摘要
对测井资料与地层可钻性级值的关系进行了分析,提出了一种基于粒子群优化支持向量机算法预测地层可钻性级值的新方法,利用测井声波时差、地层密度、泥质质量分数和地层深度进行学习训练支持向量机,并利用粒子群优化算法对支持向量机(PSO-SVM)参数进行优化,建立了预测地层可钻性级值的支持向量机模型。应用该方法对准噶尔盆地庄2井的地层可钻性级值进行了预测,并将该方法的预测结果与BP神经网络方法的预测结果进行了比较。结果表明,该方法优于BP神经网络方法,具有预测精度高、收敛速度快、推广能力强等优点。
After the relationship between well-log data and formation drillability was analyzed,a novel method for predicting formation drillability based on particle swarm optimization and support vector machine （PSO-SVM） was proposed. A prediction model for formation drillability was established using the data of well-log acoustic velocity, formation density, fraction of shale and formation depth by training the SVM optimized by PSO algorithm. The proposed method was applied to Zhuang 2 Well in Junggar Basin. The experimental results show that it has higher prediction precision, faster convergence speed and better generalization effect than BP neural network.
出处
《石油学报》
EI
CAS
CSCD
北大核心
2008年第5期761-765,共5页
Acta Petrolei Sinica
基金
中国石化科技攻关项目(JP04014)“基于钻井工程地质数据库的钻井模拟”
中国石化科技攻关项目(JP03009)“地质导向钻井工艺技术研究”联合资助
关键词
地层可钻性
粒子群优化算法
支持向量机
测井资料
参数优化
预测模型
formation drillability
particle swarm optimization algorithm
support vector machine
well-log data
parameter optimization
prediction model
