Well leakage working condition prediction method and system based on non-uniform data and intelligent optimization algorithm
The invention discloses a well leakage working condition prediction method and system based on non-uniform data and an intelligent optimization algorithm, and the method comprises the steps: carrying out the analysis of various types of well drilling data, selecting well logging data, and carrying o...
Saved in:
Main Authors | , , , , , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
10.11.2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a well leakage working condition prediction method and system based on non-uniform data and an intelligent optimization algorithm, and the method comprises the steps: carrying out the analysis of various types of well drilling data, selecting well logging data, and carrying out the processing of the well logging data; improving the logging data by using an SMOTE interpolation algorithm, and establishing a sample set; a sparrow search algorithm is improved and optimized by using double-chaos mapping and an elite reverse strategy, BP neural network parameters are optimized by using the improved sparrow search algorithm, and a prediction model is constructed; and performing modeling verification by using field data. The well leakage complex prediction model based on the improved hybrid sparrow search algorithm-BP neural network is constructed, modeling verification is carried out by using field data in order to better verify the effectiveness of the model, the result shows that the proble |
---|---|
Bibliography: | Application Number: CN202310795093 |