Noise Components Research Articles

Methods for separating the noise produced by the wheels and track during a train pass-by

AbstractRolling noise is produced by vibration of the wheels and track, induced by their combined surface roughness. It is important to know the relative contributions of the different sources, as this affects noise control strategies as well as acceptance testing of new rolling stock. Three different techniques are described that aim to use pass-by measurements to separate the wheel and track components of rolling noise. One is based on the TWINS model, which is tuned to measured track vibration. The second is based on the advanced transfer path analysis method, which provides an entirely experimental assessment. The third is based on the pass-by analysis method in combination with static vibroacoustic transfer functions which are obtained using a reciprocity method. The development of these methods is described and comparisons between them are presented using the results from three experimental measurement campaigns. These covered a metro train, a regional train and a high-speed train at a range of speeds. The various methods agree reasonably well in terms of overall trends, with moderate agreement in the mid-frequency region, and less consistent results at low and high frequency.

End-to-End Intelligent Fault Diagnosis of Transmission Bearings in Electric Vehicles Based on CNN

Environmental noise and transmission components can cause significant interference in vibration signals, rendering the extraction of bearing fault features challenging in service scenarios. Traditional fault diagnosis methods rely heavily on professional domain knowledge, prior models, and signal preprocessing methods. The accuracy of fault diagnosis depends on the quality of the fault-sensitive features extracted by vibration signal preprocessing methods. An improved convolutional neural network (CNN) end-to-end intelligent fault diagnosis model based on raw vibration data (RVDCNN) is proposed. The time-domain vibration signal of the transmission bearing is converted into a continuous two-dimensional numerical matrix, and a two-dimensional CNN model is constructed through network structure optimization. The original time-domain vibration signal numerical matrix of the bearing is trained and tested to extract and learn abstract fault features of different fault types, and then the fault classification of the bearing is achieved. To verify the generalizability of the RVDCNN intelligent fault diagnosis model, it is applied to the recognition of rolling bearings in the two-speed mechanical automatic transmission of electric vehicles, achieving recognition accuracy of 99.11% for seven types of bearings.

S&P 500 microstructure noise components: empirical inferences from futures and ETF prices

By studying the differences between futures prices and exchange‐traded fund prices for the S&P 500 index, original results are obtained about the distribution and persistence of the microstructure noise component created by positive bid‐ask spreads and discrete price scales. The bivariate density of this component for futures and exchange‐traded fund prices is estimated from high‐frequency prices, to provide estimates of the marginal noise densities and measures of noise dependence across the markets studied. Properties of the residual microstructure noise, created by factors other than discrete prices, are also estimated. The residual component has more variation and less persistence than the discrete‐price component during the period examined, from January 2010 to December 2012.

Vital signs detection of moving targets using FMCW radar

Abstract Respiratory and heartbeat rates are crucial indicators for human health assessment. Compared to contact-based measurements, millimeter-wave radar detection of these vital signs avoids the discomfort caused by physical contact and better protects personal privacy, making it highly valuable for home health monitoring. The use of millimeter-wave radar for vital sign detection of the human body is mostly focused on targets in a stationary state at present. However, the human body may sway or even move during actual detection. This article proposes a non-contact vital sign detection method for moving targets. Compared with methods for detecting vital signs of stationary targets, detecting vital signs of moving targets requires determining the range bin where the targets are located continuously and extracting target phase information. The noise components such as movement and sway contained in the phase signal need to be removed. In this paper, moving target indication is used to remove static components, an adaptive range bin selection method is proposed to determine the range bin where the targets are located, and range bin selection fluctuation is smoothed using a moving average filter. The wavelet transform is used to decompose the phase signal, remove swaying noise based on autocorrelation function, and reconstruct the life signal for different scale factors. A bandpass filter is used to separate the respiratory and heartbeat signals, and a notch filter is designed to suppress respiratory harmonic signals. The experimental results show that the proposed method can separate vital signs signals from the phase signals of moving targets, achieving detection of respiration and heartbeat rate. The average accuracy of respiration and heartbeat rate detection is 94.7% and 95.5%, respectively.

