论文部分内容阅读
本文用某地区46口井的实测孔隙度、纵横波速度资料，建立了该地区孔隙度与纵横波速度关系的线性公式。由此，仅用纵横波速度便可预测孔隙度而不需要任何其它参数，并且精度明显高于wyllie公式的预测精度；但此线性公式预测的孔隙度有平均化现象。以46口井实测资料供神经网络学习，用神经网络作函数逼近，由此建立了该地区孔隙度与纵横波速度关系的非线性公式，非线性公式明显地更逼近于实测数据，提高了精度、并消除了孔隙度预测值的平均化现象。
In this paper, using the measured data of 46 wells in a certain area from the measured porosity, P- and S-wave velocities, a linear formula for the relationship between porosity and P- and P- Thus, porosity can be predicted using only P- and S-wave velocities without any other parameters, and the accuracy is significantly higher than predicted by the Wyllie formula; however, the porosity predicted by this linear equation is averaged. Based on the measured data of 46 wells for neural network learning and neural network as a function approximation, a nonlinear formula was established for the relationship between porosity and P- and S-wave velocities in this area. The nonlinear formula was significantly closer to the measured data and the accuracy was improved , And eliminated the averaging of porosity predictions.