论文部分内容阅读
常规的基于贝叶斯理论的稀疏脉冲反演中,各约束项的拉格朗日算子均采用的是恒定的常系数。反演实际资料发现,波阻抗反演剖面与钻井资料的油气显示并不能很好地对应。考虑到反演不同地震道数据时,低频趋势模型所起的约束作用应当不同。本文在常规反演的基础上做出了改进,假定阻抗约束系数是一个空间变量,由各地震道的实际地震数据与合成记录之间的振幅残差来确定该道的阻抗约束系数。实际资料应用表明,改进后的反演结果更稳定,能更准确地反映地下阻抗信息。
In conventional sparse pulse inversion based on Bayesian theory, the Lagrange operators of each constraint adopt a constant constant coefficient. The inversion of the actual data shows that the inversion of wave impedance and the oil and gas of drilling data do not correspond well. When considering the retrieval of different seismic trace data, the constraints imposed by the low frequency trend model should be different. In this paper, we make improvements based on the conventional inversion, and assume that the impedance constraint coefficient is a spatial variable. The impedance constraint coefficient of the channel is determined by the amplitude residual between the actual seismic data of each channel and the synthetic record. The application of actual data shows that the improved inversion result is more stable and can reflect the underground impedance information more accurately.