Rotor noise in non-axial inflow conditions

This paper presents an experimental study on characterising the aerodynamic and far-field acoustic performance of rotors operating in non-axial inflow conditions. The current experiments were carried out with a two-bladed rotor, operating with a tip Mach number of 0.56, approaching the practical tip Mach numbers expected of urban air mobility vehicles. Synchronous measurements of the aerodynamic force, far-field acoustics and blade phase angle were undertaken to provide insights on both the time-averaged and phase-averaged correlation between the rotor aerodynamic loading and the emitted noise. The far-field acoustic spectra are compared across a range of advance ratios (0.01≤μ≤0.04) and forward tilting angles (0°≤α≤30°). When the inflow velocity increases, both the tonal and broadband components increase. Moreover, higher harmonics of the blade passing frequency are observed to become more prominent at higher advance ratios. As the rotor tilts forward from edgewise flight conditions, the broadband noise component reduces significantly, with the tones at fundamental blade passing frequency and higher harmonics also reducing in magnitude, agreeing well with the steady and unsteady thrust coefficients measured. From the aerodynamic loading, acoustic spectra and the directivity results, a transition from near-edgewise to near-axial modes of operation can be discerned at a tilt angle of approximately 16°. More interestingly, the phase-averaged thrust and acoustic power results show notable variations of the blade loading and acoustic emission through one full rotor rotation. At higher advance ratios and shallow tilting angles, the phase location of peaks in the root-mean-square of the fluctuating thrust coefficient agrees very well with those from the acoustic power, suggesting a strong correlation between the flow unsteadiness and the noise at these non-axial conditions, contributing to higher overall sound pressure levels.

XU-NetFullSharp: The Novel Deep Learning Architecture for Chest X-ray Bone Shadow Suppression

Background and objectivesChest X-ray image (CXR) is vital for screening, preventing, and monitoring various lung diseases. In particular, the early detection of lung cancer can significantly improve patients’ chances of survival and quality of life. Unfortunately, approximately 82–95 % of missed pulmonary nodules are estimated to be obscured by rib shadows, making them difficult to recognize. This study addresses this problem by considering the rib shadows in CXRs as noise that can be reduced using deep learning. The result of the proposed model is a CXR with improved clarity for easier and more accurate analysis by radiologists or computer algorithms. MethodsAn automated deep learning-based model for bone shadow suppression from frontal CXRs, called xU-NetFullSharp, was proposed. This network is inspired by the most modern U-NetSharp architecture and was modified using different approaches to preserve as many details as possible and accurately suppress bone shadows. For comparison, recent state-of-the-art models were implemented and trained. JSRT, VinDr-CXR, and Gusarev DES datasets were utilized for the experiments, where the JSRT dataset was extensively augmented. ResultsThe performance of the proposed xU-NetFullSharp was analyzed using statistical measures and compared with that of other architectures. The proposed model significantly outperformed the others, reaching the best values of the most used metrics (0.9846 SSIM; 0.9870 MS-SSIM). It also achieves a correlation of 96.31 % and an intersection of 10.0285 between the predicted and ground truth histograms, together with the smallest value of the Bhattacharyya distance. The obtained results were validated by experts from the University Hospital Olomouc with positive feedback, thus achieving the best objective and subjective results. The proposed method has the potential to be implemented in hospital environments. ConclusionA comprehensive comparison of the proposed architecture with state-of-the-art methods proves its efficiency in suppressing noise in CXRs and its ability to distinguish the signals of important tissues from noise components. This methodology can potentially improve the performance of the existing CXR processing methods. The source code is released in a GitHub repository that can be accessed from the following link: https://github.com/xKev1n/xU-NetFullSharp.

