摘要
由于矿床定位影响因子较多,并且这些因子与矿床位置间存在着非常复杂的非线性关系,所以一直以来,矿床定位只能实现定性化预测。RBF神经网络可以实现从输入到输出的高度非线性映射,能够使矿床预测由定性预测发展为定量预测。这里应用RBF神经网络法对浙江某地区铅锌矿进行了综合物化探找矿预测,结果表明,该方法具有很高的准确率,与实际地质情况吻合率很高,可以定量用作对隐伏矿床的预测。
Many factors affect mineral deposit location, there is a very complicated nonlinear relationship that has been achieved only qualitative predictions between these factors and the deposit location. RBF neural network can achieve a high degree nonlinear mapping from input to output.In this paper,RBF neural network is used to predict lead-zinc mineral deposit in synthetic geophysical and geochemical exploration prediction in an area of Zhejiang province.The results show that the method has high accuracy,consistent with the high rate of the actual geological conditions,so that can be used as the prediction method of the concealed deposit.
出处
《物探化探计算技术》
CAS
CSCD
2012年第2期182-185,8,共4页
Computing Techniques For Geophysical and Geochemical Exploration
基金
中央地质勘查项目(2009330001)
关键词
RBF神经网络
综合物化探找矿预测
矿床定位
RBF neural network
synthetic geophysical and geochemical exploration prediction
mineral deposit location