贾晓峰
开通时间:..
最后更新时间:..
点击次数:
摘要:Lattice spring model (LSM) provides an alternative numerical approach for simulating seismic wave propagation in heterogeneous media. This method has gained great popularity in fractured media due to its intuitive physical representation. Originating from the discrete element method, the LSM allows particles to achieve micromechanical interactions through springs rather than directly solving the differential equation. The most important issue in the LSM is calibrating the spring coefficients, which can be derived through experiments or physical principles. By simply removing the springs that exceed their strength, the LSM can easily simulate the entire failure process of materials, a task that is challenging for continuum-based methods such as the finite difference method (FDM) and finite element method. In this paper, we propose a new LSM for seismic wave simulation in heterogeneous anisotropic media, which yields more accurate results compared to the regular particle-based methods. Unlike the conventional LSM, which calibrates spring coefficients using the wave equation with an implicit homogeneous approximation, our new LSM calibrates the coefficients using a modified wave equation in heterogeneous media. Compared with the conventional LSM, whose spring coef ficients onl y contain the elasticity tensor itself, our ne w LSM additionall y takes the first deri v ati ve terms of the elasticity tensor into account, and thus can accurately handle the scattering waves in seismic wave simulation. We investigate the spring coefficients of the two LSMs and derive the numerical dispersion and stability condition. To validate the accuracy of the new LSM, we test several scattering, layered and complex heterogeneous anisotropic models, respecti vel y, comparing their results with those obtained using the high-accuracy FDM. Numerical experiments demonstrate the high quality of the new LSM in complex media compared with the conventional LSM. Finally, two fracture models are simulated to illustrate the new LSM’s capability in modelling the complex failure process.
论文类型:期刊论文
是否译文:否
上一条:Sanfu Li, Yaxing Li(*), Yunzhi Shi, Xinming Wu and Xiaofeng Jia, 2025, Deep learning for seismic imaging in the presence of velocity errors, IEEE Geoscience and Remote Sensing Letters, 22: 3000805
下一条:Zi'ang Li, Xiaofeng Jia(*), Jie Zhang, Peimin Zhu, Wei Cai, Hao Zhang and Zhiying Liao, 2024, Characterizing Near-surface Velocity Structures via Deep Learning, IEEE Transactions on Geoscience and Remote Sensing, 62: 5927913.