Transient flow and noise characteristics of vortex-turbulence-noise interaction in centrifugal pump

Centrifugal pump operation noise is an avoidable common phenomenon in engineering practices, and its characteristics are closely influenced by the flow characteristics of the fluid. In this paper, a numerical simulation of centrifugal pump operation with different flow rates is carried out using the detached eddy simulation turbulence model and compared with the experimental results. Then the noise source and far-field radiated noise characteristics of the centrifugal pump are investigated using Lighthill’s analogy. The results show that the noise of centrifugal pump is closely related to the flow characteristics of impeller, especially the generation and development of vortex. The relationship between turbulent kinetic energy (TKE) and vortex is complex, involving the generation, development and shedding of vortex. The influence of flow rate variation on vortex structure in impeller is studied by vortex transport equation. It is found that the relative vortex stretching (RVS) term is the main driving force of vortex distribution. The distribution of the Coriolis force (CORF) term is mainly concentrated in the inlet and outlet areas of the flow channel, which is related to the rotational motion of the impeller and the dynamic characteristic of the fluid. The power spectral density inside the impeller is mainly concentrated in the low frequency region, indicating that the low frequency noise component is dominant in energy. The directivity curve of far-field noise pressure level is influenced by hydrodynamic effect, impeller rotation direction and pump design structure, and will have a certain deviation. The spectral curve analysis of velocity, vorticity, pressure and sound pressure level shows that these physical quantities have strong correlation on the spectrum, in which the low frequency component is large, and the high frequency component is complex and volatile.

Mechanisms of multi-layered Rayleigh noise in Brillouin optical correlation-domain reflectometry

In Brillouin optical correlation-domain reflectometry (BOCDR), sinusoidal modulation is applied to the output frequency of a light source, with spatial resolution inversely related to the modulation amplitude. We have developed an effective method to estimate the modulation amplitude using the width of the noise spectrum caused by Rayleigh scattering, eliminating the need for an optical spectrum analyzer or modifications to existing equipment. However, the Rayleigh noise spectrum often displays a three-layered structure, complicating the identification of the appropriate spectral components for estimating the modulation amplitude. In this work, we investigate the origins of this three-layered Rayleigh noise spectrum and identify the directivity of an optical circulator as the source of the third noise component. As replacing the circulator with alternative optical components is not easy, it remains an essential part of the system. Our analysis shows that the third noise component exhibits significantly small variation in spectral width with changes in modulation frequency compared to the first and second components. This characteristic allows for the effective separation and identification of the third noise component, thereby enhancing the accuracy and convenience of modulation amplitude estimation in BOCDR.

EI-ISOA-VMD: Adaptive denoising and detrending method for nuclear circulating water pump impeller

As a key component of the nuclear circulating water pump, it is difficult to extract effective information from low-quality impeller signal because of the interference noise and non-stationary trend, such as water flow impact, vibrations from unrelated components, and sensor noise. Traditional signal denoising methods, particularly those utilizing variational mode decomposition, often require manual parameter tuning, resulting in unsatisfactory outcome. To address these issues, this paper proposes a signal denoising and detrending method based on the evaluation index driven the improved seagull algorithm optimized for variational mode decomposition (EI-ISOA-VMD). During the optimization process, sinusoidal chaotic mapping is utilized for initializing the seagull population, while a tanh function is employed to design nonlinearly decreasing inertia weights. Furthermore, the Levy flight mechanism enhances the randomness of position updates and optimizes seagull individuals beyond the boundary range in order to form the improved seagull optimization algorithm. The denoising, detrending process, and evaluation index proposed in this paper are utilized as the fitness function for the improved seagull optimization algorithm to optimize key parameters of variational mode decomposition. During the denoising and detrending process, the trend component is initially selected using the mean ratio method. The useful, low noise and high noise intrinsic mode functions are screened by correlation coefficient and weighted permutation entropy. Subsequently, the low noise components are denoised by wavelet thresholding method, and the high noise components are discarded to finally reconstruct the signal. The experimental results demonstrate that the proposed method effectively eliminates noise and trend components in low-quality signal while preserving more valuable information.

