模糊模式识别用于找矿预测

METALLOGENIC PROGNOSIS BY FUZZY PATTERN RECOGNITION

本文提出两类模糊模式识别用于找矿预测的具体方法．用Ｆ－ＰＦＳ法和调节特征因素及其权重以获取最佳分类，聚类中心即作为标准模式．根据单因素判对率确立了因素逆距离权重的概念．在标准模式的模糊向量与已知单元模糊向量之间关系的基础上可以建二线性不等式方程组，从而可解不同因素的距离权重，并进而用贴近度对未知单元进行识别．以上方法应用于鄂东南地区的铜及多金属的找矿预测，结果表明方法有效。成果较好。

On the basis of fuzzy pattern recognition,some mathematical methods for metallogenic prognosis are proposed. An optimum classification may be obtained through the adjustment of the characteristic factors and weights and by using the F-PFS method. The cluster centre is considered as a standard pattern. According to the rate of right judgement of a single factor,we present an idea about the weight of complement distance of the factor. On the basis of the relationship between the fuzzy vector of the standard pattern and that of the known cells, a system of linear algebraic inequalities can be established, hence the distance weights of different factors are solved. Thus the recognition of the unknown cells is achieved by proximity degree.These methods have been applied to the prognolication of copper and multimetal ores in southeastern Hubei Province, China, and the results show that the methods are effective.