Consistent Array-forming And Staking Study Of High-Density Seismic Data

High-density seismic is a technology which conducts point-receiver acquisition with digital geophones in the field and carries out digital array-forming processing indoor. During the practical implementation, high-density seismic acquires data in the manner of point source/receiver, small bin size, small spatial interval, uniform sampling, symmetrical sampling, large dynamic range, multi-channel receiver, full azimuth and high fold number, which can obtain seismic information of higher frequency, better fidelity and higher resolution, and can also improve imaging precision and resolution, refine reservoir description. High-density seismic technology plays an important role in the oil/gas field exploration and development.Seismic forward-modeling is a very efficient means used in seismic wavefield analysis. Based on high-order finite- difference method, this paper simulated high-density acquisition of a complex model, studied the advantage of high-density data in describing small structure; then Rayleigh wave propagation is simulated using stagger-grid finite-difference method. What's more, we analyzed the characteritics of high-density data, including the influence of sample interval, bin size and fold number on SNR(signal to noise ration) and resolution of seismic data.Taking advantage of better uniformity and symmetry in high-density seismic data, we developed F-Kx-Ky filtering method based on cross-spreads extracted from high-density seismic geometry, improved the SNR of the pre-stack data. The key point of this paper focuses on consistent array-forming technology of high-density seismic data. First, we conducted pre-stack time and phase correction, removed intra-array unconsistencies which may be caused by some factors such as layout error, sensitivity difference, elevation difference between geophones,etc., then array-forming processing is carried out.We studied the key technologies in non-repeat seismic data processing, of which conventional data is treated as reference data and high-density data as monitoring data. We built two geophysical models which are different in reservoir parameters, then designed two geometries for data modeling using high-order finite-difference method. In order to remove the data inconsistencies caused by difference in geometry and excitation conditions, we applied data matching and consistency correction to the two datasets, obtained the actual seismic response of fluid changes in reservoir.