Extended Short-Time Fourier Transform for Ultrasonic Velocity Profiler on Two-Phase Bubbly Flow Using a Single Resonant Frequency

A novel ultrasonic inspection method of the heat exchangers based on circumferential waves and deep neural networks

The heat exchanger (HE) is an important component of almost every energy generation system. Periodic inspection of the HEs is particularly important to keep high efficiency of the entire system. In this paper, a novel ultrasonic water immersion inspection method is presented based on circumferential wave (CW) propagation to detect defective HE. Thin patch-type piezoelectric elements with multiple resonance frequencies were adopted for the ultrasonic inspection of narrow-spaced HE in an immersion test. Water-filled HE was used to simulate defective HE because water is the most reliable indicator of the defect. The HE will leak water no matter what the defect pattern is. Furthermore, continuous wavelet transform (CWT) was used to investigate the received CW, and inverse CWT was applied to separate frequency bands corresponding to the thickness and lateral resonance modes of the piezoelectric element. Different arrangements of intact and leaky HE were tested with several pairs of thin piezoelectric patch probes in various instrumental setups. Also, direct waveforms in the water without HE were used as reference signals, to indicate instrumental gain and probe sensitivity. Moreover, all filtered CW corresponding to resonance modes together with the direct waveforms in the water were used to train the deep neural networks (DNNs). As a result, an automatic HE state classification method was obtained, and the accuracy of the applied DNN was estimated as 99.99%.

Optimization of X-ray Tube Voltage to Improve the Precision of Two Phase Flow Meters Used in Petroleum Industry

To the best knowledge of the authors, in all the former studies, a fixed value of X-ray tube voltage has been used for investigating gas–liquid two-phase flow characteristics, while the energy of emitted X-ray radiations that depends on the tube voltage can significantly affect the measurement precision of the system. The purpose of present study is to find the optimum tube voltage to increase the accuracy and efficiency of an intelligent X-ray radiation-based two-phase flow meter. The detection system consists of an industrial X-ray tube and one detector located on either side of a steel pipe. Tube voltages in the range of 125–300 kV with a step of 25 kV were investigated. For each tube voltage, different gas volume percentages (GVPs) in the range of 10–90% with a step of 5% were modeled. A feature extraction method was performed on the output signals of the detector in every case, and the obtained matrixes were applied to the designed radial basis function neural networks (RBFNNs). The desired output of the networks was GVP. The precision of the networks in every voltage and every number of neurons in the hidden layer were obtained. The results showed that 225 kV tube voltage is the optimum voltage for this purpose. The obtained mean absolute error (MAE) for this case is less than 0.05, which demonstrates the very high precision of the metering system with an optimum X-ray tube voltage.

Multi-Classification of Complex Microseismic Waveforms Using Convolutional Neural Network: A Case Study in Tunnel Engineering

Due to the complexity of the various waveforms of microseismic data, there are high requirements on the automatic multi-classification of such data; an accurate classification is conducive for further signal processing and stability analysis of surrounding rock masses. In this study, a microseismic multi-classification (MMC) model is proposed based on the short time Fourier transform (STFT) technology and convolutional neural network (CNN). The real and imaginary parts of the coefficients of microseismic data are inputted to the proposed model to generate three classes of targets. Compared with existing methods, the MMC has an optimal performance in multi-classification of microseismic data in terms of Precision, Recall, and F1-score, even when the waveform of a microseismic signal is similar to that of some special noise. Moreover, semisynthetic data constructed by clean microseismic data and noise are used to prove the low sensitivity of the MMC to noise. Microseismic data recorded under different geological conditions are also tested to prove the generality of the model, and a microseismic signal with M_w ≥ 0.2 can be detected with a high accuracy. The proposed method has great potential to be extended to the study of exploration seismology and earthquakes.

Influence of Flow Disturbances behind the 90° Bend on the Indications of the Ultrasonic Flow Meter with Clamp-On Sensors on Pipelines

The subject matter of the article concerns velocities/flow rate measurements in the area of disturbed flows-behind the 90° bend. They were conducted by means of an ultrasonic flowmeter with clamp-on sensors on pipeline, for water and two different Reynolds numbers of 70,000 and 100,000, corresponding to two velocities of approximately 1.42 m/s and 2.04 m/s. The tests were carried out at 12 distances from the disturbance. Sensors on the circumference of the pipeline were mounted 30° each. The correction factor values were calculated for the given measurement geometry. The measurements have shown that the values of this coefficient are always greater than 1, which means that the ultrasonic flow meter understates the speed values. They also showed that already at a distance of 8 nominal diameters from the disturbance, the correction factor does not exceed 1.02, so the measurement errors are within the maximum permissible error (MPE) of a typical ultrasonic flow meter. For distances less than eight nominal diameters from the disturbance, not taking the correction factor value into the account can lead to systematic errors of up to 10.8%. Studies have also proved that in each measurement plane behind the disturbance there are two mounting angles for the ultrasonic sensors, 60° and 240° respectively, for which the correction factor values are minimal. Additionally, using the laser Doppler anemometry (LDA) method, velocity solids were determined at individual distances from the disturbance, and the projections of velocity blocks on the appropriate plane represented velocity profiles and indicated the distances from the disturbance at which these profiles stabilise.

Comparison of Signal-Analysis Techniques for Seismic Detection System for High-Speed Train Data: Effect of Bridge Structures

This study evaluated the earthquake warning system for high-speed trains using onboard accelerometers instead of expensive seismometers. Onboard accelerometers measure the train data additional to the earthquake acceleration. The measured earthquake acceleration could also be modified by railroad-supporting bridges. To develop the data analysis system, the virtual onboard data sets are synthesized using the train acceleration data and earthquake data. Not only the earthquake acceleration data but also the earthquake responses of bridges are used for the virtual onboard data synthesis. For the analysis of synthesized data, the short-time Fourier Transform (STFT), the wavelet transform (WT), and Wigner–Ville Distribution (WVD) methods have been compared. Results show that WVD provides the best detection performance while the computational costs are large.

Development of a Seismic Detection Technology for High-Speed Trains Using Signal Analysis Techniques

As the occurrence of earthquakes is increasing in South Korea, the earthquake early warning (EEW) system becomes indispensable for the protection of high-speed railways. Although the importance of EEW system has been increasing, the number of installed seismic accelerometers in South Korea is not sufficient to provide rapid information. This study uses a stochastic signal analysis technique to utilize the smartphone sensors for the rapid EEW system. From the train vibration data from the low fidelity on-board accelerometer, the virtual earthquake detection data in the train by smartphone sensor has been constructed. To analyze the stochastic characteristics of the constructed data, the short time Fourier transform (STFT) approach has been applied. The study’s overall objective is to offer stochastic approaches that provide effective analysis of the low fidelity sensor data, such as smartphone sensor data, for the rapid EEW system.