摘要
储层流体识别是油气田开发的前提。针对本地区的储层流体识别,提出基于多层感知器网络模型的方法。在对储层流体常规分析中根据储层流体的测井响应,优选出储层流体的敏感测井参数,利用多层感知器的自适应和自学习功能,结合BP神经网络误差逆传播的原理,建立储层流体识别模型,对比试气结果,验证模型的准确性,同时为其他地区低渗储层流体识别问题提供借鉴。
With the development of oil and gas fields,reservoir fluid identification is the premise of development.Aiming at the reservoir fluid identification in this area,a method based on multilayer perceptron network model is proposed.In the conventional analysis of reservoir fluid,according to the logging response of reservoir fluid,the sensitive logging parameters of reservoir fluid are optimized.By using the self-adaptive and self-learning function of multi-layer perceptron,combined with the principle of BP neural network error back propagation,the reservoir fluid identification model is established,and the accuracy of the model is verified by comparing the gas test results.At the same time,it provides a reference for low permeability reservoir fluid identification in other areas It can be used for reference.
作者
张涛
赵军
韩东
王伟明
Zhang Tao;Zhao Jun;Han Dong;Wang Wei-ming
出处
《化工设计通讯》
CAS
2021年第4期178-179,共2页
Chemical Engineering Design Communications
关键词
流体识别
多层感知器
BP神经网络
逆传播
fluid recognition
multilayer perceptron
BP neural network
back propagation
