论文部分内容阅读
本文的理论方法是以几何射线理论为基础发展起来的、天然地震走时反演技术及天然地震层析成像技术。它存在着震源函数与介质参数的解耦问题。本研究采取了五种方法来改善反演结果。包括，利用，Pg，Pn等震相增大约束条件；用已有精度较高的人工地震测深结果作速度约柬：用波形反演来修改模型，把诸多物理量开发出来互为约束，以修改后的模型再作反演，使解的稳定性大大提高：采用最优化过程，选择遗传算法。可以进行震源定位，走时反演，波形反演：得到任意深度的速度分布及从地表到Moho面的速度剖面。用于在祁连山地区的结果表明，这些层析剖面对认识大地构造、重大深部事件动力学是很有益的。
The theoretical method of this paper is based on the theory of geometrical rays developed by natural seismic traveltime inversion technology and natural seismic tomography. It exists the decoupling problem between the source function and the medium parameter. In this study, we adopted five methods to improve the inversion results. Including the use of, Pg, Pn and other phases to increase the constraint conditions; with the existing high accuracy of artificial seismic sounding results for speed constraints: waveform inversion to modify the model, the development of a large number of physical constraints each other to The modified model is then inverted and the stability of the solution is greatly improved: the optimization process is adopted and the genetic algorithm is selected. Source positioning, walk-on inversion, waveform inversion can be performed: Velocity distribution at any depth and velocity profiles from the surface to the Moho surface can be obtained. The results used in the Qilian Mountains indicate that these tomographic profiles are beneficial for understanding the tectonics and major deep event dynamics.