An Automatic Horizon Interpretation Method Based on Minimal Seismic Waveform Unit Classification

Abstract

The invention provides an automatic horizon interpretation method based on minimum seismic waveform unit classification, comprising: Step 1, performing continuity enhancement preprocessing on the seismic data of the target horizon; Step 2, performing coarse grid layer on the target horizon Interpretation; step 3, use the preliminary interpretation results obtained in step 2 as horizon time control points, and translate up and down to obtain a top-bottom time window containing multiple complete seismic waveforms; step 4, perform seismic data in the time window Minimal waveform unit division; Step 5, extraction of minimum seismic waveform unit eigenvalues; Step 6, construction of a random forest minimum seismic waveform unit classification model; Step 7, optimization of the results of automatic interpretation of horizons in the whole region. The invention reduces the optimization difficulty of the automatic layer identification algorithm, imposes professional constraints, effectively improves the reliability of the automatic interpretation result of the layer, and has a good application prospect.

Description

technical field [0001] The invention relates to the field of exploration geophysics, and in particular provides an automatic layer interpretation method based on the classification of minimum seismic waveform units. Background technique [0002] In exploration and development, seismic horizon interpretation is the basis of oil and gas exploration, and its reliability directly affects the rationality of subsequent reservoir prediction and oil and gas reservoir evaluation results. In general, horizon interpretation is basically based on manual or single-section automatic tracking. Due to the limited field of view, it is often difficult to obtain satisfactory interpretation results in a short period of time, which greatly restricts the advancement of oilfield production. The traditional horizon interpolation algorithm is simple and has high execution efficiency, but it is prone to cross-axis phenomenon, and a large amount of manual intervention is still required in complex cons...

AI Technical Summary

Problems solved by technology

This method works well in areas with relatively simple structures, but when the structure is complex, it is difficult to guarantee the accuracy and reliability of horizons; the third type is the automatic tracking technology based on artificial neural networks, and Alberts (1999) and other applications This technology has achieved good results in horizon tracking, but it mainly uses a shallow neural network, and the reliability of the interpretation results is severely restricted by the number of training samples, and it is often difficult to meet the algorithm requirements in practical applications; the fourth category is The automatic horizon tracking method based on seismic attribute analysis, Emanuele (2015, 2016) used seismic phase attributes to pick horizons, but when the signal-to-noise ratio of seismic data is low, the stability of interpretation results cannot be guaranteed

In general, the above technology improves the automatic horizon tracking algorithm and has achieved certain application effects in practical applications, but it needs to be further improved in describing the mapping relationship between seismic data and horizons

Images