The present invention discloses ultrasonic nondestructive methods for pipe wall thickness measurement at high or low temperatures. An ultrasonic detection device comprises a first and a second ultrasonic waveguide. The waveguide length is selected according to the surface temperature of a pipe under inspection. A first piezoelectric plate causes generation of a plurality of ultrasonic excitation signals which is transmitted to the pipe through the first ultrasonic waveguide. The plurality of ultrasonic excitation signals has different group speeds when traveling along the first ultrasonic waveguide. The reflected ultrasonic wave signals are collected and transmitted to a second piezoelectric plate by the second ultrasonic waveguide. The pipe wall thickness is calculated using an ultrasonic wave signal which has the highest group speed. The first and second waveguides are arranged parallel and side by side. An isolation plate is disposed such that the first and second waveguides go through the plate perpendicularly.

Systems, techniques, and computer-implemented processes for acoustic signal based improvements to one or more process steps in the manufacture of battery cells. Information gathered based on an acoustic signal based analysis in one process step can be used in one or more other process steps using any suitable combination of feedback and/or feedforward of the acoustic signal based analysis. Such feedback and/or feedforward can improve the overall quality of battery cells produced using the manufacturing process, efficiency/cost of the manufacturing process, improvement in yield/reduction in wastage of the battery cells produced using the manufacturing process and/or improvements in individual process steps.

Methods, systems, and apparatuses are provided for detecting conditions associated with a fluid conduit. An apparatus includes an insert having an internal conduit to connect with the fluid conduit and a plenum volume, and a detection device including a housing connected to the insert within the plenum volume, an acoustic sensor to receive acoustic signals, an acoustic exciter to apply acoustic signals to the housing, and a controller. The controller is electrically connected to the acoustic sensor and the acoustic exciter. The controller is configured to cause the acoustic exciter to apply an input acoustic signal to the housing, receive the acoustic signals from the housing using the acoustic sensor, analyze the received acoustic signals to determine a pipe condition of a pipe defining the fluid conduit or fluidically connected to the fluid conduit, and cause data representative of the pipe condition to be transmitted to an external device.

Aspects of the present disclosure are directed to a suite of testing apparatuses and non-destructive, acoustic inspection methods for scanning and inspecting batteries to determine and characterize various physical phenomena in these batteries. In one aspect, a rastering system for non-invasive and acoustic inspection of battery cells includes a holder for placing a battery cell inside the system for the acoustic inspection, at least one transducer configured to perform acoustic measurements on the battery cell, and a controller configured with inspection parameters for performing the acoustic measurements, the inspection parameters being dynamic and interchangeable depending on at least one or more of a shape, a size, and a form factor of the battery cell.

An additive manufacturing apparatus according to one embodiment includes a manufacturing unit, an elastic wave generation unit, an elastic wave detection unit, and an inspection unit. The manufacturing unit sequentially stacks a layer formed by emitting a first energy beam to a material and solidifying the material. The elastic wave generation unit emits a second energy beam to a manufactured object including the layer and generates an elastic wave propagating in the manufactured object. The elastic wave detection unit detects the elastic wave. The inspection unit inspects the manufactured object on the basis of a detection result from the elastic wave detection unit.

A prediction method of part surface roughness and tool wear based on multi-task learning belong to the file of machining technology. Firstly, the vibration signals in the machining process are collected; next, the part surface roughness and tool wear are measured, and the measured results are corresponding to the vibration signals respectively; secondly, the samples are expanded, the features are extracted and normalized; then, a multi-task prediction model based on deep belief networks (DBN) is constructed, and the part surface roughness and tool wear are taken as the output of the model, and the features are extracted as the input to establish the multi-task DBN prediction model; finally, the vibration signals are input into the multi-task prediction model to predict the surface roughness and tool wear.

A system for nondestructive inspection of structures. The system includes an omni-directional unmanned aerial vehicle (UAV) and a support arm extending outward from the UAV body from a first end attached to the body to a second distal end, which is used to support a nondestructive testing (NDT) sensor. The system includes an autopilot module stabilizing the flight of the omni-directional platform. The autopilot includes a wall-tracking mode, which determines the normal of the structure's surface, and the omni-directional UAV is stabilized to fly with the support arm aligned with normal to the surface and with the second end proximate to the surface with the UAV body in any orientation in space. The UAV operates to follow a flight path whereby a longitudinal axis of the support arm coincides with the normal and the sensor is positioned in predefined measurement position relative to the structure surface to take the measurements.

A prediction method of part surface roughness and tool wear based on multi-task learning belong to the file of machining technology. Firstly, the vibration signals in the machining process are collected; next, the part surface roughness and tool wear are measured, and the measured results are corresponding to the vibration signals respectively; secondly, the samples are expanded, the features are extracted and normalized; then, a multi-task prediction model based on deep belief networks (DBN) is constructed, and the part surface roughness and tool wear are taken as the output of the model, and the features are extracted as the input to establish the multi-task DBN prediction model; finally, the vibration signals are input into the multi-task prediction model to predict the surface roughness and tool wear.

A method includes exciting a vibration of a solid material preferably in cylindrical shape, detecting the vibration by a pair of transducers attached at opposing positions on the solid material and preferably orthogonally aligned to the excitation force, adding the pair of sensor signals to mostly cancel bending mode signal components, and processing the added output signal to obtain a frequency spectrum. A method may also include identifying a resonance peak of an extensional mode diametrically symmetric, measuring a characteristic of the resonance peak, comparing the characteristic with a series of measured values for standard samples, and determining a defect status such as decay in a wooden pole solid material.

Methods, systems, and apparatuses are provided for detecting and determining conditions of and conditions within a fluid conduit.