Back to results
US 2026/0177530 A1Published application
Structural inspection using machine learning models
AssigneeChevron U.S.A. Inc.
InventorsRobert Kwan Meng Seah, David Hugh Fraser, Larry A. Bowden, JR.
Life of this patent
Estimating the term…
Abstract
Different machine learning models are trained to detect different types of defects in structures. A machine learning model is trained using pairs of waveforms/wavefields measured from a structure over durations of time and labels for the defect in the structure, which enables the machine learning model to learn and interpret features from acoustic wave amplitude propagation over time. The trained machine learning models serve as proxy of the physics models to detect defects in the structures. The trained machine learning models improves the efficiency and effectiveness of the inspection process by analyzing the waveform/wavefield measurements more accurately and quickly.
Claims20
A system for structural inspection, the system comprising:
- one or more physical processors configured by machine-readable instructions to:
- obtain different sets of acoustic excitation measurement from one or more structures for different defects, the different sets of acoustic excitation measurement from the one or more structures for different defects including a first set of acoustic excitation measurement from the one or more structures for a first type of defect and a second set of acoustic excitation measurement from the one or more structures for a second type of defect different from the first type of defect, the different sets of acoustic excitation measurement from the one or more structures for different defects including waveforms/wavefields measured from the one or more structures over durations of time;
- obtain different sets of defect characterization for the one or more structures, the different sets of defect characterization for the one or more structures corresponding to the different sets of acoustic excitation measurement from the one or more structures, the different sets of defect characterization for the one or more structures including a first set of defect characterization for the one or more structures corresponding to the first set of acoustic excitation measurement from the one or more structures for the first type of defect and a second set of defect characterization for the one or more structures corresponding to the second set of acoustic excitation measurement from the one or more structures for the second type of defect;
- train different machine learning models for structural inspection, the different machine learning models including a first machine learning model trained based on the first set of acoustic excitation measurement from the one or more structures for the first type of defect and the first set of defect characterization for the one or more structures and a second machine learning model trained based on the second set of acoustic excitation measurement from the one or more structures for the second type of defect and the second set of defect characterization for the one or more structures, wherein training of the different machine learning models using the waveforms/wavefields measured from the one or more structures over the durations of time enables the different machine learning models to learn and interpret features from acoustic wave amplitude propagation over time;
- and
- store the trained machine learning models for structural inspection in a non-transient storage medium.
3 more elements
A method for structural inspection, the method comprising:
- obtaining different sets of acoustic excitation measurement from one or more structures for different defects, the different sets of acoustic excitation measurement from the one or more structures for different defects including a first set of acoustic excitation measurement from the one or more structures for a first type of defect and a second set of acoustic excitation measurement from the one or more structures for a second type of defect different from the first type of defect, the different sets of acoustic excitation measurement from the one or more structures for different defects including waveforms/wavefields measured from the one or more structures over durations of time;
- obtaining different sets of defect characterization for the one or more structures, the different sets of defect characterization for the one or more structures corresponding to the different sets of acoustic excitation measurement from the one or more structures, the different sets of defect characterization for the one or more structures including a first set of defect characterization for the one or more structures corresponding to the first set of acoustic excitation measurement from the one or more structures for the first type of defect and a second set of defect characterization for the one or more structures corresponding to the second set of acoustic excitation measurement from the one or more structures for the second type of defect;
- training different machine learning models for structural inspection, the different machine learning models including a first machine learning model trained based on the first set of acoustic excitation measurement from the one or more structures for the first type of defect and the first set of defect characterization for the one or more structures and a second machine learning model trained based on the second set of acoustic excitation measurement from the one or more structures for the second type of defect and the second set of defect characterization for the one or more structures, wherein training of the different machine learning models using the waveforms/wavefields measured from the one or more structures over the durations of time enables the different machine learning models to learn and interpret features from acoustic wave amplitude propagation over time;
- and
- storing the trained machine learning models for structural inspection in a non-transient storage medium.
2 more elements
Dependencies are inferred from the claim text.