Deep learning methods for wellbore pipe inspection

    公开(公告)号:US11976546B2

    公开(公告)日:2024-05-07

    申请号:US17114591

    申请日:2020-12-08

    IPC分类号: E21B47/002

    CPC分类号: E21B47/0025 E21B2200/22

    摘要: Methods and systems for inspecting the integrity of multiple nested tubulars are included herein. A method for inspecting the integrity of multiple nested tubulars can comprise conveying an electromagnetic pipe inspection tool inside the innermost tubular of the multiple nested tubulars; taking measurements of the multiple nested tubulars with the electromagnetic pipe inspection tool; arranging the measurements into a response image representative of the tool response to the tubular integrity properties of the multiple nested tubulars; and feeding the response image to a pre-trained deep neural network (DNN) to produce a processed image, wherein the DNN comprises at least one convolutional layer, and wherein the processed image comprises a representation of the tubular integrity property of each individual tubular of the multiple nested tubulars.

    Downhole apparatus with a valve arrangement

    公开(公告)号:US11946339B2

    公开(公告)日:2024-04-02

    申请号:US18065733

    申请日:2022-12-14

    IPC分类号: E21B34/10 E21B34/14 E21B43/10

    摘要: A valve arrangement is for a downhole apparatus having a tubular body having first and second ports in a wall thereof. The valve arrangement comprises a valve member that comprises a first port associated with the first port of the tubular body; and a second port associated with the second port of the tubular body. The valve arrangement is configurable to be locked in a first configuration with the tubular body, such that the first port is closed and the second port is closed. The valve arrangement is configurable to be locked in a second configuration with the tubular body, such that the first port is open and the second port is closed. The valve arrangement is configurable to be locked in a third configuration with the tubular body, such that the first port is closed and the second port is open.

    Transfer printer and method
    69.
    发明授权

    公开(公告)号:US11919320B2

    公开(公告)日:2024-03-05

    申请号:US17590323

    申请日:2022-02-01

    摘要: A method for monitoring a characteristic of a printed image of a thermal transfer printer. The method comprises providing a ribbon and a substrate at a printing location of the thermal transfer printer. The method further comprises printing an image on the substrate at the printing location by transferring ink from a region of the ribbon in a printing operation, a negative image being formed on the region of ribbon. The method further comprises transporting the region of ribbon, by a ribbon transport system, from the printing location towards an imaging location along a ribbon transport path. The method further comprises when a characteristic of the ribbon transport meets a predetermined criterion, obtaining, by an image capture system, a ribbon image of the negative image. The method further comprises processing said ribbon image to generate data indicative of the characteristic of the printed image.

    Classifying downhole test data
    70.
    发明授权

    公开(公告)号:US11891882B2

    公开(公告)日:2024-02-06

    申请号:US16932056

    申请日:2020-07-17

    发明人: Jiazuo Zhang

    摘要: Disclosed embodiments include methods and systems for classifying test data. In one embodiment a method includes determining one or more variable types in a multivariate test vector within a data set, and for a plurality of machine-learning models, determining a closest match between variable types used by (to train) the machine-learning models and the determined variable types for the test vector. In response to determining a closest match for one machine-learning model, a corresponding machine-learning model is selected and the test vector is classified using the selected model. In response to determining a closest match for multiple machine-learning models, a similarity is determined between a probability distribution for the test data set and the probability distributions for the multiple machine-learning models to generate similarity values for each of the models. In response to one of the similarity values exceeding a threshold value, a machine-learning model is selected that corresponds to the exceeding similarity value and the test vector is classified using the selected model.