Frequency division iteration constrained random inversion method

Abstract

The invention provides a frequency division iteration constrained random inversion method. The random inversion method comprises the steps of optimizing a drilled well needed by inversion; carrying out fine well-to-seismic calibration; expanding an existing seismic data frequency band; performing frequency division processing on the broadband seismic data after frequency expansion; extracting high-frequency wavelets; obtaining a three-dimensional apparent sand-to-ground ratio distribution body. Through an inversion method mainly based on seismic data driving, the precision of a random inversion result dominated by seismic data driving is improved, the obtained three-dimensional apparent sand-to-ground ratio distribution body serves as prior information to be added into a random inversionprocess driven by high-frequency seismic data, and therefore a high-precision frequency division iteration constraint random inversion result is obtained; according to the method, the reservoir prediction precision is effectively improved, the transverse thickness change of a reservoir is described more accurately, the real overlapping communication relation of the reservoir is recovered, and an important reference basis is provided for pre-drilling well position deployment in the exploration target evaluation stage and well position optimization in the oil field comprehensive adjustment stage.

Description

technical field [0001] The invention relates to the technical field of interpretation of oil and gas exploration seismic data, in particular to a random inversion method constrained by frequency division iterations. Background technique [0002] As an important part of the seismic data interpretation process, reservoir prediction provides an important reference for oil and gas field exploration well location deployment and development well pattern design. Reservoir prediction is a technical means to predict the development of underground reservoirs based on the original seismic data, using seismic attribute method or seismic inversion technology, and using various attribute information or inversion wave impedance results. At present, the oil and gas fields discovered in China are mainly continental reservoirs, and the continental depositional environment makes the development of reservoirs complex and diverse, which brings great challenges to reservoir prediction and reservo...

AI Technical Summary

Problems solved by technology

At present, the oil and gas fields discovered in China are mainly continental reservoirs, and the continental depositional environment makes the development of reservoirs complex and diverse, which brings great challenges to reservoir prediction and reservoir description.

Seismic attributes, 90-degree phase shift, and constrained sparse pulse inversion are effective and important technical means in traditional seismic reservoir prediction methods, which play an important role in the early exploration stage of oil and gas fields. Accuracy Requirements for Fine Exploration and Development

[0003] Seismic inversion techniques commonly used in reservoir prediction can be divided into two categories: deterministic inversion and stochastic inversion. Among them, constrained sparse pulse inversion belongs to the deterministic inversion method, and the inversion result is the best wave impedance volume. However, this method is based on the assumption of sparsity, and it is difficult to effectively reflect thin layers; while the geostatistical inversion method belongs to the category of stochastic inversion, and the inversion result is the average result of a series of equal probability realizations. Combining simulation and seismic inversion technology, it is an inversion method that makes full use of geological, seismic and logging information to obtain high-resolution inversion results. Information requirements are high. In areas where the number of participating wells is insufficient, the distribution of well patterns is uneven, and the quality of seismic data is poor, such as during the exploration and development of offshore oil and gas fields, the number of wells and the distribution range of well patterns are limited, and it is difficult for random inversion methods to fully utilize their potential. Advantage

At present, there is no relatively mature and perfect inversion technology in the industry for fine prediction and description of reservoirs under the conditions of insufficient prior information such as a small number of drilling wells or insufficient fineness of the grid model.

Images