CN-116990861-B - Converted wave multiple prediction method and related device

CN116990861BCN 116990861 BCN116990861 BCN 116990861BCN-116990861-B

Abstract

The invention discloses a converted wave multiple prediction method and a related device, wherein the method comprises the steps of respectively establishing multiple prediction models corresponding to various types of converted wave multiple propagation paths according to various types of converted wave multiple propagation paths, transforming longitudinal wave seismic time domain data and converted wave seismic time domain data to obtain longitudinal wave seismic frequency domain data and converted wave seismic frequency domain data, predicting converted wave multiple frequency domain data according to the longitudinal wave seismic frequency domain data, the converted wave seismic frequency domain data and the multiple prediction models corresponding to various types of converted wave multiple propagation paths, and inversely transforming the converted wave multiple frequency domain data to obtain converted wave multiple time domain data. According to the method, the multiple prediction model is built according to the longitudinal wave seismic frequency domain data, the converted wave seismic frequency domain data and various converted wave multiple propagation paths, and the converted wave multiple of the complex multiple propagation paths is predicted efficiently and accurately.

Inventors

XU QIANG
CHENG YAO
JIAO XUMING
WANG WEI
WANG HAIKUN
SUN LEIMING
MA DEZHI
ZHOU MI
ZHANG MINGQIANG
LI CHUNLEI

Assignees

中海油田服务股份有限公司

Dates

Publication Date: 20260508
Application Date: 20230704

Claims (8)

1. A converted wave multiple prediction method, comprising: establishing a first multiple prediction model according to a pp-ps propagation path, wherein the pp-ps propagation path consists of downlink longitudinal waves, uplink longitudinal waves, downlink longitudinal waves and uplink transverse waves; respectively establishing a second multiple prediction model and a third multiple prediction model according to two ps-ps propagation paths of different types, wherein the ps-ps propagation paths consist of downlink longitudinal waves, uplink transverse waves, downlink longitudinal waves and uplink transverse waves; Transforming the longitudinal wave seismic time domain data and the converted wave seismic time domain data to obtain longitudinal wave seismic frequency domain data and converted wave seismic frequency domain data; Predicting converted wave multiple frequency domain data according to the longitudinal wave seismic frequency domain data, converted wave seismic frequency domain data, a first multiple prediction model, a second multiple prediction model and a third multiple prediction model; Performing inverse transformation on the converted wave multiple frequency domain data to obtain converted wave multiple time domain data; Wherein the first multiple prediction model is: The second multiple prediction model is: the third multiple prediction model is as follows: Wherein, the In order to be of an angular frequency, For the depth delay of the gun focus, In order for the detector point depth to be delayed, Is an imaginary number, and is used for the purpose of calculating, 、、 As a function of the complex variables, Respectively an inter-gun x coordinate, an inter-gun y coordinate, a detector x coordinate and a detector y coordinate, For the summed x-and y-coordinates, As longitudinal wave seismic frequency domain data, Is converted wave seismic frequency domain data.
2. The converted wave multiple prediction method according to claim 1, wherein transforming the longitudinal wave seismic time domain data and the converted wave seismic time domain data to obtain longitudinal wave seismic frequency domain data and converted wave seismic frequency domain data further comprises: performing Fourier transform on the longitudinal wave seismic time domain data and the converted wave seismic time domain data to obtain longitudinal wave seismic frequency domain data P and converted wave seismic frequency domain data R, Wherein, the In order to be of an angular frequency, For the number of samples in the frequency domain, In order to be able to take time, Is an imaginary number, and is used for the purpose of calculating, As a function of the complex variables, In the form of longitudinal wave seismic time domain data, Is converted wave seismic time domain data.
3. The converted wave multiple prediction method according to claim 1, wherein predicting converted wave multiple frequency domain data from the longitudinal wave seismic frequency domain data, converted wave seismic frequency domain data, a first multiple prediction model, a second multiple prediction model, and a third multiple prediction model further comprises: Inputting the longitudinal wave seismic frequency domain data and the converted wave seismic frequency domain data into the first multiple prediction model, the second multiple prediction model and the third multiple prediction model respectively to obtain prediction results of the first multiple prediction model, the second multiple prediction model and the third multiple prediction model; Accumulating the prediction results of the first multiple prediction model, the second multiple prediction model and the third multiple prediction model to obtain total converted wave multiple frequency domain data , 。
4. The method of claim 2, wherein the inverse transforming the converted wave multiple frequency domain data to obtain converted wave multiple time domain data further comprises: and carrying out Fourier inverse transformation on the converted wave multiple frequency domain data to obtain converted wave multiple time domain data.
5. The method for predicting multiple of converted wave according to claim 4, wherein the specific formula for performing inverse fourier transform on the multiple frequency domain data of converted wave to obtain multiple time domain data of converted wave is: Wherein, the For prediction at Converted wave multiple time domain data of time instant, In order to be of an angular frequency, For the number of samples in the frequency domain, In order to be able to take time, Is an imaginary number, and is used for the purpose of calculating, As a function of the complex variables, Is the total converted wave multiple frequency domain data.
6. A converted wave multiple prediction apparatus comprising: The model construction module is used for building a first multiple prediction model according to a pp-ps propagation path, wherein the pp-ps propagation path consists of downlink longitudinal waves, uplink longitudinal waves, downlink longitudinal waves, uplink transverse waves, respectively building a second multiple prediction model and a third multiple prediction model according to two different types of ps-ps propagation paths, wherein the ps-ps propagation path consists of downlink longitudinal waves, uplink transverse waves, and the first multiple prediction model is as follows: The second multiple prediction model is as follows: the third multiple prediction model is as follows: Wherein, the In order to be of an angular frequency, For the depth delay of the gun focus, In order for the detector point depth to be delayed, Is an imaginary number, and is used for the purpose of calculating, 、、 As a function of the complex variables, Respectively an inter-gun x coordinate, an inter-gun y coordinate, a detector x coordinate and a detector y coordinate, For the summed x-and y-coordinates, As longitudinal wave seismic frequency domain data, Seismic frequency domain data for converted waves; The transformation module is used for transforming the longitudinal wave seismic time domain data and the converted wave seismic time domain data to obtain longitudinal wave seismic frequency domain data and converted wave seismic frequency domain data; the prediction module is used for predicting converted wave multiple frequency domain data according to the longitudinal wave seismic frequency domain data, the converted wave seismic frequency domain data, the first multiple prediction model, the second multiple prediction model and the third multiple prediction model; And the inverse transformation module is used for carrying out inverse transformation on the converted wave multiple frequency domain data to obtain converted wave multiple time domain data.
7. A computing device comprising a processor, a memory, a communication interface, and a communication bus, the processor, the memory, and the communication interface completing communication with each other over the communication bus; The memory is configured to store at least one executable instruction, where the executable instruction causes the processor to perform the operations corresponding to the converted wave multiple prediction method according to any one of claims 1 to 5.
8. A computer storage medium having stored therein at least one executable instruction for causing a processor to perform operations corresponding to the converted wave multiple prediction method of any one of claims 1-5.

