基于改进PSO-SVM的钻井液侵入储层深度预测

开展钻井液侵入储层深度预测,对于测井评价以及提高油井产能具有一定的现实意义。在分析钻井液侵入储层的机理和特征的基础上,提出了钻井液侵入储层的影响因素指标体系,建立了改进PSO-SVM的钻井液侵入储层深度预测模型,以塔里木塔中35口井为例进行了实证分析,并与传统BP神经网络和SVM模型预测结果进行了对比。研究结果表明:侵入深度与泥饼的渗透率、钻井液与储层压差以及侵入时间正相关,与储层孔隙度和钻井液粘度负相关,改进的PSO-SVM模型预测结果误差小,准确率高,能够用于钻井液侵入储层深度预测,具有广泛的应用前景。

PREDICTION OF DRILLING FLUID DAMAGE DEPTH BASED ON IMPROVED PSO-SVM

It is of practical significance for logging evaluation and improving oil well productivity to carry out prediction of drilling fluid invasion depth.Based on the mechanism and characteristics of drilling fluid invasion into reservoir,index system of influencing factors of drilling fluid invasion into reservoir are put forward in this paper,the prediction model of drilling fluid polluted reservoir depth is established by using improved PSO-SVM.Taking 35 wells in Tazhong,Tarim as examples,this paper makes an empirical analysis,and compares the evaluation results with the traditional BP neural network and SVM model.The results show that the invasion depth is positively related to the permeability of mud cake,the pressure difference between drilling fluid and formation and the invasion time,negatively related to the formation porosity and the viscosity of drilling fluid.The improved PSO-SVM model has small error and high accuracy,which can be used to predict the depth of reservoir polluted by drilling fluid and has a wide application prospect.