摘要
针对只有孔隙度、渗透率等基础测试数据的井难以求取其分形维数的问题,提出一种基于粒子群算法的致密砂岩储层分形维数求解方法。首先对于拥有高压压汞、核磁资料的井,利用汞饱和度法、核磁共振法计算其分形维数值,并将其设置为训练样本;其次对于只有孔渗数据的井,基于粒子群算法(PSO)思想将目标函数定义为孔、渗数据和训练样本数据的偏差平方和,使用8个岩心样品的孔渗数据进行分形维数计算;最后将计算结果与汞饱和度法和核磁共振法计算结果进行对比分析。结果表明,使用粒子群算法求得的分形维数具有较高精度,该计算过程具有高自动化性、低主观性,该方法可在分形维数参数求解计算中推广应用。
In the cause of solving the problem which is knotty to obtain the fractal-dimension parameter of wells with only basic test data such as porosity and permeability,an approach based on a particle swarm optimization algorithm(PSO)is planned to resolve the fractal dimension of tight sandstone reservoir’s pore structure.Firstly,for wells with nuclear magnetic and high pressure mercury injection data,the fractal dimension values were calculated by nuclear magnetic resonance method and mercury saturation method,and were set as training samples.Secondly,for wells with only porosity and permeability data,the objective representation was defined as the sum of the squares of the deviations of porosity and permeability data and training sample data based on the PSO idea,and the porosity and permeability data of 8 core samples were used to calculate the fractal dimension.Finally,the calculated results were compared with those of mercury saturation method and nuclear magnetic resonance method.The results show that the fractal dimension obtained by particle swarm optimization algorithm has a high precision,the calculation process has high automation and low subjectivity,and it can be well applied in the calculation of fractal dimension parameters.
作者
孙和美
邢雪杰
刘琦
邓琳
张正朝
杨宏涛
嵇雯
SUN Hemei;XING Xuejie;LIU Qi;DENG Lin;ZHANG Zhengchao;YANG Hongtao;JI wen(College of the Geoscience and Engineering,Xi’an Shiyou University,Xi’an 710065,Shaanxi,China;Shaanxi Key Laboratory of Petroleum Accumulation Geology,Xi’an 710065,Shaanxi,China;PetroChina Coalbed Methane Company Limited,Xi’an 715400,Shaanxi,China)
出处
《石油地质与工程》
CAS
2024年第3期61-69,共9页
Petroleum Geology and Engineering
关键词
分形维数
粒子群算法
高压压汞
核磁共振
致密储层
孔渗数据
fractal dimension
particle swarm optimization algorithm
high-pressure mercury injection
nuclear magnetic resonance
tight sandstone reservoir
porosity and permeability data