Impact of road noise components (rolling/propulsion) by vehicle categories (light-duty, heavy-duty and two-wheeled vehicles) on population's noise exposure and associated adverse health effects

Exposure to road traffic noise can lead to various adverse health effects, ranging from annoyance and night-time sleep disturbances to ischemic heart diseases. The aim of this paper is to identify the contribution of the road noise components (rolling/propulsion) for each vehicle category (light-duty, heavy-duty and two-wheeled vehicles) on population's noise exposure and the associated adverse health effects. This analysis focuses on the case study of Lyon, France. The input data align with the requirements of the 4th deadline of Directive 2002/49/CE and adhere to the CNOSSOS calculation standard. The results reveal that the rolling noise component of light-duty vehicles is predominant due to vehicle speed and traffic volume. It is the sole noise component (i) significantly exposing a portion of the population to Lden > 68 dBA, noise limit set by the French decree of April 4, 2006, and (ii) where over half of the population is exposed to Lden > 53 dBA, recommended threshold by the World Health Organization (2018). By using transfer functions established between noise components for each vehicle category, and population's noise exposure, noise source-abatement solutions can be quantified in terms of their impact on population's noise exposure, thus determining the reduction in health impacts.

Methods to estimate efficiency of Low Noise Pavement for specific vehicle categories

Low Noise Pavements (LNPs) are efficient in reducing the rolling noise component of road-tire noise generation mechanisms. Open (porous) asphalt can be also efficient in reducing the propulsion noise component. As rolling/propulsion noise ratio varies with vehicles' category, the mitigation effect of a LNP varies between categories as confirmed by experimental data gathered in the LIFE NEREIDE. A proper model needs to predict and simulate the effects of LNPs in real conditions in the implementation sites. At present, LNPs can be modeled according to CNOSSOS-EU methodology using available reference surfaces, or by tuning user-defined surfaces. The paper discusses the limitations of mapping methods in delivering reliable values for specific vehicle categories, even when using ad-hoc tuned surfaces. The results are based on case studies in which modeling results are compared to experimental data. The study found that differences in measurements are due to the way the source model is conceived and cannot be completely eliminated.

CNOSSOS-FR: Are current road traffic noise emission coefficients still representative of today's vehicle fleet and roads?

CNOSSOS-EU is the common noise prediction method for producing strategic noise maps in Europe, according to Directive 2002/49/CE. Its road vehicle noise emission model involves a propulsion noise component and a rolling noise component, with coefficients specific to each vehicle category. As currently applied in France, it relies on the original EU coefficients published in the 2015 CNOSSOS-EU version and the rolling noise contribution is detailed in road surface classes in agreement with the national NMPB2008 method. As provided in the European method, corrective coefficients in frequency bands were determined per road surface class. The objective was to minimise the overall difference between NMPB2008 and CNOSSOS-FR emission predictions under similar conditions, the former method being previously used for periodic mapping. Now considering the later release of new EU reference coefficients and the difficulty of making the two methods involving different expressions and assumptions coincide, as well as questioning the need to update emission values resulting from measurements of the vehicle fleet taken 20-25 years ago, the study MINOOS is being carried out to assess the noise emission of the current vehicle fleet and reconsider French prediction coefficients accordingly. The paper describes the procedure and presents the first results.

Dual-stream Noise and Speech Information Perception based Speech Enhancement

In real-world scenarios, dynamic ambient noise often degrades speech quality, highlighting the need for advanced speech enhancement techniques. Traditional methods, which rely on static embeddings as auxiliary features, struggle to address the complexities of varying noise conditions. To overcome this, we propose a Dual-stream Noise and Speech Information Perception (DNSIP) approach that dynamically detects and processes both noise and speech through innovative information extraction and suppression mechanisms. Initially, non-speech segments predominantly contain environmental noise, while speech segments carry information about the intended speaker. To handle this dynamic nature, real-time voice activity detection (VAD) is employed to accurately differentiate between speech and noise components. Building on VAD estimates, we propose an innovative information extraction framework that selectively extracts relevant noise and speech features from the noisy input, establishing a dual-stream network for concurrent noise and speech learning. To account for the temporal and spectral variability of noise and speech, a frequency-sequence attention mechanism is integrated, enhancing the model’s ability to learn contextual and spectral dependencies. Additionally, an information suppression module is introduced to minimize cross-stream interference by attenuating noise within the speech stream and suppressing speech content within the noise stream. The derived noise and speech spectrograms are then utilized to formulate a minimum mean square error log-spectral amplitude (MMSE-LSA) estimator for robust speech enhancement. Experimental evaluations on the WSJ0 and VCTK+DEMAND datasets demonstrate that our DNSIP approach surpasses existing state-of-the-art methods, underscoring its efficacy in challenging acoustic environments.

