节点文献
阻尼LSQR算法在多波地震反演中的应用
The Application of Damped LSQR Algorithm in the Multi-wave Seismic Inversion
【摘要】 多波VTI介质AVA储层参数联合反演的关键是解Jacobi矩阵。本文从Lanczos迭代出发,结合双对角化和QR分解,推导出了阻尼Lanczos算法流程。通过比较SVD算法和LSQR算法,显示了阻尼Lanczos算法在效率和精度方面的优越性。该算法应用于多波地震勘探资料反演中,获得的目标层储层参数:纵波速度、横波速度和密度,证明了该算法的有效性。
【Abstract】 The key of the joint inversion of reservoir parameters for the Multi-wave VTI media AVA is the solution of the Jacobi matrix.This paper based on the Lanczos iteration,deduced Lanczos algorithm flow damping by combining double diagonalization and QR decomposition.Compared with SVD algorithm and LSQR algorithm,it was showed that the damped Lanczos algorithm was superior in its efficiency and accuracy.Then,we applied this algorithm to the Multi-wave seismic data inversion to prove the effectiveness of the algorithm.
- 【文献出处】 中国西部科技 ,Science and Technology of West China , 编辑部邮箱 ,2012年01期
- 【分类号】P631.44
- 【被引频次】1
- 【下载频次】126