Advances in passive acoustic detection, localization, and tracking applied to unmanned underwater vehicles

Detection, classification, localization, and tracking (DCLT) of unmanned underwater vehicles (UUVs) in the presence of shipping traffic is a critical task for passive acoustic harbor security systems. In general, vessels can be tracked by their unique acoustic signature due to machinery vibration and cavitation noise. However, cavitation noise of UUVs is considerably quieter than ships and boats, making detection significantly more challenging. In this thesis, I demonstrated that it is possible to passively track a UUV from its highfrequency motor noise using a stationary array in shallow-water experiments with passing boats. First, causes of high frequency tones were determined through direct measurements of two UUVs at a range of speeds. From this analysis, common and dominant features of noise were established: strong tones at the motor’s pulse-width modulated frequency and its harmonics. From the unique acoustic signature of the motor, I derived a high-precision, remote sensing method for estimating propeller rotation rate. In shallow-water UUV field experiments, I demonstrated that detecting a UUV from motor noise, in comparison to broadband noise from the vehicle, reduces false alarms from 45% to 8.4% for 90% true detections. Beamforming on the motor noise, in comparison to broadband noise, improved the bearing accuracy by a factor of 3.2×. Because the signal is also high-frequency, the Doppler effect on motor noise is observable and I demonstrate that range rate can be measured. Furthermore, measuring motor noise was a superior method to the “detection of envelope modulation on noise” algorithm for estimating the propeller rotation rate. Extrapolating multiple measurements from the motor signature is significant because Bearing-Doppler-RPM measurements outperform traditional bearing-Doppler target motion analysis. In the unscented Kalman filter implementation, the tracking solution accuracy for bearing, bearing rate, range, and range rate improved by a factor 2.2×, 15.8×, 3.1×, and 6.2× respectively. These findings are significant for improving UUV localization and tracking, and for informing the next-generation of quiet UUV propulsion systems.

Assessment of Noise Signature for a Cavitating Centrifugal Pump

In recent days, sophisticated instruments have emerged to obtain an online measurement of performance parameters from centrifugal pump of different kinds and the signals can be directed to the hands of pump user through mobile application. With this in mind, a centrifugal pump of low specific speed was chosen for cavitation studies from 80% to 120% of nominal flow rate and for three different speeds. An assessment was carried out for cavitation noise signature from those operating condition of that pump. The result of cavitation noise based on peak magnitude as well as average revealed a nature in relation to cavitation coefficient and it greatly depends on the flow rate with respect to nominal flow rate. The noise envelope for the flow rate at best efficiency and above was having similar trend whereas at flows less than the nominal, it was totally different. So the criteria for finding the deviation in noise cannot be uniform for all flow rates. In this paper, the method adapted was to impose a trend line to the measured cavitation noise information and to find out the deviation with respect to normal operating condition. It was concluded that detection of abnormality in pumps due to cavitation effects requires the current operating condition to be diagnosed first and then proper criteria for deviation in noise has to be imposed.

Numerical investigation into effects of gas concentration and bubble collapse on tip vortex cavitation noise of NACA16-020 wing

In this study, the effects of gas concentration and bubble collapse on tip vortex cavitation noise of NACA16-020 wings are investigated using coupled Eulerian-Lagrangian method based on sequential application of Reynolds averaged Navier-Stokes (RANS) solver, bubble dynamics model and acoustic analogy. The bubble dynamics model used in the preceding study (Ku et al., 2020) is modified by including the gas pressure terms and the bubble collapse model, which depends on the timing and threshold of bubble collapse, the number, initial radius and location of divided bubbles. The validity of the modified bubble dynamics model is confirmed through its application to a benchmark problem where single bubble is triggered by laser. Then, the coupled Eulerian-Lagrangian method based on the modified bubble dynamic model is applied for the prediction of tip-vortex cavitation noise of NACA16-020 wing. The predicted results of the tip vortex pattern and acoustic pressure spectrum are compared with the measured results, which shows closer agreements between two results than those in the previous study.

