Method and device for predicting missing logging curve

Abstract

The invention discloses a missing logging curve prediction method and device. The method comprises the steps of obtaining a logging curve, horizon data, well location data and geological information of a research area; preferably selecting a logging curve combination for establishing a machine learning network model according to correlation between a preset to-be-predicted curve of a well section with complete logging curves in a research area and other logging curves; determining the weight of sample well data for machine learning model training according to the logging curve, the horizon data, the well location data and the geological information of the research area; constructing a machine learning network model; training and verifying the machine learning network model by using the sample well data; and according to a known logging curve of a to-be-processed well in the research area, performing prediction by using the trained and verified machine learning network model to obtain a missing logging curve of the to-be-processed well. According to the invention, the accuracy of predicting missing curves by using a machine learning method can be improved.

Description

technical field [0001] The invention relates to the technical field of logging curve prediction of complex lithologic reservoirs such as carbonate rock, volcanic rock and mud shale, in particular to a method and device for predicting missing logging curves. Background technique [0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. The descriptions herein are not admitted to be prior art by inclusion in this section. [0003] Well logging is an important technical means for oil and gas exploration and development. By processing and analyzing the rock geophysical characteristic curve data such as downhole acoustics, radioactivity, and electricity obtained by measurement, the qualitative identification of oil and gas layers and the quantitative calculation of parameters can be realized, and the oil and gas reservoirs can be analyzed qualitatively. Comprehensive evaluation provides key data support. Howe...

AI Technical Summary

Problems solved by technology

However, due to the complex underground conditions, various unpredictable and unavoidable problems such as hole diameter expansion and instrument failure during the measurement process, as well as human reasons such as improper logging implementation and consideration of economic factors, some problems often occur in practical applications. In the case of distorted or missing well logging data, these missing parts or even the entire logging curve will bring great challenges to reservoir logging evaluation, and curve prediction is a common technical method to solve such problems

[0004] The traditional missing logging curve prediction mainly relies on the internal relationship between various logging data directly, such as determining the empirical relationship between the curve to be predicted and one or several known curves through methods such as crossplot and multiple regression, but due to the underground conditions Complexity and heterogeneity are strong, there is often a strong nonlinear relationship between logging data, and the mapping relationship between data is also extremely complex, and the actual application effect is poor

In recent years, with the widespread application of machine learning methods in the fields of science and engineering, many researchers have also proposed to use data-driven methods to solve geological problems, such as using support vector machines (SVM), fuzzy logic models (FLM) and artificial neural networks ( ANN) and other methods to predict well logging curves, but these methods essentially construct a mapping relationship between point-to-point or depth sequences, without taking into account the relationship between the sample data used to establish the prediction model and the well to be predicted The correlation and difference in the geological structure of oil and gas reservoirs, formation lithology changes, etc., which are contrary to the actual geological analysis experience and geological thinking, so the accuracy of predicting and generating logging curves is low

Images