摘要
针对传统变差函数拟合方法的不足,采用改进算法对变差函数进行拟合,进而用于储层非均质性预测。利用柯西变异能快速跳出局部极小值的优点,采用线性变化的收缩扩张因子对量子粒子群算法进行改进,将改进的算法应用于变差函数球状模型的拟合,最后利用该方法对某储层平面非均质性进行分析。实验结果表明,改进算法实现了参数的自动拟合,拟合的变差函数较传统的加权最小二乘拟合方法具有更高的精度,较好地反映了不同方向上储层平面非均质性的差异。基于改进量子粒子群算法拟合的变差函数可以应用于储层平面非均质性的预测。
The improved algorithm is used for the fitting of the variation function in order to predict the reservoir heterogeneity and overcome the shortcomings of the traditional variation function fitting method. The advantage of quickly jumping out of the local minimum by Cauchy mutation is used to improve the quantum swarm optimization (PSO) algorithm by using the linear expansion factor, and the improved algorithm is applied to the fitting of the spherical model of the variance function and to the analysis of the plane heterogeneity of a reservoir. The experimental results show that the improved algorithm achieves the automatic fitting of the parameters, and the fitted variable function has higher accuracy than the traditional weighted least squares fitting method, and it well reflects the difference of the heterogeneity of reservoirs in different directions. Therefore, the variation function based on improved quantum particle swarm optimization can be applied to the prediction of reservoir plane heterogeneity.
作者
谢源
李琼
陈杰
何荣胜
XIE Yuan;LI Qiong;CHEN Jie;HE Rongsheng(College of Geophysics,Chengdu Unieersity of Technology,Chengdu 610059,China)
出处
《成都理工大学学报(自然科学版)》
CAS
CSCD
北大核心
2018年第3期379-385,共7页
Journal of Chengdu University of Technology: Science & Technology Edition
基金
国家自然科学基金项目(41274129)
国家科技重大专项(2011ZX05035-005-003HZ)
关键词
改进量子粒子群算法
变差函数
球状模型
自动拟合
improved quantum particle swarm optimization
variogram
spherical function model
automatic fitting