Improvement method for sound quality of pharyngeal speech by using Bayes' theorem

In general, a speech signal can be measured by a microphone and a pharyngeal microphone etc.. However, speech signal measured by a microphone often contains the background noise. On the other hand, though a pharyngeal microphone is effective for background noise, it involves body conducted noise. In this study, we propose an improvement method for sound quality of speech signal measured trough a pharyngeal microphone to achieve well the speech recognition. The relationship between the original speech signal and the speech measured by the pharyngeal microphone is not clear. Therefore, we consider the relationship as multiplicative and additive models of the original speech signal and noise components with unknown parameters. An algorithm to estimate simultaneously the original speech signal and the unknown parameters is proposed by using Bayes' theorem based on the measured speech through the pharyngeal microphone. Finally, a speech recognition experiment is conducted to confirm the effectiveness of the proposed algorithm.

The impact of active components wear on psychoacoustic parameters of noise in a car cabin

The automotive industry is undergoing an electromobility revolution. Changing a car's drive unit influences the vibroacoustics of a car's interior. Drive-train noise at low car speeds and in the low-frequency range is no longer dominant. There is no masking effect due to the noise of the combustion engine, and a passenger paradoxically may perceive the noise of a car as less pleasant. Moreover, existing vibroacoustic requirements have to some extent taken into account the change in the noise of components with the usage time, but there are still areas that require further research. This paper focuses on the study of electric adjustable seats, power windows and power exterior mirrors in different wear states, to determine the significant psychoacoustics changes of the selected car component's noises. The study is part of extensive research that aims to revise the current vibroacoustic requirements for new components and to determine how they could be tested and evaluated.

Numerical evaluation of rail track noise and vibration and components optimization

In the context of a project commissioned by the Swiss Federal Office for the Environment, an open-source modelling toolbox was developed to simulate and analyze rail tracks dynamics. In particular, a vibro-acoustic multi-sleeper model, which is based on advanced finite element modelling techniques, was used to investigate the effects of components properties on track noise and vibration. Considering changes around a reference construction (60E2 rail, SBB hard pad, B91 sleeper), parametric studies were performed on the dominant parameters identified for the acoustic and ballast protection performance indicators defined. The following sweet spot for the component properties of the track have been identified as using a rail pad of medium stiffness (1000 kN/mm at 1kHz, 300 kN/mm static stiffness) and high damping (at least 0.5) combined with soft Under Sleeper Pads (250 kN/mm). Based on the present models, such combination could provide both a significant noise reduction and a marked improvement in ballast protection compared to current track construction.

Vibration and noise reduction on a cover for power electronics for electric vehicles with vibroacoustic metamaterials

Electric vehicles represent a particular challenge in terms of noise, vibration and harshness, since there is no masking of the combustion engine anymore. For consumer acceptance, long distances need to be covered with electric vehicle, which makes lightweight structures necessary. Consequently, vibration problems occur during operation, as lightweight structures are more susceptible to vibrations. Common measures for vibration and noise reduction add a considerable amount of mass to the vehicle, counteracting the lightweight efforts. Vibroacoustic metamaterials as a novel approach promise greater vibration reductions than common measures like Alubutyl or bitumen and are accompanied by a lower weight. This study deals with the design of vibroacoustic metamaterials for the noise and vibration reduction of a cover for power electronics of an electric vehicle. The design process is demonstrated for two different kinds of vibroacoustic metamaterials made from sheet metal and a plastic film. The structural dynamic behavior of both approaches is experimentally investigated and compared with a cover equipped with Alubutyl. The vibroacoustic metamaterial, which is made from sheet metal, is furthermore tested in a vehicle under driving condition. A reduction of tonal noise components by 11 dB is achieved within the passenger compartment.