Description

Converted wave multiple prediction method and related device Technical Field The invention relates to the field of marine petroleum geophysical exploration, in particular to a converted wave multiple prediction method and device, computing equipment and a computer storage medium. Background In the seismic exploration of reflected waves, the seismic waves are reflected at the underground interface or the sea surface more than once, and finally the multiple reflected waves, namely multiple waves, are received by a wave detector at the ground or the sea surface, and the multiple waves can travel back and forth between strong reflection interfaces, so that the propagation path and the amplitude change are more complex. The developing multiples can severely reduce the accuracy of seismic data processing, thereby misleading interpreters to give erroneous interpretation results. Multiple prediction processing has long been the focus and difficulty of seismic data processing. For the converted wave, the multiple propagation paths are more complex (the multiple propagation paths of the converted wave mainly comprise two types, namely downlink longitudinal wave, uplink longitudinal wave, downlink longitudinal wave, uplink transverse wave, downlink longitudinal wave, uplink transverse wave, and the like), and the prediction of the multiple is more difficult due to severe multiple interference, so that the imaging quality of the converted wave seismic data is seriously affected. The existing multiple wave suppression method mostly assumes that all scattering sources only reflect once at the underground interface, and often cannot effectively suppress multiple waves of complex multiple propagation paths, or limits the application of the method in actual data due to large calculated amount. Therefore, there is a need to efficiently and accurately predict converted wave multiples and remove them from seismic data to avoid misunderstanding in the seismic data. Disclosure of Invention The present invention has been made in view of the above-described problems, and has as its object to provide a converted wave multiple prediction method and apparatus, a computing device, and a computer storage medium capable of efficiently and accurately predicting converted wave multiple. According to an aspect of the present invention, there is provided a converted wave multiple prediction method including: According to the multiple propagation paths of the converted waves of various types, respectively establishing multiple prediction models corresponding to the multiple propagation paths of the converted waves of various types; Transforming the longitudinal wave seismic time domain data and the converted wave seismic time domain data to obtain longitudinal wave seismic frequency domain data and converted wave seismic frequency domain data; predicting converted wave multiple frequency domain data according to the longitudinal wave seismic frequency domain data, converted wave seismic frequency domain data and multiple prediction models corresponding to various converted wave multiple propagation paths; and performing inverse transformation on the converted wave multiple frequency domain data to obtain converted wave multiple time domain data. In an optional manner, the establishing the multiple prediction model corresponding to each type of converted wave multiple propagation path according to the multiple types of converted wave multiple propagation paths respectively further includes: establishing a first multiple prediction model according to a pp-ps propagation path, wherein the pp-ps propagation path consists of downlink longitudinal waves, uplink longitudinal waves, downlink longitudinal waves and uplink transverse waves; And respectively establishing a second multiple prediction model and a third multiple prediction model according to two different ps-ps propagation paths, wherein the ps-ps propagation paths consist of downlink longitudinal waves, uplink transverse waves, downlink longitudinal waves and uplink transverse waves. In an alternative manner, the first multiple prediction model is: The second multiple prediction model is: the third multiple prediction model is as follows: Wherein ω is angular frequency, t s is gun focus depth delay, t k is detector depth delay, i is imaginary, As complex function, x s,ys,xr,yr is x coordinate, y coordinate, detector x coordinate and detector y coordinate between cannons, x k,yk is x coordinate and y coordinate summed, P is longitudinal wave seismic frequency domain data, and R is converted wave seismic frequency domain data. In an optional manner, the transforming the longitudinal wave seismic time domain data and the converted wave seismic time domain data to obtain longitudinal wave seismic frequency domain data and converted wave seismic frequency domain data further includes: performing Fourier transform on the longitudinal wave seismic time domain data and the converted wave seismic