径向基函数神经网络法预测海上油田储层敏感性

Prediction of Reservoir Sensitivity In Offshore Oil Field by Radial Basis Function Neural Network

在石油勘探开发过程中,引入不配伍工作液会引起储层敏感性损害。目前预测储层敏感性常参考行业标准通过实验评价,但实验评价周期长,需消耗大量人力物力,特别是对于探井和缺乏岩心样品的井,如何快速、准确地预测储层敏感性是亟需解决的问题。径向基函数神经网络法作为一种新兴预测方法,具有网络结构简单、逼近能力强和学习速度快等优点,已成为陆地油田的储层敏感性智能预测方法之一。文章基于径向基函数神经网络研究储层敏感性,实现了对海上油田储层速敏、水敏、酸敏、碱敏损害的定量诊断,且精度高,可操作性强,可推广应用。

In the process of oil exploration and development,incompatibility of working fluid will cause damage to reservoir sensitivity.At present,the prediction of reservoir sensitivity often refers to the industry standard through the experimental evaluation,but the experimental evaluation cycle is long,which needs to consume a lot of manpower and material resources,especially for exploration wells and wells without core samples,how to predict reservoir sensitivity quickly and accurately is an urgent problem to be solved.As a new prediction method,radial basis function neural network has the advantages of simple network structure,strong approximation ability and fast learning speed,which has become one of the intelligent prediction methods of reservoir sensitivity in land oilf ield.Based on the radial basis function neural network,this paper studies the reservoir sensitivity,and realizes the quantitative diagnosis of the damage of the reservoir speed sensitivity,water sensitivity,acid sensitivity and alkali sensitivity in the offshore oil f ield.