论文部分内容阅读
以岩心分析资料及多种测井信息为依据，首先利用样本信息的神经元模型（ＣＵＳＩ）解决了储层参数的计算问题，并利用改进后的自适应神经元模型（ＡＣＵＳＩ）提高了分析精度。最后利用前馈神经网络的误差反向传播模型（ＢＰ）网络的外延和信息表达能力解决了非储层的定量识别。应用上述方法对辽河油田四口井进行了逐点参数分析，分析结果与实际情况吻合很好。
Based on the core analysis data and various logging information, firstly, the calculation of reservoir parameters is solved by the sample information neuron model (CUSI), and the improved adaptive neuron model (ACUSI) is used to improve the analysis accuracy . Finally, the quantitative identification of non-reservoir is solved by the extension of the error back propagation model (BP) network of the feedforward neural network and the ability of information expression. Applying the above method, the four wells in Liaohe Oilfield were analyzed by point-by-point parameters. The analysis results are in good agreement with the actual situation.