Industrial Noise Environments Research Articles

A reference learning network for fault diagnosis of rotating machinery under strong noise

The strong noise often masks the fault characteristics of equipment, which reduces the accuracy of fault diagnosis and even leads to the inability of intelligent fault diagnosis algorithms to be applied in industrial environments. This has always been a challenge in the field of mechanical fault diagnosis. As known that equipment failure results from the continuous degradation of the equipment’s state, with the failure state evolving from the healthy state. Considering that both healthy signals and fault signals contain similar noise, this paper proposes a Reference Learning Network (RLNet) model. The model aims to enhance the distinguishing features between healthy and faulty samples through reference units, thereby eliminating the influence of noise on feature distribution. Firstly, the impact of variable speed on the model’s robustness is mitigated using the computed order tracking method. Then, the difference features between healthy samples and a class of fault samples are extracted through the binary classification reference learning unit (RLU). Next, the extracted differential features are used to train the state classifier. Finally, membership weights are employed to effectively combine the feature recognition results, reducing the influence of fault features from mismatched RLUs. The robustness and superiority of the proposed method were verified by comparing it with five other intelligent fault diagnosis methods on the gear and bearing datasets. RLNet is of great significance for the engineering application of intelligent fault diagnosis methods in industrial noise environments.

Assessing the multidimensional comfort of earplugs in virtual industrial noise environments

Earplugs’ comfort is primarily evaluated through cost-effective laboratory evaluations, yet these evaluations often inadequately capture the multidimensional comfort aspects due to design limitations that do not replicate real-world conditions. This paper introduces a novel laboratory method for comprehensive assessment of the multidimensional comfort aspects of earplugs, combining questionnaire-based evaluations and objective perceptual tests within virtual industrial sound environments replicating in-situ noise exposure. Objective perceptual results confirm that the sound environment affect participants’ ability to detect alarms in a noisy environment and comprehend speech-in-noise while wearing earplugs. Subjective questionnaire results reveal that the earplugs family has an effect on the primary attributes of the acoustical, physical and functional comfort’s dimension. Participants reported the physical dimension as the most important factor they take into account when evaluating earplugs’ comfort. The functional dimension was considered the second most important factor by the participants, followed by the psychological dimension, and the acoustical dimension.

Assessment of hearing loss risk due to impact noise in industrial environments

Impact noises are often found in industrial environments and they are predominant in mining, construction, factories, workshops, and shipyards. It is well-known that impact noises are more likely to cause noise-induced hearing loss than continuous noise of equal energy. Impulse noises are characterized by their high intensity over a short period of time and many countries have defined impulse noise exposure limits and criteria in occupational settings. They are based on the sound level measurements made using standard sound pressure level meters and dosemeters. However, because of their metrological limitations, it is not appropriate to use these instruments when dealing with such high peak levels and short duration times. Extensive research on hearing damage produced by impulse noise generated by firearms has been presented, mainly on police and military personnel. These studies have led to damage risk criteria contained in various versions of the standard MIL-STD-1474. Although industrial noises can reach similar peak sound pressure levels, not many results have been published on the subject. In this work, several common sources of industrial impact noise were measured in-situ, at the worker locations, using a specialized system equipped with high-dynamic-range microphones and a very high data acquisition rate. The signals were post-processed to obtain the main metrics defined for impulse noise exposure assessment. Then, occupational hearing loss risk was estimated using different criteria. It is shown that many common industrial processes reported a very high risk of impulsive noise to human hearing.

Measurement and Modeling of Electromagnetic Noise in Industrial Environments

Measurement and Modeling of Electromagnetic Noise in Industrial Environments

Noise Prediction in Industrial Workrooms Using Regression Modeling Methods Based on the Dominant Frequency Cutoff Point

Noise pollution is one of the major problems in industrial environments. The physiological response to the noise in industrial environments depends on the characteristics of the noise and environment. This study aimed to develop an empirical model for predicting the level of noise in closed industrial spaces using regression modeling based on the dominant frequency cutoff point. After identifying and determining the effective input variables in the prediction of noise level, the relevant data were collected from 56 industrial workrooms and the model was developed using multiple regression technique. The two models were best fitted to estimate the noise level for workrooms with a dominant frequency of less or equal to and more than 250 Hz (R2 = 0.86, R2 = 0.85, respectively). Based on the results of this study, it is less costly and requires less equipment for noise evaluation and monitoring by the mentioned models during the design, implementation, and operation of industrial environments. These experimental models can be used as suitable measures for screening closed industrial spaces and ranking them in terms of the amount of noise pollution.

