论文部分内容阅读
为了提高电法测井反演的速度,提出了一种快速计算雅可比矩阵的方法,即只考虑每个待反演参数的微小改变对其所在层及其邻层视电阻率有影响,而对其他层的视电阻率没有影响,从而节约了大量计算雅可比矩阵的时间,其反演的时间与待反演的参数个数基本上呈线性关系。在计算精度基本不受影响的情况下,使得应用阻尼最小二乘方法反演侵入半径和原状地层真电阻率的速度提高了至少一个数量级。人工模型和实际资料的处理结果表明,该方法能有效地消除环境因素的影响,恢复地层的实际电阻率。同时也证实了阵列感应测井数据具有探测深度大、分辨率高的特点,在层不太薄(层厚超过1 m)、侵入不太深(侵入半径不超过0.25 m)时,深探测数据能较准确地反映地层真电阻率。
In order to improve the speed of electrical logging inversion, a fast method of calculating Jacobian matrix is proposed, in which only small changes of parameters to be inverted affect the apparent resistivity of the layer and its adjacent layers Has no influence on the apparent resistivity of the other layers, thus saving a great deal of time for calculating the Jacobian matrix, and the inversion time basically has a linear relationship with the number of parameters to be inverted. With the computational accuracy largely unaffected, the application of the damped least squares method to invert the true radius of resistivity and the original resistivity is increased by at least an order of magnitude. The results of artificial model and actual data show that the method can effectively eliminate the influence of environmental factors and restore the actual resistivity of the formation. At the same time, it is also confirmed that the array induction logging data has the characteristics of large exploration depth and high resolution. When the layer is not too thin (over 1 m in thickness) and the intrusion is not too deep (the intrusion radius does not exceed 0.25 m), the deep exploration data Can more accurately reflect the real formation resistivity.