗传算法一理论、应用与软件实现[M].西安：西安交通大学出版社,2002.
[63]贺勇,诸克军,徐思新,刘婷,李明.一种Pso和ANN集成的品位优化模型[J],运筹与管理,2009,18(2)：91-94.
[64]刘鹏飞等.基于多目标粒子群的对接函数综合[J].微计算机信息,2009,25(3)：197-199.
[65]Rumelhart, D E, G. E. Hinton, R. J. Williams, Learning representations by back-Propagating errors [J]. Nature,1986,323(9):533-536.
[66]Moody J, Darken C. Learning with Localized receptive fields [C]. Proeeedings of the 1988 Connectionist Models Summer School, San Mateo, CA,1988, Carnegie Mellon University, Morgan Kaufmann.
[67]Moody J, Darken C. Fast learning in networks of locally-tuned processing, units [J]. Neural Computation,1989,1(2):281-294.
[68]Ma Xiaomin. Inverse Identification and Closed-Loop Control of Dynamic Systems Using Neural Networks [J].控制理论与应用,1997,14(6)：829-836.
[69]Rovithakis. G.A, Christodoulou. M.A. Adaptive Control of Unknown Plants Using Dynamical Neural Networks. IEEE Trans Syst Man Cyber,1994,24(3):400-412.
[70]Hopfield. J.J. Neural Networks and Physical Systems with Emergent Collective Computational Abilities [C]. In:Proc. Natl. Acad. Sci.,1982,79:2554-2558.
[71]Hopfield. J. J. Neurons with graded response have collective computational properties like those of two state neuron[C]. In:Proc. Natl. Acad. Sci.,1984,81:3088-3092.
[72]Hopfield. J. J., Tank. D. W., Neural computation of decision in optimization problem [J]. Biological, Cybernetics,1985,52:141-152.
[73]Roman R.J., The Use of Dynamic Programming for Determining Mine-mill Production Schedules[J], Application of Computer Methods in the Mineral Industry (Johannesburg: South African), IMM 1973,65-70.
[74]Jorgen Elbrond etc., Use of an Interactive Dynamic Programming System as an Aid to Mine Evaluation[J], Proeeeding,17th Apcom,463-474.
[75]Yves Lizotte, Jorgen Elbrond. Choice of Mine-mill Capacities and Production Schedules Using Open-ended Dynamic Programming[J], CIM Bulletin,1982(3),154-163.
[76]Dowd P. A., The Use of Dynamic Programming in Cutoff Grade Determination[J], Tech. Rep. DeP. Genie Miner,1975(4),11-29.
[77]Dowd P. A., Application of Dynamic and Stochastic Programming to Optimize Cutoff Grades and production rates[J], Trans. Inst. Mining,1976(1), Sect. A, VOL.85, A22-31.
[78]Noren N.E., Long-Range Decision Models in Mining[J], Economic Research Institute, 1969(12),240-253.
[79]N.M. Satybaldiev, A.M. Freidin, Optimization of Mineral mining Geometry and Mine Production Capacity[J], Journal of Ming Science, VOL.42,2006(5),27-34.
[80]Qing Wang, Long-term Open-pit Production Scheduling Through Dynamic Phrase-bench Scheduling[J], Trans. Inst. Min. Metal. (Section A:Mine Industry),105, May-August, 1996, A99-103.
[81]初道忠,王青,郑怀昌.边界品位的动态规划优化方法及其应用[J].中国矿业,2008,17(9)：68.
[82]梁中扬,李克庆.湖北三鑫金铜股份有限公司生产技术指标的优化研究[J].黄金,2006,27(12)：31
[83]黄晓乃,曹鼎阶等.应用数理统计估值法测算320铀矿床的边界品位[J].有色金属,2005,57(3)：86—90.
[84]解世俊,矿床地下开采理论与实践[M],北京：冶金工业出版社,1990,27-36.
[85]张玉清,矿山技术经济学[M],北京：冶金工业出版社,1987,62-67.
[86]Robert D. Cairns, Takayoshi shinkuma, The choice of the cutoff grade in mining[J], Resource Policy,2003(29),75-81.
[87]解世俊,金属矿床地下开采[M],北京：冶金工业出版社,1986,89-101.
[88]王青,史维祥主编,采矿学[M],北京：冶金工业出版社,2001,327,377-387.
[89]李祥仪、李仲学,矿业经济学[M],北京：冶金工业出版社,2001,89-90.
[90]彭兆德、云庆夏等,采矿手册第七卷[M],北京：冶金工业出版社,1991,38-39.
[91]谢英亮,露天矿边际品位最优化的经济分析[D],长沙：中南大学,2004.
