基于主成分分析的BP神经网络在岩性识别中的应用

提出一种将主成分分析和BP神经网络相结合的方法对测井资料进行岩性识别。首先将原始测井数据进行主成分分析，分析结果作为PCABP神经网络的学习样本进行训练，建立测井解释的PCA—BP神经网络岩性识别模型．并用该模型对测试样本进行识别。结果表明该方法同传统的BP神经网络相比．不仅简化了网络结构（网络的输入神经元个数由5个减少为2个），网络收敛速度也加快了21％．而且识别的准确率提高了25％。

Application of Principal Component Analysis and BP Neural Network in Identifying Lithology

A lithology identification method based on principal component analysis （PCA） and back propagation neural network is presented. First, using the learning samples, which are obtained after the PCA of original well-logging data, to train the PCA-BP neural network and establishing the PCA-BP neural network model of lithology identification, then using the model to forecast the lithology of unknown samples. Compared with common BP neural network model in lithology identification, the results indicate that the PCA-BP neural network model could not only predigest the network structure （the number of input neurons reduces from five to two） and accelerate the network convergence speed by 21 percent, but also increase the precision of recognition by 25 percent.