Method for automatically recognizing reservoir cracks based on electric imaging logging porosity spectrum information

Abstract

The invention relates to a method for automatically recognizing reservoir cracks based on electric imaging logging porosity spectrum information, and belongs to the technical field of geophysical exploration methods. The method is characterized by including the following steps of calculating electric imaging logging porosity distribution spectrums, calculating statistical characteristics of the porosity distribution spectrums, calculating the crack probability of a single characteristic, establishing a crack recognition probability model integrated with information of the multiple characteristics and recognizing the reservoir cracks. According to the method, electric imaging logging data are utilized to analyze the porosity spectrums and extract the statistical characteristics of energy, entropy, the contrast ratio and the expected value from the porosity spectrums, the crack recognition probability model integrated with the information of the multiple characteristics is established according to the characteristics, the reservoir cracks are automatically recognized, and the method has the advantages that background interference can be eliminated, the cracks can be qualitatively recognized, and reliability of crack recognition is high.

Description

Technical field: [0001] The invention relates to a method for automatically identifying reservoir fractures based on electrical imaging logging porosity spectrum information, and belongs to the technical field of geophysical exploration methods. Background technique: [0002] The key to evaluating fractured reservoirs is to detect fractures accurately. In fractured reservoirs such as carbonate rocks, fractures are important fluid storage spaces and oil and gas migration pathways. Studying the development and distribution of fractures is of great significance for effective evaluation of fractured reservoirs. Due to the complex distribution of fractures and poor regularity, as well as the limitations of observation, detection means and research methods, how to effectively identify reservoir fractures such as carbonate rocks is one of the difficulties in exploration and development research. However, there is still a lack of effective evaluation methods for the prediction of f...

AI Technical Summary

Problems solved by technology

[0004] (1) Using drilling coring for fracture identification, which can directly observe the development of reservoir fractures; but the disadvantages of this method are: first, the cost is too high, and it is impossible to conduct large-scale drilling and coring for each well; The azimuth homing of reservoir fracture development is uncertain; the third is affected by reservoir fractures, the cores taken out are easily broken and difficult to effectively use

[0005] (2) Use conventional logging methods to identify fractures; according to the response of different logging sequences to reservoir fractures, conventional logging data generally used for fracture identification include acoustic logging, resistivity logging, nuclear logging, etc.; Since the sensitivity of various logging methods to reservoir fractures is not exactly the same, and some non-fracture factors may also cause the same abnormal response as reservoir fractures, it is often difficult to use one or two logging methods to identify reservoir fractures. Difficult to make definitive judgments, especially in poor borehole conditions

[0007] At present, the effectiveness of conventional logging is not enough to make it a reliable data resource for fractured reservoir evaluation: on the one hand, because many conventional logging resolutions are small, the measurement results are limited by the surrounding conditions of the wellbore; On the other hand, the logging response of reservoir fractures is a comprehensive reflection of many rock properties, and is easily affected by other conditions such as filling, mud, dissolution, etc.

[0008] (3) Use imaging logging method to identify fractures; imaging logging method has been published since the early 1990s, and belongs to the logging method that can directly detect fracture properties; imaging logging data can be displayed in an intuitive, vivid and clear manner. The geological characteristics of the two-dimensional space of the wall; however, the processing of logging data is currently mainly based on manual interactive processing; basically, the user first draws the trace and boundary of the fracture, and then the computer uses the trace and boundary information to automatically Calculate various fracture parameters; this method has high requirements for users, and manual processing has the problems of heavy workload and inaccurate positioning. The identification results are easily affected by the subjective factors and level of the interpreters, and the identification efficiency is very low.

When fully mining the information reflecting fractures, there is still a lack of modeling systems that can provide multi-information fusion of logging

Images