Seismic data interpolation system
AssigneeGeo-X Systems, Ltd.
InventorXishuo Wang
Life of this patent
Estimating the term…
Abstract
Seismic traces of 3D surveys are interpolated using polynomials in 4D space from a less densely sampled real data set of source points S(x,y) and receiver points R(x,y). Input seismic traces should be pre-processed by normal move out (NMO) and static corrections. Input traces contributing to each single interpolated trace are chosen from a selected number of traces closest to each interpolated trace in 4D space. The methodology in accordance with the invention is designed to minimize computational load. The amplitude of an interpolated trace is modeled with a two-term Amplitude vs. Offset (AVO) formulation.
Claims11
A method of generating output seismic trace data at a specified location from input seismic trace data from known locations comprising the steps of:
- selecting a specified location for the output data;
- selecting input data at known locations for fitting to a multi-order polynomial equation;
- obtaining polynomial fit parameters of the multi-order polynomial equation;
- and
- utilizing the polynomial fit parameters to obtain seismic trace data at the selected location by interpolation.
A method of creating a shot gather from known seismic trace data obtained from a plurality of known locations within an x,y grid, the method comprising the steps of:
- a. selecting a source location within the x,y grid to which the shot gather will correspond;
- b. selecting a plurality of receiver positions that will form receiver positions for the shot gather wherein at least one of the source location or receiver positions is an unknown position;
- c. selecting a plurality of receiver traces having known positions as input traces and utilizing the known receiver traces to interpolate receiver trace data at a location adjacent the known receiver positions utilizing a least-squares fitting model;
- d. repeating step c for all selected receiver positions so as to form a shot gather corresponding to the source location.
Dependencies are inferred from the claim text.