Precise reservoir prediction method based on waveform classification and retrieval under forward constraints

Abstract

The invention discloses a precise reservoir prediction method based on waveform classification and retrieval under forward constraints. The method comprises the steps of selecting an effective range and converting large-region multi-phase spread into micro-region single phase spread; dividing a reservoir in the single phase spread into different types, and summarizing sedimentary characteristics of different types of reservoirs and post-stack seismic reflection characteristics; simulating a typical waveform of each type of reservoirs via wave equation forward modeling; performing frequency expanding treatment on the single phase spread seismic data by using well control mixed phase wavelet deconvolution; performing primary waveform classification on a single phase spread range by using a non-supervision neural network classification method; and reconstructing a waveform model trace, re-classifying the waveforms, and implementing waveform retrieval. According to the method provided by the invention, a corresponding relation between the reservoir types and the waveforms is built by wave equation forward modeling, and the waveforms are primarily classified and then retrieved to achieve precise prediction of the different types of reservoirs in the phase spread, so that the method is the effective and rapid precise prediction technology for recognizing the different types of reservoirs in the same phase spread in a small range.

Description

technical field [0001] The present invention relates to the technical field of precise prediction of reservoirs in seismic data, in particular to the combination of advantages of unsupervised waveform classification and supervised waveform classification of drilled wells, the use of wave equation forward modeling to establish the corresponding relationship between reservoir types and waveforms, and the preliminary classification of waveforms Afterwards, the waveform retrieval is used to realize the fine prediction method of different types of reservoirs in a certain facies zone. Background technique [0002] The waveform feature of a signal is a comprehensive representation of information such as amplitude, phase, and time window. Seismic waves are a type of signal. Therefore, in a physical sense, seismic waveforms are a comprehensive representation of seismic reflection wave amplitude, phase, and frequency. In the geological sense, the seismic waveform is the comprehensive ...

AI Technical Summary

Problems solved by technology

[0005] The second problem is that the unsupervised neural network seismic waveform classification technology currently mainly used can realize automatic clustering without well supervision, but considering that seismic data from acquisition to final processing and mapping are affected by underground geological conditions, acquisition equipment, noise and The combined effects of other interferences, post-processing methods and other factors lead to a certain degree of multi-solution in seismic data, and completely unsupervised neural network waveform clustering often leads to mismatches between seismic classification and geological body classification

So how can we not only effectively retain the advantages of neural network waveform classification not being limited by models and wells, and truly reflect the waveform changes of seismic data volumes, but also increase the supervision of wells drilled to avoid earthquakes, geology, etc. There is no effective method for repeated test classification when the body classification is not uniform.

Images