Laser stripe extraction method in industrial environments utilizing self-adaptive convolution technique.

A line-structured laser scanner is widely applied for 3D reconstruction in industrial environments with ubiquitous various luminance, complicated background, diverse objects, and instable lasers. These elements will show up as noise in the obtained laser stripe images. Therefore, the basic and key point for a line-structured laser scanner is to accurately extract the laser stripe from noise. This paper proposes an effective laser stripe extraction procedure with two steps. First, a novel laser stripe center extraction method based on the geometry information and correlation in the laser stripe, is designed to significantly eliminate noise and accurately extract the laser stripe centers. In addition, the robustness, speed, and accuracy of this method are respectively analyzed in detail. Second, piecewise fitting is adopted to acquire a smooth and continuous laser stripe centerline. In order to select the optimal fitting method, the characteristics of two spline methods, Akima spline and cubic Hermite spline, are deeply analyzed and compared. Finally, an experiment is carried out by using a rough metal step and a line-structured laser scanning system. The experiment results demonstrate that the proposed self-adaptive convolution-mass method can significantly eliminate noise in industrial environments. In addition, the cubic Hermite spline is a better choice for 3D reconstruction, rather than the Akima spline.

A Novel Framework for Speech Signal Denoising using PSO Optimized ICA-DWT Algorithm

Objectives: In this paper a hybrid approach of ICA-DWT algorithm optimized using Particle Swarm Optimization (PSO) is proposed to deal with problems of aperiodic and period noises in industrial noise environment. Method/analysis: The feature of Independent Component Analysis (ICA) for separating the signals of various channels is exploited to separate noise peaks from the speech channel. To reserve the original signal and discern the noise, the speech is segmented in various levels of frequencies via discrete wavelet transform. The adaptive filtration through wavelet filters has been a powerful tool for signal segmentation into various frequencies. The output of ICA is sourced to Discrete Wavelet Transform (DWT) and is optimized using PSO to generate threshold value and number of wavelets for it. Findings: The results indicate that additional overhead computation of DWT has a better Signal-to-Noise Ratio (SNR) value compared to clean fast ICA algorithm and thus validate the improvement in speech signal intelligibility and quality.For the range of input signals and noise environment, the optimization of PSO to filter the speech signals has best SNR compared to conventional algorithm. Novelty/ Improvement: In the proposed denoising model two stages optimized filtering is presented. The number of wavelet levels and value of threshold is depicted using objective function to minimize Spectral Noise Density (SND). The objective function is optimized with PSO in constraints of SND to generate the best possible levels of DWT and thus maximize the SNR ultimately.

Using Cognitive Radio for Interference-Resistant Industrial Wireless Sensor Networks: An Overview

Industrial wireless sensor networks (IWSNs) have to contend with environments that are usually harsh and time-varying. Industrial wireless technology, such as WirelessHART and ISA 100.11a, also operates in a frequency spectrum utilized by many other wireless technologies. With wireless applications rapidly growing, it is possible that multiple heterogeneous wireless systems would need to operate in overlapping spatiotemporal regions. Interference such as noise or other wireless devices affects connectivity and reduces communication link quality. This negatively affects reliability and latency, which are core requirements of industrial communication. Building wireless networks that are resistant to noise in industrial environments and coexisting with competing wireless devices in an increasingly crowded frequency spectrum is challenging. To meet these challenges, we need to consider the benefits that approaches finding success in other application areas can offer industrial communication. Cognitive radio (CR) methods offer a potential solution to improve resistance of IWSNs to interference. Integrating CR principles into the lower layers of IWSNs can enable devices to detect and avoid interference, and potentially opens the possibility of utilizing free radio spectrum for additional communication channels. This improves resistance to noise and increases redundancy in terms of channels per network node or adding additional nodes. In this paper, we summarize CR methods relevant to industrial applications, covering CR architecture, spectrum access and interference management, spectrum sensing, dynamic spectrum access (DSA), game theory, and CR network (CRN) security.

Role of the kurtosis metric in evaluating hearing trauma from complex noise exposures—From animal experiments to human applications

A number of animal experiments and epidemiologic studies in humans have demonstrated that current noise standard underestimated hearing trauma by complex noise. While energy and exposure duration are necessary metrics they are not sufficient to evaluate the hearing hazard from complex noise exposure. The temporal distribution of energy is an important factor in evaluating NIHL. Kurtosis incorporates in a single metric all the temporal variables known to affect hearing (i.e., peak, inter-peak interval, and transient duration histogram) that makes kurtosis as one of the candidate metrics. Our previous animal studies show that both kurtosis and energy are necessary to evaluate the hazard posed to hearing by a complex noise exposure. In this study, we focus on how to use the knowledge from animal model into humans. Methods are presented to solve the following questions: (1) How to calculate the kurtosis? (2) What is the relation between kurtosis and the hearing trauma? (3) How to extract a single number from the distribution of kurtosis that would best correlate with noise trauma in real industrial noise environments? Human data study shows that the kurtosis metric may be a reasonable candidate for evaluating the risk of hearing trauma complex noise exposures.

