多元统计储后预测
|
摘要
以地震信息为主进行储层预测是常现有效的方法。然而，当地震资料表现为低分辨率、低保真度和低估噪比时，几乎所有的地震属性多数（振幅、频率等）都失去了自身的地质意义。在这种情况下，若工区内具有较为富集的钻井，则可利用统计学理论，从工区所有并同层段的各种测井曲线数据中分别提取均值、标准差、偏斜度和峭度。这些统计特征量表征了不同的储层特性，将同一种统计特征在平面上展开，可直观地反映出相应储层特性的横向变化。为了准确地描绘储层的发育程度，本文应用模糊数学方法，综合多种测井、钻井和孔隙率等资料，概括提炼出一个综合判别标准，来确定储层的发育程度和储集能力；然后，结合沉积相和地震信息，对储层作多参数、多视角的立体描述，从而指出含油气的有利地区。应用实例表明，预测结果与钻井反馈信息的符合率很高，说明此方法是可行的。
Reservoir prediction based mainly on seismic informations is a general effective method. Nearly all the seismic attribute paratneters (amplitude, frequency, etc. )lose their geological significance when seismic data are characterized by the low resolution, low fidelity and low S/N ratio. When there are dense drilling in an area ,some characteristic data such as mean value, standard difference, divergence and steepness can be statistically derived from the same interval of various log curves.These statistical characteristic values indicate different reservoir properties respec tively, and the plane distribution of one kind of characteristic values exhibits visual lateral variation of the corresponding reservoir property.The volurne and reservoiring ability of a reservoir can be determined by using a comprehensive discretion criterion that is derived frorn various log data, drilling da ta and porosity data with the use of fuzzy mathematical method. Then 3D overall description of a reservoir is conducted by analysing the sedimentary facies, various seismic informations, Volume and reservoiring ability, so that favourable hydrocarbon areacan be mapped out reasonably. It is practically shown that this method is feasible because the prediction result coincides with the drilling result very well.
Reservoir prediction based mainly on seismic informations is a general effective method. Nearly all the seismic attribute paratneters (amplitude, frequency, etc. )lose their geological significance when seismic data are characterized by the low resolution, low fidelity and low S/N ratio. When there are dense drilling in an area ,some characteristic data such as mean value, standard difference, divergence and steepness can be statistically derived from the same interval of various log curves.These statistical characteristic values indicate different reservoir properties respec tively, and the plane distribution of one kind of characteristic values exhibits visual lateral variation of the corresponding reservoir property.The volurne and reservoiring ability of a reservoir can be determined by using a comprehensive discretion criterion that is derived frorn various log data, drilling da ta and porosity data with the use of fuzzy mathematical method. Then 3D overall description of a reservoir is conducted by analysing the sedimentary facies, various seismic informations, Volume and reservoiring ability, so that favourable hydrocarbon areacan be mapped out reasonably. It is practically shown that this method is feasible because the prediction result coincides with the drilling result very well.
引文
1普雷斯 W H等数值方法大全,兰州大学出版社,1991
版权所有：© 2023 中国地质图书馆 中国地质调查局地学文献中心