[92]任毅、孙英健,市场价格波动与选别式地下开采边界品位控制[J],有色金属,1998(8),21-27.
[93]谢英亮、高阳,论矿产品价格波动与采矿边际品位的调整[J],中国矿业,2003(6),70—72.
[94]Taylor, H.K, Cutoff grade-some further reflections[J], Trans. Inst. Mining Metallurgy,1985, Vol.94, A204-216.
[95]高福德、吴双生等,矿山开发地质经济概论[M],北京：冶金工业出版社,1991,10-15.
[96]曹乐农等.攀枝花铁矿露采境界内表外矿合理利用[J].矿山地质,1966,(6)
[97]袁怀雨,舒航.矿产经济理论与实践[M].北京：冶金工业出版社,1993.152-158.
[98]赵文才.我国重点铁矿平衡表外矿量和顺便开采[J].矿山地质,1993,14(1)：60-61.
[99]Lane, K.F., Economic Defintion of ore-Cutoff grade in Theory and Practice[J], Mining Journal books, London 1988,211-232.
[100]Lane,K.F., Commercial Aspeets of Choosing Cutoff grade[J],16th APCOM Sylnposium,1979,280-285.
[101]谢英亮.露天矿边际品位最优化的经济分析[D].长沙：中南大学,2004：54-56.
[102]初道忠.地下开采边界品位动态优化研究及其应用[D].沈阳：东北大学,2008：62-64.
[103]赵增辉.地下采矿边界品位优化研究[D].沈阳：东北大学,2008：2.
[104]赵德孝,柳小波等.金山店铁矿入选品位动态优化数学模型建立[J].金属矿山,2004(4)：8.
[105]陈清运,杨从兵等.金山店铁矿崩落法过渡分段放矿指标优化研究[J].金属矿山,2010(4)：24-25.
[106]李讯.金山店铁矿出矿优化模型及应用研究[D].武汉：武汉科技大学,2010：20.
[107]程爱平,许梦国等.金山店铁矿低贫化放矿数值模拟[J].金属矿山,2011(1)：31.
[108]秦茂林.金山店铁矿采矿运输系统优化研究[D].武汉：武汉科技大学,2009.
[109]罗文斌.德瑞克高频振动细筛在金山店铁矿选矿厂的应用[J].现代矿业,2009(9)：4-5.
[110]黄晓东.多金属矿开采的经济性评价[J].南方冶金学院学报,2003,24(4)：5-8.
[111]李柏林.采矿权价值评估研究[D].昆明：昆明理工大学采矿工程,2001.
[112]李菊平.矿石可采品位及可采厚度的技术经济分析[J].冶金经济分析,1987,(4)：23-29.
[113]樊春辉.低品位矿资源量的归类、圈定和估算问题的探讨[J].中国矿业,2010,19(7)：47-48.
[114]江四海,李德斌.运用盈亏平衡品位进行低品位矿采选决策[J].江西有色金属,2006,20(1)：22-25.
[115]向鹏程,谢英亮.应用盈亏平衡分析法确定低品位矿利用的临界品位[J].冶金矿山设计与建设,2001,23(6)：24-25.
[116]Taylor, H.K, Mine valuation and Feasibility studies[J], Mineral Industry Cost, Northwest Mining Association,1977(2),1-17.
[117]Halls. K. F., Determination of optimum ore reserves and plant size by Incremental financial analysis. CIM.72. August.1969.
[118]谢英亮,高阳.露天矿边际品位优化模型—边际分析和机会成本分析方法的应用[J].江西有色金属,2003,7(3)：13-17.
[119]Lane, K.F., Economic Defintion of ore-Cutoff grade in Theory and Practice[J], Mining Journal books, London 1988,211-232.
[120]魏一鸣、童光煦,矿床技术经济评价现状及未来发展[J],黄金,1995(8),10-12.
[121]Robert D.Cairns, Takayoshi Shinkuma, The Choice of Cutoff Grade in Mining[J], Resource policy[J],2003(29),75-81.
[122]Erickson J.D. Long-range Openpit Planning. Min. Eng., N.Y 20. April,1968,5-78.
[123]张晓,戴冠中,徐乃平.一种新的优化搜索算法一遗传算法[J].控制理论与应用,1995,12(3)：265-273.
[124]韩力群.人工神经网络教程[M].北京：北京邮电大学出版社,2006.
[125]黄光球.复杂信息条件下矿石品位推断的神经网络方法[J].西部探矿工程,1996,8(1)：28-32.
[126]王淑红.基于神经网络的选矿厂预测模型及应用研究D],昆明：昆明理工大学,2002.
[127]邓衍义.Matlab在露天矿品位参数优化中的应用[J].江西理工大学学报,2008,29(3)：13-16.