Individual monitoring of hearing status: Development and validation of advanced techniques to measure otoacoustic emissions in suboptimal test conditions

Scientific evidence suggests that measurement of otoacoustic emissions (OAEs) is a reliable tool to detect the early onset of noise induced hearing loss. Nevertheless, the individual field measurement of otoacoustic emissions on industrial workers is very challenging in practice because the high level of ambient noise usually disturbs the OAE measurement. The use of OAE measurement probes with high passive noise isolation allows the attenuation of most of the high frequency ambient noise, but it is often insufficient for the low frequency content of the ambient noise. In the described research, a new type of OAE system, suitable for the continuous monitoring of OAE levels on an individual worker, has been designed as a pair of earpieces each featuring an external microphone, an internal microphone and a pair of miniature receivers. Adaptive Noise Reduction (ANR) processing the measured Distortion Product OAE (DPOAE) is used to further improve the signal-to-noise ratio in frequencies mostly where passive isolation remains insufficient. The ANR technique relies on a Normalized Least-Mean-Square (NLMS) algorithm that uses the ipsilateral external microphone and the contralateral internal microphone to denoise the measured DPOAE signals for each in-ear OAE probe. A side-by-side comparison with commercially available clinical OAE equipment on 8 test-subjects successfully confirmed that the developed OAE system would be suitable for the continuous monitoring of workers’ hearing capabilities in industrial noise environments with levels up to 75dB(A).

Effect of cigarette smoking on noise-induced hearing loss in workers exposed to occupational noise in China

Excessive exposure to high noise level environments has the potential to cause noise-induced hearing loss (NIHL), and cigarette smoking has also been shown to have a potential adverse effect on hearing. The aim of this study was to determine whether smoking interacts with noise in the development of hearing loss, and if so, the extent of the contribution from smoking on NIHL. A cross-sectional study was designed to assess the effect of smoking on NIHL in 517 male workers (non-smokers: N = 199; smokers: N = 318) exposed to a high-level industrial noise environment in China. Shift-long temporal waveforms of the noise that workers were exposed to for evaluation of noise exposures, and audiometric threshold measures were obtained on all selected subjects. The subjects used hearing protection devices only within the last 1-2 years. The results suggest that smoking has an adverse effect on NIHL in workers exposed to high level industrial noise, i.e., the median high frequency hearing thresholds were significantly greater in smokers than non-smokers exposed to noise for more than 10 years. This effect was observed at 4.0 and 6.0 kHz. Smoking did not have an adverse effect on NIHL in workers exposed to noise less than 10 years. Multivariate regression analysis revealed that the odds ratio (OR) for high frequency hearing loss (i.e., hearing threshold greater than 40 dB at 4.0 kHz) were 1.94 for smokers in comparison to non-smokers. The results suggest that: (1) smokers have a higher risk of developing high frequency hearing loss than non-smokers with a similar occupational noise exposure, and (2) the interaction between cigarette smoking and high-level noise exposure may be additive. There is a need to develop and analyze a larger database of workers with well-documented exposures and smoking histories for better understanding of the effect of smoking on NIHL incurred from high-level industrial noise exposures. A better understanding of the role of smoking may lead to its incorporation into hearing risk assessment for noise exposure.

The use of the kurtosis metric in the evaluation of occupational hearing loss in workers in China: Implications for hearing risk assessment

This study examined: (1) the value of using the statistical metric, kurtosis [β(t)], along with an energy metric to determine the hazard to hearing from high level industrial noise environments, and (2) the accuracy of the International Standard Organization (ISO-1999:1990) model for median noise-induced permanent threshold shift (NIPTS) estimates with actual recent epidemiological data obtained on 240 highly screened workers exposed to high-level industrial noise in China. A cross-sectional approach was used in this study. Shift-long temporal waveforms of the noise that workers were exposed to for evaluation of noise exposures and audiometric threshold measures were obtained on all selected subjects. The subjects were exposed to only one occupational noise exposure without the use of hearing protection devices. The results suggest that: (1) the kurtosis metric is an important variable in determining the hazards to hearing posed by a high-level industrial noise environment for hearing conservation purposes, i.e., the kurtosis differentiated between the hazardous effects produced by Gaussian and non-Gaussian noise environments, (2) the ISO-1999 predictive model does not accurately estimate the degree of median NIPTS incurred to high level kurtosis industrial noise, and (3) the inherent large variability in NIPTS among subjects emphasize the need to develop and analyze a larger database of workers with well-documented exposures to better understand the effect of kurtosis on NIPTS incurred from high level industrial noise exposures. A better understanding of the role of the kurtosis metric may lead to its incorporation into a new generation of more predictive hearing risk assessment for occupational noise exposure.

