摘要
本文利用神经网络计算方法，建立了一个具有自适应、复杂非线性的储层预测模型，对储层的物性参数（孔隙度、渗透率）进行预测和储层特性的识别，并对其预测精度进行检验，获得了明显优于常规解释方法的地质效果。
By the computation methods of the neural network in this paper, a predictive model for the layer with complicated nonlinearities is set up, the reservoir characters （such as porosity, permeability） are forecasted and organized, and the precision has been tested. The geographic efficiency obviously superior to traditional explanation methods obtained.
出处
《工程数学学报》
CSCD
北大核心
2005年第8期25-28,共4页
Chinese Journal of Engineering Mathematics
基金
国家自然科学基金（40274019）
关键词
神经网络
孔隙度
渗透率
岩性识别
neural networks
porosity
permeability
lithology recognition
