利用变参数回归方法反演泥质岩裂缝密度

本文利用变参数回归分析建立了多口井的泥质岩储层裂缝密度模型。该模型的平均相对误差为１３．５％，较常系数回归分析、ＢＰ网络模型进行裂缝预测的精度高（平均相对误差分别为３８．７％、１７．９％）。通过实际资料处理认为该油田在纵向上随时代变老，深度加深，裂缝密度降低。即从Ｎ２２、Ｎ１２到Ｎ１，平均裂缝密度从０．７８条／ｍ、０．５条／ｍ降低到０．３条／ｍ。在平面上，沿构造轴部裂缝最发育，平均裂缝密度Ｎ２２、Ｎ１２、Ｎ１层分别为０．５８条／ｍ、０．６条／ｍ、０．３条／ｍ。

The fracture density model for pelite reservoirs based on the regression analysis of variable parameters

The fracture density model for the pelite reservoirs in several wells in the Gaojiaya oil field,Xinjiang is presented by means of the regression analysis of variable parameters.The mean relative errors of the model only reach 13.5%.So its precision is higher than those of the regression analysis of constant coefficients and BP network model.The mean relative errors of''the latter two are 38.7% and 17.9%,respectively.The data available have re-vealed that vertically the fracture densities decrease from 0.78 through 0.5 to 0.3 per me-tre as a function of depth and age(from late and early Middle Piiocene to Miocene),and horizonally the fractures tend to be most developed along the structural axis.The mean fracture densities range from 0.58 to 0.6 and to 0.3 per metre,respectively in the upper and lower Plioeene and Miocene strata of the oil field.