The use of the kurtosis metric in the evaluation of industrial noise exposures.

Data from a chinchilla model of noise-induced hearing loss will demonstrate the value of using the statistical metric, kurtosis, along with an energy metric to evaluate complex industrial noise environments for hearing conservation purposes. Complex noises are non-Gaussian noises consisting of a combination of Gaussian continuous noise and noise transients, either impacts or noise bursts. Chinchillas were exposed to an interrupted and intermittent 97-dB sound pressure level broadband, non-Gaussian noise for 8 h/day for 3 weeks. The experimental noises were designed to model an industrial exposure. Groups of animals with 8/group were exposed to 1 of 12 different complex noise environments having a fixed spectrum and energy but with different temporal structures achieved by varying the transient peak, interval, and duration histograms. Preliminary results show that (1) hearing and sensory cell loss increase with increasing kurtosis despite a fixed energy; (2) at a given value of kurtosis, hearing trauma is invariant with respect to changes in the temporal structure of the noise exposure. A metric such as the kurtosis may be a necessary adjunct to spectral energy in the evaluation of industrial noise exposures for the purposes of hearing conservation. [Research supported by NIOSH Grant R01 OH 002317.]

Role of the Kurtosis Statistic in Evaluating Complex Noise Exposures for the Protection of Hearing

To highlight a selection of data that illustrate the need for better descriptors of complex industrial noise environments for use in the protection of hearing. The data were derived using a chinchilla model. All noise exposures had the same total energy and the same spectrum; that is, they were equal energy exposures presented at an overall 100 dB(A) SPL that differed only in the scheduling of the exposure and the value of the kurtosis, beta(t), a statistical metric. Hearing thresholds were determined before and after noise exposure using the auditory-evoked potential measured from the inferior colliculus in the brain stem. Cochlear damage was estimated from sensory-cell counts (cochleograms). (1) For equivalent energy and spectra, exposure to a high-kurtosis, non-Gaussian noise produced substantially greater hearing and sensory-cell loss in the chinchilla model than a low-kurtosis, Gaussian noise. (2) beta(t) computed on the amplitude distribution of the noise could clearly differentiate between the effects of Gaussian and non-Gaussian noise environments. (3) beta(t) can order the extent of the trauma as determined by hearing thresholds and sensory-cell loss. The noise level in combination with the statistical properties of the noise quantified by beta(t) clearly differentiate the effects between both continuous and interrupted and intermittent Gaussian and non-Gaussian noise environments. For the same energy and spectrum, the non-Gaussian environments are clearly the more hazardous. The use of both an energy and kurtosis metric can better predict the hazard of a high-level complex noise than the use of an energy metric alone (as is the current practice). These results point out the need for a new approach to the analysis and quantification of industrial noise for the purpose of hearing conservation practice.

Performance of Different Types of Hearing Protectors Undergoing High-Level Impulse Noise

The paper describes the problems that may occur when hearing protectors, usually designed for industrial noise environments, are used for high-level impulse (weapon) noise. The military impulse noise environment is described, as are the different types of passive and active hearing protectors and the measurement procedures. The different mechanisms that may alter the effectiveness of the hearing protectors as well as their global efficiency when submitted to high-level impulse noise are presented. The paper also discusses how the performance values accessible to the user may be used in different damage risk criteria for continuous and impulse noise.

Laboratory synthesis of industrial noise environments with predetermined statistical properties

High‐level nonGaussion noise is commonly found in a variety of industrial environments. Recent experiments have shown that for a given energy level, the statistical properties of a noise can have a strong effect on the extent of hearing loss produced in exposed individuals. In order to study, in an animal model, the effects on hearing of such noise environments, the statistical properties of the noise as embodied in the kurtosis metric must be under experimental control. For a fixed value of kurtosis and energy level the following four variables will have a strong effect on hearing loss: (1) peak histogram; (2) interval histogram; (3) duration of noise transients; and (4) level of any background Gaussian noise. Simulations have shown that the relations among kurtosis and these variables are nonlinear. However, under certain restricted conditions, these relations may be linear. Accordingly, two strategies for designing controlled industrial noise exposures are presented: (1) the interval‐priority model and (2) the duration‐priority model. Computer simulations and measurements of actual acoustic environments showed that these two models could be effectively used to simulate a wide variety of realistic industrial noises.

