Yield prediction method and device for single-cave oil and gas well

Abstract

The invention provides a yield prediction method for a single-cave oil and gas well, which comprises the following steps: 1, selecting and collecting related parameters influencing the yield of the single-cave oil and gas well as an original data set in a well testing industrial process; 2, calculating the weight of each related parameter in the original data set by using an accelerated genetic optimization projection pursuit method to obtain a weight value corresponding to each related parameter; and 3, taking the weight value and the original data set as input information, and performing training iteration by using a neural network to realize effective prediction of the single-cave oil and gas well yield. The invention is high in calculation speed, high in real-time performance and high in precision, and the yield of the single-karst-cave oil and gas well explained by well testing can be predicted in real time; in the well testing industrial process, the combined network model is high in applicability, reliability and stability and has wide application prospects.

Description

technical field [0001] The invention relates to the technical field of production capacity of single karst cave oil and gas wells interpreted by well testing, in particular to a production prediction method and device for single karst cave oil and gas wells. Background technique [0002] The production capacity of oil and gas wells is aimed at a single well. It mainly studies the relationship between the bottom hole pressure and the surface production during the production of a single well. The production capacity test can only be applied to the depletion development method without external energy supplement during production. However, for fractured-vuggy reservoirs, oil wells generally have connectivity, and water injection, polymer injection, and other external energy supplement methods are commonly used in oil and gas development, making productivity prediction difficult. [0003] The purpose of productivity well testing is to obtain a productivity equation (the relations...

AI Technical Summary

Problems solved by technology

[0005] However, the productivity prediction of the above-mentioned dynamic data is mainly based on the assumption of constant bottomhole flow pressure, and the production curve analysis with time is formed by establishing certain models and equations. These models and methods need to meet certain conditions, such as only one well in the formation. Depletion production, sufficient production time in the early stage to ensure that the flow reaches the boundary, etc., are very difficult for fractured-vuggy reservoirs with strong heterogeneity and possible connectivity, so it is necessary to explore the process of well testing, production, etc. The method of comprehensive data for productivity analysis shortens the time for oil and gas well productivity prediction and improves the practicability and scope of application of the method

Images