欢迎访问《金属矿山》杂志官方网站,今天是 分享到:

金属矿山 ›› 2009, Vol. 39 ›› Issue (07): 66-68.

• 地质与测量 • 上一篇    下一篇

基于BP人工神经网络的成矿预测模型

毛政利1,闫继涛1,赖健清2   

  1. 1.河南城建学院;2.中南大学
  • 出版日期:2009-07-15 发布日期:2011-04-26

ANN-Based Metallogenic Prognosis Model

Mao Zhengli1,Yan Jitao1,Lai Jianqing2   

  1. 1.Henan University of Urban Construction;2.Central South University
  • Online:2009-07-15 Published:2011-04-26

摘要: 在传统的成矿预测研究中,一般只能定性地研究矿床的形成与定位受哪些地质因素的控制,但这些因素对成矿的贡献程度则很难给出一个定量的值,而BP人工神经网络的高度非线性映射功能则能很好的拟合成矿作用过程这样的高度非线性耦合关系,因此,基于BP人工神经网络的成矿预测模型相对于传统模型具有更高的预测精度。

关键词: 成矿预测模型, 成矿作用, 非线性耦合关系, BP人工神经网络

Abstract: Conventional Metallogenic prognosis is only capable of a qualitative research of deposit formation and controlling geological factors but fails to give quantitative values of their contribution degrees to the metallogenesis.BP ANN's highly nonlinear reflection function can well fit the highly nonlinear coupling relation involved in the metallogenesis process.Therefore, the Ann-based metallogenic prognosis model has higher prognosis accuracy than conventional models

Key words: Metallogenic prognosis model, Metallogenesis function, Nonlinear coupling relation, BP ANN