Prediction and evaluation of the low-frequency noise impact from industrial noise sources

German legislation requires an assessment of low-frequency noise in addition to the regular noise assessment for industrial plants. The detailed prediction and assessment of low-frequency noise immissions is carried out according to the administrative regulation for noise TA Lärm and the German standard DIN 45680. However, a standardised procedure for predictions of low-frequency noise does not exist. For this reason, different approaches for the prediction of low-frequency noise immissions have been developed over the last two decades. Typically, the method according to DIN ISO 9613-2 for sound propagation outside is extended heuristically to the lower frequency range. The predicted low-frequency noise level outside the room is taken as a basis for determining the noise immission level inside the room. The level of knowledge about the quality of sound propagation modeling of low-frequency noise is still debatable. A forecast method of low-frequency noise components should be on the "safe side" in terms of noise control; however, the scope of the required measures to meet the noise requirements must be comprehensible and proportionate. We present an overview of the building blocks of a forecasting process and the challenges that arise. The topic is illustrated by a practical example.

Comprehensive investigations on spectral and temporal features of GX 5−1 using AstroSat observations

ABSTRACT Comprehensive spectrotemporal analyses of the Z-type neutron star low-mass X-ray binary GX 5−1 were performed using 10 broad-band observations from AstroSat/Soft X-ray Telescope and Large Area X-ray Proportional Counter (LAXPC) instruments. The LAXPC-20 hardness–intensity diagram showed horizontal and normal branches (HBs and NBs) of the Z track which exhibited secular motion. The time-averaged spectra in the energy range 0.7–25.0 keV could be fitted with the model combination – $\tt {constant}\, \times \, \tt {tbabs}\, \times \, \tt {edge}\, \times \, \tt {edge}\, \times \, \tt {thcomp}\, \times \, \tt {diskbb}$. This yielded $\Gamma \, \sim$ 2, $kT_{\mathrm{ e}}\, \sim$ 3.3 keV, and $F_{\mathrm{ disc}}$/$F_{\mathrm{ total}}\, \sim$ 0.8 indicating the soft/intermediate spectral state of the source during the observations. Flux-resolved spectral analysis revealed a positive correlation between $kT_{\mathrm{ in}}$ and $F_{\mathrm{ bol}}$. However, a negative correlation was observed between them in one of the NBs. Time-averaged temporal analysis revealed multiple HB oscillations (HBOs) and NB oscillations (NBOs), and peaked noise components in the $\sim$5–50 Hz range. Furthermore, flux-resolved temporal analysis showed that the frequency of the HBOs correlates positively whereas the strength of HBOs correlates negatively with $F_{\mathrm{ bol}}$, indicating their probable origin from the accretion disc. In contrast, the frequency and strength of NBOs remain fairly constant with $F_{\mathrm{ bol}}$, suggesting that they originate from a different region in the system. Using the relativistic precession model along with highest frequency of the HBO, the upper limits of the magnetic dipole moment ($\mu$) and field strength (B) at the poles of the neutron star in the system were found to be 25.60$\times \, 10^{25}$ G cm3 and 3.64$\times \, 10^{8}$ G, respectively, for $k_{\mathrm{ A}}$ = 1.

WPConvNet: An Interpretable Wavelet Packet Kernel-Constrained Convolutional Network for Noise-Robust Fault Diagnosis.

Deep learning (DL) has present great diagnostic results in fault diagnosis field. However, the poor interpretability and noise robustness of DL-based methods are still the main factors limiting their wide application in industry. To address these issues, an interpretable wavelet packet kernel-constrained convolutional network (WPConvNet) is proposed for noise-robust fault diagnosis, which combines the feature extraction ability of wavelet bases and the learning ability of convolutional kernels together. First, the wavelet packet convolutional (WPConv) layer is proposed, and constraints are imposed to convolutional kernels, so that each convolution layer is a learnable discrete wavelet transform. Second, a soft threshold activation is proposed to reduce the noise component in feature maps, whose threshold is adaptively learned by estimating the standard deviation of noise. Third, we link the cascaded convolutional structure of convolutional neutral network (CNN) with wavelet packet decomposition and reconstruction using Mallat algorithm, which is interpretable in model architecture. Extensive experiments are carried out on two bearing fault datasets, and the results show that the proposed architecture outperforms other diagnosis models in terms of interpretability and noise robustness.