A computer-based method and system for predicting the propagation of cracks along a pipe is provided, wherein successive time-indexed ultrasound images of a pipe surface are captured and digitized. A computer vision algorithm processes the images to identify defects in the pipe, including cracks. At least one blob detection module is used to identify groups of cracks on the pipe surface that have created detectable areas of stress concentration or a prescribed likelihood of crack coalescence or crack cross-influence. The center locations and radial extents of respective blobs are each parametrized as a function of time and pipe surface location by determining parity relationships between successive digital data sets from successive captured images. The determined parity relationships are then used as training data for a machine learning process to train a system implementing the method to predict the propagation of cracks along the pipe.

A system and method apply an input noisy ultrasonic test (UT) scan image to an input layer of a convolutional neural network, generate a feature map using a convolutional layer, pool the feature map using a pooling layer, apply the pooled feature map to a fully connected layer, generate a de-noised UT scan image, and output the de-noised UT scan image from an output layer.

A method is provided for characterizing a part includes: a) carrying out measurements using a sensor, the sensor being placed on the part or facing the part; b) forming at least one measurement matrix using the measurements performed in step a); c) using the matrix as input datum of a convolutional neural network including an extracting block, configured to extract features from each input datum; a classifying block, configured to classify the features extracted, the classifying block outputting to at least one node; and d) depending on each node, detecting the presence of a defect in the part. The neural network employed in step c) is established using the extracting block of another previously parametrized neural network.

The present disclosure provides a system and method for real-time visualization of a material during ultrasonic non-destructive testing. The system includes a graphical user interface (GUI) capable of showing a three-dimensional (3-D) image of a composite laminate constructed of a series of two-dimensional (2-D) cross sections. The GUI is capable of displaying the 3-D image as each additional 2-D cross section is scanned by an ultrasonic testing apparatus in real time or near real time, including probable defect regions that contain a flaw such as a hole, crack, wrinkle, or foreign object within the composite. Furthermore, in one embodiment, the system includes an artificial intelligence capable of highlighting defect areas within the 3-D image in real time or near real time and providing data regarding each defect area, such as the depth, size, and/or type of each defect.

An inspection device includes a storage unit for storing a mechanical property inference model generated by machine learning based on mechanical property information and nondestructive inspection information of a fiber-reinforced first molded article region for which the mechanical property information and the nondestructive inspection information are known, the mechanical property inference model being configured to be input nondestructive inspection information of a second molded article region which is reinforced with reinforcing fibers for predicting unknown mechanical property information of the second molded article region. The inspection device acquires the nondestructive inspection information of the second molded article region; inputs the nondestructive inspection information to the mechanical property inference model; acquire the mechanical property information of the second molded article region predicted by the mechanical property inference model; and outputs the mechanical property information of the second molded article region.

A system for detecting machining errors for a laser machining system for machining a workpiece includes: a detection unit for detecting image data and height data of a machined workpiece surface; and a computing unit. The computing unit is designed to generate an input tensor based on the detected image data and height data and to determine an output tensor on the basis of the input tensor using a transfer function. The output tensor contains information on a machining error.

A system for non-destructive testing of a bond condition of concrete beams reinforced by steel rods is described. The system includes a transducing transmitter, a transducing receiver, and an ultrasonic pulse generator configured to generate drive signals for the transducing transmitter and receive a plurality vibrational waves at the transducing receiver. The system further includes a computing device including a measurement circuit configured to record a transit time for each vibrational wave and divide a distance between the transducing transmitter and the transducing receiver by the transit time to determine a pulse velocity of each vibrational wave, a comparison circuit configured to identify a highest pulse velocity of the vibrational waves and compare each highest pulse velocity to a first reference pulse velocity, and a decision circuit including an artificial neural network configured to identify a compromised bond condition around a steel rod.

A system for non-destructive testing of a bond condition of concrete beams reinforced by steel rods is described. The system includes a transducing transmitter, a transducing receiver, and an ultrasonic pulse generator configured to generate drive signals for the transducing transmitter and receive a plurality vibrational waves at the transducing receiver. The system further includes a computing device including a measurement circuit configured to record a transit time for each vibrational wave and divide a distance between the transducing transmitter and the transducing receiver by the transit time to determine a pulse velocity of each vibrational wave, a comparison circuit configured to identify a highest pulse velocity of the vibrational waves and compare each highest pulse velocity to a first reference pulse velocity, and a decision circuit including an artificial neural network configured to identify a compromised bond condition around a steel rod.

An example system for monitoring a characteristic of a material. The system includes a stimulator to provide a stimulus signal to the material. The stimulus signal includes at least one of an electrical signal, a magnetic signal, an optical signal, and an acoustic signal. The system includes a sensor to measure a response signal from the material. The response signal includes at least one of an electrical signal, a magnetic signal, an optical signal, and an acoustic signal. At least one of the stimulus signal and the response signal includes an optical signal or an acoustic signal. The system further includes a controller in communication with the stimulator and the sensor to apply machine learning to determine a characteristic of the material based on the stimulus signal and the response signal, wherein the characteristic is not directly measurable.

A smart acoustic information recognition-based welded weld impact quality determination method and system, comprising: controlling a tip of an ultrasonic impact gun (1) to perform impact treatment on a welded weld with different treatment pressures, treatment speeds, treatment angles and impact frequencies, obtaining acoustic signals during the impact treatment, calculating feature values of the acoustic signals, and constructing an acoustic signal sample set including various stress conditions; marking the acoustic signal sample set according to impact treatment quality assessment results for the welded weld; establishing a multi-weight neural network model, and using the marked acoustic signal sample set to train the multi-weight neural network model; obtaining feature values of welded weld impact treatment acoustic signals to be determined, inputting the feature values into the trained multi-weight neural network model, and outputting determination results for welded weld impact treatment quality to be determined.