Oil well oil-water relationship mode judgment method and device

Abstract

The invention provides an oil well oil-water relationship mode judgment method and device. The judgment method comprises the following steps: obtaining historical production dynamic data of an oil well, and establishing a reservoir body evaluation index system of the oil well; establishing a neural network model according to the reservoir body evaluation index system; carrying out training on the neural network model to obtain a stable model; and applying the production dynamic data of the oil well to be judged to the stable model to obtain an oil-water relationship mode of the oil well to be judged. Through a semi-quantitative division method based on the genetic algorithm neural network, the evaluation result is high in reliability; and the method and device can provide theoretical basis for water breakthrough risk early warning stage control of oil wells corresponding to different types of oil-water relationship modes and remaining oil tapping potential strategy giving.

Description

technical field [0001] The invention relates to the field of oil reservoir exploitation, in particular to a method and device for judging the oil-water relationship mode of an oil well. Background technique [0002] The reservoir space types of fractured-vuggy carbonate reservoirs are mainly large caves, secondary dissolution vugs and high-angle fractures. The distribution of fractures and caves is uneven and the reservoir heterogeneity is serious. There is very active edge and bottom water, and because fractures are the storage space and the main fluid flow channel, this often leads to violent water flooding of oil wells, oil well stop blowout, production decline, invalid pressure water cone, and production failure in the actual production process. Complex water characteristics and other problems. Therefore, it is of great guiding significance to improve the recovery of this type of reservoirs by integrating a variety of dynamic and static data, classifying and evaluating ...

AI Technical Summary

Problems solved by technology

However, for extremely complex fracture-vuggy media structures, most experimental studies cannot effectively describe the water-flooding characteristics of actual reservoir conditions; at the same time, how to determine a reasonable indoor experimental scale and accurately describe the complex flow laws in reservoirs still remains to be solved. is a technical problem

On the other hand, due to the existence of multiple flow forms such as seepage, free flow, and pipe flow in fractured-cavity reservoirs, the flow mechanism is extremely complex, which makes the existing Darcy flow theory and commercial simulation software unable to accurately describe this type of flow. Coupled flow characteristics in media

Therefore, whether the research understanding based on laboratory experiments, numerical simulation or mathematical theory derivation has field application value remains to be verified.

Images