Active control of noise radiated through rectangular plates using piezeletric patches

Due to problems caused by noise in industrial environment and in human daily life, techniques of noise control have received increasing attention from engineers and researchers lately. More recently, the use of piezeletric elements as sensors and/or actuators in noise and vibration control systems has been extensively investigated. The main advantage of the use of such devices is that they can be easily integrated to the mechanical system with little added mass and relatively high control authority. The present paper addresses a technique of active control of sound transmitted through a rectangular, thin, simply supported plate by employing multiple piezeletric patches bonded to the plate’s surface. A harmonic plane wave incident on one side of the plate is considered to be the primary noise source. Aiming at minimizing the noise transmitted to the other side of the plate, bending motion is induced through the piezeletric patches so that the plate behaves as a secondary sound source. The paper brings the development of the system mathematical model which enables to obtain the spatial distribution of sound pressure radiated through the plate in the far field. An optimal control technique providing the voltage control signals for the activation of the piezoelectric patches is presented, based on the minimization of a cost function representing the mean square integral of the sound pressure radiated in a semi-sphere in far field. It is also proposed a methodology for the optimal placement on the piezelectric patches using Genetic Algorithms. (To be presented in Portuguese.)

Complex noise exposures: an energy analysis.

Industrial noise environments usually present a complex stimulus to the exposed individual. These environments often contain mixtures of multiply reflected impact noises and a relatively Gaussian broadband noise. Noise exposure standards do not consider the possibility of interactions between the two classes of noise that can exacerbate the amount of hearing trauma. This paper presents the results of a large series of experiments designed to document the hazard posed to hearing from complex noise exposures. Twenty-three groups of chinchillas with 5 to 11 animals per group (total N = 135) were exposed for 5 days to either octave bands of noise, impacts alone, or combinations of impact and octave bands of noise. Evoked potential measures of hearing thresholds and cochleograms were used to quantify the noise-induced trauma. The results show that, for sound exposure levels (SEL) which produce less than approximately 10 dB PTS (permanent threshold shift) or 5% total sensory cell loss, equal-energy exposures tend to produce equivalent effects on hearing. However, there is a range of at least 10 dB in the SEL parameter where hearing loss from equal-energy exposures at a particular SEL can be exacerbated by increasing the repetition rate of the impacts or by the addition of a Gaussian low-level noise. The exacerbation of trauma from the addition of a Gaussian continuous noise is dependent upon the spectrum of that noise.

The use of frequency domain kurtosis in the assessment of industrial noise environments for hearing conservation purposes.

Industrial noise environments are frequently characterized by combinations of noise impacts superimposed on a steady-state continuous noise background. Digital techniques were used to synthesize complex noise environments containing random impulses with random peak SPL’s up to 125 dB. Four groups of chinchillas were exposed to one of three different complex noise environments whose sound exposure levels and spectra were similar to a fourth Gaussian noise exposure. Conventional methods were used to measure energy spectra and exposure levels. Measures of permanent threshold shift showed significant differences between the complex noise exposures and the Gaussian exposure. An application of frequency domain kurtosis [R. F. Dwyer, IEEE J. Oceanic Eng. OE-9, 85–92 (1984)] emphasized the region of the noise spectrum where audiometric differences were measured. [Work supported by NIOSH.]

Kurtosis measurements: Implications for noise exposure criteria

Two conventional measures of an industrial noise environment are the overall sound-pressure level (SPL) and the power spectrum. While these two measures are necessary, they may not be sufficient to describe a noise environment for the purposes of evaluating the hazard to hearing. This presentation will describe the results of an experiment designed to test the hypothesis that for equal SPL and power spectrum a high-kurtosis noise exposure is more hazardous to hearing than is a low-kurtosis noise. Two groups of chinchillas, with two animals per group, were exposed continuously for 5 days, to one of two classes of noise at a 90 dB SPL. The power spectrum of both noises was identical but the noises had different values of kurtosis. The results clearly showed that there was up to a 20 dB greater permanent hearing loss for the animals exposed to the high-kurtosis noise. Detailed results of asymptotic and permanent threshold shifts and sensory cell losses will be presented. These results would appear to be a clear indication that, even for moderate levels of exposure, energy is not a sufficient variable upon which to base noise exposure standards. [Work supported by NIOSH.]