The present invention provides a method for analyzing a liquid sample that solves a problem of a kinetically slow equilibrium at a liquid-solid interface, the problem occurring when liquid sample analysis is performed with a chemical sensor. In the method for analyzing a liquid sample according to an embodiment of the present invention, a component to be analyzed in a liquid sample is adsorbed on a receptor layer of a chemical sensor, one or more kinds of gases are then supplied to the chemical sensor, and a response thereof is measured. As a result, since a slow equilibrium at a liquid-solid interface is not used, a high-sensitivity measurement can be performed in a short time, and existing findings regarding analysis of gas samples on which much progress in research has been achieved can be used.

A multi-scale inspection and intelligent diagnosis system and method for tunnel structural defects includes: a traveling section; a supporting section, disposed on the traveling section, and including a rotatable telescopic platform, where two mechanical arms working in parallel are disposed on the rotatable telescopic platform; an inspection section, mounted on the supporting section, and configured to perform multi-scale inspection on surface defects and internal defects in different depth ranges of a same position of a tunnel structure, and transmit inspected defect information to a control section; and the control section, configured to: construct a deep neural network-based defect diagnosis model; construct a data set by using historical surface defect and internal defect information, and train the deep neural network-based defect diagnosis model; and receive multi-scale inspection information in real time, and automatically recognize types, positions, contours, and dielectric attributes of the internal and surface defects.

An ultrasonic inspection system, method, and software. In one embodiment, the ultrasonic inspection system includes an ultrasonic probe that directs ultrasound waves into a structure from a front wall, and receives reflected waves to generate a response signal. The system further includes a processor that rectifies the response signal to generate a rectified signal, integrates a portion of the rectified signal within a detection time window to determine an energy sum, and generates output based on the energy sum. The detection time window is restricted to a front wall reflection and at least a portion of a near-surface dead zone following the front wall reflection.

Systems and methods for prediction of state of charge (SOH), state of health (SOC) and other characteristics of batteries using acoustic signals, includes determining acoustic data at two or more states of charge and determining a reduced acoustic data set representative of the acoustic data at the two or more states of charge. The reduced acoustic data set includes time of flight (TOF) shift, total signal amplitude, or other data points related to the states of charge. Machine learning models use at least the reduced acoustic dataset in conjunction with non-acoustic data such as voltage and temperature for predicting the characteristics of any other independent battery.

Systems and methods for prediction of state of charge (SOH), state of health (SOC) and other characteristics of batteries using acoustic signals, includes determining acoustic data at two or more states of charge and determining a reduced acoustic data set representative of the acoustic data at the two or more states of charge. The reduced acoustic data set includes time of flight (TOF) shift, total signal amplitude, or other data points related to the states of charge. Machine learning models use at least the reduced acoustic dataset in conjunction with non-acoustic data such as voltage and temperature for predicting the characteristics of any other independent battery.

A motor noise detecting device according to an embodiment of the present disclosure includes a signal sensing part for sensing an acoustic signal generated from an object to be tested, a data acquisition part for receiving the acoustic signal sensed by the signal sensing part and converting it into an acoustic digital signal, and a data analysis part for receiving and analyzing the acoustic digital signal to perform a detection on whether the object to be tested is abnormal. In addition, the signal sensing part includes an AE (Acoustic Emission) sensor for sensing an elastic wave included in the acoustic signal, and the data analysis part generates result data of analyzing the acoustic digital signal, analyzes the generated result data through a pre-learned model, and detects whether the object to be tested is abnormal.

A processing state detecting device for detecting a processing state of a workpiece processed by laser processing includes: a sound collecting unit that measures sound while the workpiece is being processed by laser processing; an installation position evaluating unit that determines whether an installation position of the sound collecting unit needs to be changed, on the basis of the sound measured by the sound collecting unit; and an evaluation result informing unit that provides information on a result of evaluation of the installation position evaluating unit.

A method of creating and characterizing a representative image of the surface of an object from acoustic emissions of a multimode ultrasonic probe tip and transducer integrated into a micro tool, such as a nano indenter or a nano indenter interfaced with a Scanning Probe Microscope (SPM). The representative image may be utilized to predict mechanical properties or characteristics of the sample, including topography, fracture patterns, indents and artifacts. The tip component is configured to operate at multi-resonant frequencies providing sub-nanometer vertical resolution. The tip component may be quasi-statistically calibrated and deep learning iterative image comparison and characterization may be utilized to derive mechanical properties of a sample.

A method is provided for determining the geometry of one or more real, examined defects of a metallic and in particular magnetizable object, in particular a pipe or a tank, by means of at least two reference data sets of the object generated on the basis of different, non-destructive measuring methods.

Methods and systems for anomaly detection include training a neural network model to identify a form of data augmentation that has been performed on a waveform. Multiple forms of data augmentation are performed on a sample waveform to generate data augmentation samples. The data augmentation samples are classified with the neural network model. An anomaly score is determined based on the classification of the data augmentation samples.