公开(公告)号：US20230384470A1

公开(公告)日：2023-11-30

申请号：US18031693

申请日：2021-10-15

Applicant: SHANDONG UNIVERSITY ,  SHANDONG HI-SPEED GROUP CO., LTD.

Inventor： Peng JIANG ,  Yuxiao REN ,  Qifeng WANG ,  Zhiwu ZUO ,  Xinji XU ,  Kai WANG ,  Lei CHEN ,  Chuanyi MA ,  Shuai CAO ,  Senlin YANG ,  Qingyang WANG ,  Xianglong MENG

IPC: G01V1/28

CPC classification number: G01V1/282 ,  G01V2210/642 ,  G01V2210/66 ,  G01V2210/6222

Abstract: A method for three-dimensional velocity geological modeling with structures and velocities randomly arranged, including determining base points in three-dimensional space, building equation according to the base points to determine planar layered model, complicating a tilt layer of planar layered model, and building a fold layer model of a surface in three-dimensional space; building three-dimensional fault folded model based on the three-dimensional surface fold layer model combined with a fault plane of a random reference point and displacement of each point in a global coordinate system; building a velocity model containing a salt body based on the three-dimensional fault folded model, and simulating salt body intrusion in a geological body of a certain depth; and performing a random velocity amplitude to realize three-dimensional velocity modeling according to the layered type which has been set and according to the set velocity range and the velocity difference range between each layer of geology.

公开(公告)号：US20230305177A1

公开(公告)日：2023-09-28

申请号：US18031289

申请日：2021-12-14

Applicant: SHANDONG UNIVERSITY

Inventor： Bin LIU ,  Yuxiao REN ,  Peng JIANG ,  Senlin YANG ,  Qingyang WANG ,  Xinji XU ,  Duo LI

IPC: G01V1/30

CPC classification number: G01V1/303 ,  G01V2210/6222 ,  G01V2210/66

Abstract: A multi-scale unsupervised seismic velocity inversion method based on an autoencoder for observation data. Large-scale information is extracted by the autoencoder, which is used for guiding an inversion network to complete the recovery of different-scale features in a velocity model, thereby reducing the non-linearity degree of inversion. A trained encoder part is embedded into the network to complete the extraction of seismic observation data information at the front end, so it can better analyze the information contained in seismic data, the mapping relationship between the data and velocity model is established better, then the inversion method is unsupervised, and location codes are added to the observation data to assist the network in perceiving the layout form of an observation system, which facilitates practical engineering application. Thus a relatively accurate inversion result of the seismic velocity model when no real geological model serves as a network training label can be achieved.