神经网络技术在碳氧比能谱测井中的应用

APPLICATION OF NEURAL NETWORK TECHNIQUE TO THE CARBON/OXYGEN SPECTRAL LOG

为了充分利用碳氧比能谱测井中的各种信息,以BP神经网络基本原理为理论基础,发挥神经网络自动识别的优势,研究并建立了预测碳氧比含油饱和度的非线性神经网络模型,通过对实际取心井资料的应用,说明该方法是一种有效的方法。

In order to make best use of various information of carbon/oxygen spectral log,the authors take the basic principles of BP neural network as the theoretical basis,bring the automatic identification of the neural network into full play,and establish a nonlinear neural network model for predicting carbon/oxygen oil saturation.Proved by the application of data from actual sampling,this log technique is an effective method.