弹性波被动源数据一次波估计方法研究

与声波被动源数据相比，弹性波被动源数据含有更加丰富的地下地质信息.弹性波被动源数据合成的虚拟炮记录中，同样存在表面相关多次波，影响数据的后续处理以及对地下地质构造的准确判断.对弹性波被动源数据合成的虚拟炮记录进行多次波的预测与匹配相减，将比声波数据更加困难.本文提出弹性波被动源数据一次波估计方法，针对各向同性介质条件下的弹性波被动源数据进行研究，依据声波被动源数据一次波估计方法的研究思路，推导弹性波被动源数据一次波估计方法的理论过程，充分考虑弹性波被动源数据所包含的波场成分，包括纵波、横波以及在传播过程中产生的转换波.在传播过程中，由于能量的损失，略去多次传播的高阶项，进而给出相应的目标函数.利用数据的稀疏性，结合L₁范数约束的最优化问题求解方法，避免对弹性波虚拟炮记录进行多次波的预测与匹配相减，解决多次波影响问题.本文通过合成的弹性波被动源数据，进行一次波估计的测试，并与主动源和合成的虚拟炮记录进行对比，验证本文提出方法的有效性与准确性.

Study of estimating primaries in elastic passive seismic data

Comparing with acoustic passive-seismic data, elastic waves from a passive source contains more abundant information of underground geological structure. The virtual-shot gathers reconstructed from elastic passive data also include multiple waves, which seriously influence the subsequent data processing and accurate determination of geological structure in the subsurface. Thus the multiple-wave prediction and matching subtraction for the virtual-shot gathers of elastic passive data would be more difficult than acoustic one. To solve this problem, this paper proposes a method of estimating primaries in elastic passive seismic data based on the study of the elastic passive seismic data in an isotropic medium. In light of the derived process of acoustic passive data, we derive the theoretical process of estimating primaries in elastic passive seismic data, which adequately takes into account the wave-field components, including longitudinal waves, shear waves, and converted waves. Due to the loss of energy during the propagation, the high-order items of multiple reflections can be omitted. Then the corresponding objective function is obtained. We use the sparsity of data, and combine with the solving method of the optimization problem by the L₁-norm constraint, which avoids multiple-wave prediction and matching subtraction for the virtual-shot gathers of elastic passive data, and solves the problem of the multiple-wave influence. Tests on the synthetic elastic passive data and comparing with control-shot gathers and virtual-shot gathers demonstrate the effectiveness and accuracy of the proposed method.