Methods for predicting the noise of transport vessels propellers

Object and purpose of research. The object of the research is the cavitation noise of transport vessels propellers. The purpose is to analyze and generalize modern research and methods for predicting broadband noise induced by the operation of the transport vessels cavitating propellers. Materials and methods. A review of the research materials (published in the XXI century) on the cavitation noise of transport vessels propellers, in particular broadband noise, as well as the author's own research on the causes and physical aspects of the occurrence of broadband noise is carried out. Based on the Fourier analysis of the time functions of velocities and pressures in the flow near the propeller, the conditions for the occurrence of broadband noise at (4–10)-blade harmonics and the parameters that relate this noise to various types of cavitation on the propeller are found. Main results. A review of recent studies results devoted to the cavitation noise of transport vessels propellers has shown that the levels of the so-called broadband noise that occurs when the cavitating propeller operates in nonuniform flow at high (4–10)-blade harmonics are used as the basis for modern noise predicting methods. At the same time, the authors of the methods consider broadband noise only as a result of the tip vortices development. The paper considers and generalizes an alternative hypothesis of the broadband noise occurrence caused by the occurrence and collapse of cavities within the blade with the formation of a double-headed peak on the time function of pressures. The analysis of publications on the development of the vortex wake behind the propeller and the modeling of pressures at the buckling failure of the tip vortex spirals allowed to show that double pressure peaks can be formed in the flow during the development of the vortex wake behind the propeller. This made it possible to combine both hypotheses of the broadband noise formation – from the development of the tip vortex and from the occurrence and collapse of cavities on the blades – and to link the assessment of the growth possibility of high blade harmonics with the cavitation parameters. Conclusions. It is shown that modern research methods have made it possible to obtain new data on the mechanism of the occurrence and collapse of cavitation both on the blade and in the tip vortex. At the same time, the currently used methods for predicting cavitation and, in particular, broadband noise are very approximate and require refinement to assess the effect of various cavitation characteristics on broadband noise. The hypotheses of broadband noise occurrence and the proposed physical model covering both vortex cavitation and the formation and collapse of cavities can serve as a basis for such clarification.

Classification of Surface Vehicle Propeller Cavitation Noise Using Spectrogram Processing in Combination with Convolution Neural Network

This paper proposes a method to enhance the quality of detecting and classifying surface vehicle propeller cavitation noise (VPCN) in shallow water by using the improved Detection Envelope Modulation On Noise (DEMON) algorithm in combination with the modified Convolution Neural Network (CNN). To improve the quality of the VPCN spectrogram signal, we apply the DEMON algorithm while analyzing the amplitude variation (AV) to detect the fundamental frequencies of the VPCN signal. To enhance the performance of the traditional CNN, we adapt the size of the sliding window in accordance with the properties of the VPCN spectrogram data, and also reconstruct the CNN layer structure. As for the results, the fundamental frequencies contented in the VPCN spectrogram data can be detected. The analytical results based on the measured data show that the accuracy of the VPCN classification obtained by the proposed method is above 90%, which is higher than those obtained by traditional methods.

Study of the relationship of cavitation erosion activity and cavitation noise intensity

The study of the erosion activity of cavitation is of considerable interest for clarifying the mechanism of the effect of cavitation on biological tissues and cells. This paper proposes an improved technique for assessing the erosion activity of acoustic cavitation. The results of testing this technique in relation to the problem of studying the distribution of erosion activity in the cavitation region, generated by a radiator with a rod waveguide, are presented. The experiments were carried out using a submersible emitter with a resonant frequency of 32 kHz. It was found that erosion activity rapidly decreases with distance from the emitter and depends on the distance to the emitter L as 1/L3 when the diameter of the emitter is less than or of the order of the wavelength in the used liquid. It was shown that there is a correlation between the erosion activity of cavitation and the readings of the cavitometer with the output signal being the integral intensity of the highfrequency component of the cavitation noise in the frequency range up to 10 MHz. Piezoelectric sensors were used to register cavitation noise. In particular, in liquids characterized by a higher level of erosion activity, the output signal of the cavitometer is also higher. In this case, the readings of the cavitometer change depending on the distance to the radiator as 1/L. Based on the data obtained, a method is proposed for assessing the erosion activity of cavitation by the magnitude of the intensity of cavitation noise in a cube. It is shown that this parameter is linearly related to the results of measurements of the erosional activity of cavitation. The results obtained will be used in the development of a specialized cavitometer designed to assess the erosion activity of cavitation during in vitro studies of the effect of ultrasound on cells.