Deep shale oil ground microseism weak signal extraction and three-dimensional positioning method
By calculating the characteristic difference degree and frequency domain characteristic evaluation in deep shale oil exploration, the target signal set was screened and a deep neural network model was adopted to solve the problem of weak seismic signals being difficult to extract and locate in traditional methods, and to achieve high-precision three-dimensional location of hydraulic fracturing fractures.
Patent Information
- Application Number
- CN202610545701.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-04-23
- Publication Date
- 2026-05-26
AI Technical Summary
In the exploration and development of deep shale oil, traditional microseismic monitoring methods are difficult to effectively extract weak seismic signals, resulting in large deviations in the location of hydraulic fracturing fractures and a high misjudgment rate, which affects development efficiency.
Signals are classified by calculating the characteristic differences between detector channels, and a target signal set is selected. Then, three-dimensional localization is performed by combining frequency domain feature evaluation and deep neural network model to eliminate spurious signal interference and ensure the integrity and accuracy of the signal.
It improves the accuracy of three-dimensional positioning of fracturing fractures in deep shale oil, reduces the interference of pseudo signals on positioning, ensures the signal-to-noise ratio and the purity of positioning input, and achieves high-precision spatial coordinate mapping.