人工神经网络在多金属结核研究中的应用

本文通过对太平洋CC区多金属结核成矿背景的综合分析,利用人工神经网络随机BP算法,建立能够反映控矿因素的地质因子与结核分布参数间关系的人工神经网络模型,用于研究海底多金属结核的分布。本文对BP算法加以改进,提高了BP网络的收敛速度,加强了BP网络的学习和记忆功能,开发出人工神经网络的软件包,该软件包能够快速估算矿区内每个位置结核的丰度值,且具有较高的容错性。

Application of artificial neural network in study of polymetallic nodules

An artificial neural network model is founded by analysing the background of the Pacific poly- metallic nodules, that is used to study the relationship between geologic factors and distribution of polymetal- lic nodules in the area studied by BP method. In this paper, we advanced random BP method, and developed a software package of artificial neural network that can rapidly calculate abundance of polymetallic nodules in any site of the area studied.