An ultrasonic NDT method and system, which can extract and analyze a defect signal even when a signal reflected from a defect interferes with a unique initial pulse of an ultrasonic transducer or a signal reflected from the surface of a test object, and an autoencoder-based prediction model training method used therefor. The method may include acquiring a measured signal by transmitting an ultrasonic wave to a test object and receiving an ultrasonic wave reflected from the test object; inputting the measured signal to an autoencoder-based prediction model and predicting a reference signal which is to be expected to be measured from a test object with no defect; calculating a residual signal as the absolute value of a difference between the measured signal and the reference signal; and analyzing information on a defect contained in the test object by analyzing the residual signal.

The present disclosure provides a detection device of a display panel. The detection device includes: an image receiver configured to receive a detection image of a display panel to be detected; a detector configured to input the detection image of the display panel to be detected into a detection model and generate a detection result by the detection model, the detection model is pre-constructed and configured to detect the display panel. The disclosure also provides a detection method of the display panel, an electronic device and a computer readable medium.

To provide an anomalous sound detection training technique by which a feature amount extraction function for detecting anomalous sound can be generated irrespective of whether training data for anomalous signals is available or not. An anomalous sound detection training apparatus includes: a first function updating unit 3 that updates a feature amount extraction function and an feature amount inverse transformation function, which are input, based on an optimization index of a variational autoencoder; an acoustic feature extraction unit 4 that extracts an acoustic feature of normal sound based on training data for normal sound; a normal sound model updating unit 5 that updates a normal sound model by using the acoustic feature that is extracted; a threshold updating unit 6 that obtains a threshold .sub. corresponding to a false positive rate , which has a predetermined value, by using the training data for normal sound and the feature amount extraction function that is input; and a second function updating unit 8 that updates the feature amount extraction function that is updated, based on a Neyman-Pearson-type optimization index defined by the threshold .sub. that is obtained, and repeatedly performs processing of each of the above-mentioned units.

A method for detecting damage in a component of a rotary machine includes collecting, via one or more sensors, vibration data relating to the component. The method also includes identifying energy of at least one harmonic or sideband series within the at least one region indicative of a damaged component and identifying energy within the at least one region excluding the at least one harmonic or sideband series. Further, the method includes determining at least one damage ratio based on the energy of at least one harmonic or sideband series within the at least one region indicative of a damaged component and the energy within the at least one region excluding the at least one harmonic or sideband series. Moreover, the method includes calculating a damage factor of the component as a function of, at least, the at least one damage ratio. In addition, the method includes comparing the damage factor to a predetermined damage threshold, wherein a damage factor exceeding the predetermined damage threshold is indicative of a damaged component.

Various embodiments relating to resonance inspections and in-service parts are disclosed. One protocol (150) includes conducting a resonance inspection of an in-service part (152). The frequency response of the in-service part may be compared with a resonance standard (154) for purposes of determining whether or not the in-service part is changing abnormally (156). An in-service part that is identified as changing abnormally may be characterized as being rejected (160). An in-service part that is no identified as changing abnormally may be characterized as being accepted (158).

A method for detecting deterioration of a structural part includes: detecting a waveform of time domain of the structural part by a sensor disposed on the structural part; performing a conversion of time domain to frequency domain for the waveform of time domain by a processor electrically connected to the sensor so as to obtain an actual modal parameter of each of a plurality of modals related to a waveform of frequency domain of the structural part; comparing the actual modal parameter of each of the plurality of modals to modal parameter information stored in a database to determine whether a deterioration defect exists in the structural part; and determining a degree and a position of the deterioration defect when the deterioration defect exists in the structural part.

The invention provides methods and systems for real-time monitoring of the insulation state of wind-powered generator windings comprising the following steps: a) capturing in real-time, during a predetermined time period (in situations where the generator is synchronized to the electrical network but is not yet couplet thereto as well as in situations where the generator is producing energy) the values of one or more electrical and vibration variables of the generator; b) obtaining in real-time, the temporal evolution of the vibration and of the inverse components of electrical variables at one or more predetermined frequencies; c) identifying a possible generator insulation fault when the inverse component of at least one electrical variable and/or one vibration at a predetermined frequency exceeds an absolute threshold or a temporal increase threshold pre-sets.

A sensor assembly (2) is configured to perform fault detection in a pump assembly that includes an electric motor (70) and a fluid pump (30). The sensor assembly (2) includes a housing (4) configured to be mechanically attached to the pump (30) and configured to be attached into a bore provided in the pump (30). One or more vibration sensing element(s) (16) is/are arranged in the housing (4). The sensor assembly (2) includes a calculation unit (84) configured to receive sensor signals (V.sub.1, V.sub.2, V.sub.3) from the vibration sensing element(s) (16) and perform calculations and thereby detect motor bearing faults and cavitation.

A system and method perform automated detection, classification, and remediation of defects in a structure using ultrasound testing. The system includes an autoencoder is trained and configured to generate a de-noised UT scan image from a noisy UT scan image of a structure, a support vector machine configured to detect a defect in the structure, a convolutional neural network configured to classify the defect, and a remediation subsystem configured to remediate the defect. The method implements the system.

According to an embodiment, a diagnostic apparatus includes a sound-emitting unit, at least one measurement unit, and a processor. The sound-emitting unit includes a plurality of speakers arranged at equal angular intervals on a circumference of a circle, and is configured to emit a first vibration sound to a target by using the speakers. The at least one measurement unit is arranged on a central axis of the circle, and is configured to measure a vibration of the target generated in response to the first vibration sound, or a second vibration sound radiated from the target due to the vibration. The processor is configured to diagnose the target based on an output from the at least one measurement unit.