• The Journal of the Acoustical Society of Korea
• /
• v.42 no.6
• /
• pp.603-616
• /
• 2023

In this study, machine learning-based clustering and classification of residential noise in apartment buildings was conducted using frequency and temporal characteristics. First, a residential noise source dataset was constructed . The residential noise source dataset was consisted of floor impact, airborne, plumbing and equipment noise, environmental, and construction noise. The clustering of residential noise was performed by K-Means clustering method. For frequency characteristics, L_eq and L_max values were derived for 1/1 and 1/3 octave band for each sound source. For temporal characteristics, Leq values were derived at every 6 ms through sound pressure level analysis for 5 s. The number of k in K-Means clustering method was determined through the silhouette coefficient and elbow method. The clustering of residential noise source by frequency characteristic resulted in three clusters for both L_eq and L_max analysis. Temporal characteristic clustered residential noise source into 9 clusters for L_eq and 11 clusters for L_max. Clustering by frequency characteristic clustered according to the proportion of low frequency band. Then, to utilize the clustering results, the residential noise source was classified using three kinds of machine learning. The results of the residential noise classification showed the highest accuracy and f1-score for data labeled with L_eq values in 1/3 octave bands, and the highest accuracy and f1-score for classifying residential noise sources with an Artificial Neural Network (ANN) model using both frequency and temporal features, with 93 % accuracy and 92 % f1-score.

• Journal of the Society of Cosmetic Scientists of Korea
• /
• v.42 no.2
• /
• pp.103-109
• /
• 2016

Efficacy of cosmetics has been mainly evaluated by qualitative and quantitative methods based on visual sense, tactile sense and skin structure until now. In this study, we suggested a novel evaluation method for skin status based on sound; measuring and analyzing the rubbing noise generated by applying cosmetics. First, the rubbing noise was measured at a close range by a high-sensitivity microphone in anechoic environment, and the noises were analyzed by 1/3 octave band analysis in frequency-domain. Three conditions, 1) before washing, 2) after washing and 3) after application of cosmetics, were compared. As a result, sound pressure level (SPL) of rubbing noise after washing was larger than that of before washing, and the SPL of rubbing noise after cosmetic application was the smallest. Furthermore, the energy of rubbing noise after application was higher than that of the before and after washing conditions in a low frequency band (lower than 2 kHz region). Conversely, the energy of rubbing noise after application was much lower than the others in a high-frequency band (upper than 2 kHz region). This change of energy distribution was described as a balloon-skin model. High SPL in the low frequency region after the cosmetic applications was due to the increase of "flexibility index", while SPL in the high frequency region significantly decreased because of the attenuation which is related to "softness index". Therefore, we developed two indices based on the spectrum-energy difference for evaluating skin conditions. This proposed method and indices were verified via skin flexibility and roughness measurement using cutometer and primos respectively. These results suggest that acoustic measurement of skin friction noise may be a new skin status evaluation method.

• International Journal of Highway Engineering
• /
• v.14 no.2
• /
• pp.19-28
• /
• 2012

A measuring technique for tire-pavement interaction noise that uses a proposed noble close proximity(NCPX) method, which has been proofed in terms of the reliability and consistency of interaction noise measurement through several researches, equipped with surface microphones has been adopted in order to perform bridge deck pavement noise evaluations on four different pavement surfaces. Through field testing measurement of bridge deck pavement in Seoul inner ring road, the appropriate noise-measuring procedures have been used for evaluating the noise characteristics of four different surfaces. Measuring results show that tire-pavement noise levels vary depending on the surface types and vehicle speeds. Furthermore, the different characteristics of tire-pavement interaction noise can be found before and after the new surface construction of bridge deck pavements in terms of the 1/3 octave band analysis of vehicle speed.

• Journal of the Korean Society of Safety
• /
• v.12 no.4
• /
• pp.191-200
• /
• 1997

This paper discusses prediction of the sound pressure level produced by simple engine exhaust systems(plain pipe, plain expansion chamber pipe, plain expansion chamber with internally extended inlet and outlet pipe, perforated pipe enclosed in a plain expansion chamber) and a computer program has been developed which predicts the sound pressure level and the frequency spectrum. The program utilizes unsteady flow gas dynamic theory and acoustic theory to predict the pressure-time history in the exhaust system and the mass flow rate-time history at the open end of the system and the sound pressure levels(1/3 Octave band levels) and the frequency spectrum in semi-anechoic room. The predictions are compared with measured levels and show a high degree of correlation.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.15 no.9
• /
• pp.820-825
• /
• 2004

In this paper an octave tuning voltage controlled oscillator which is used in set-top TV tuner was designed. Oscillation frequency range is 0.9 GHz～2.2 GHz with 1.3 GHz bandwidth. By using 4 varactor diodes in base and emitter of transistor, wide-band tuning, sweep linearity and low phase noise could be achieved. Designed VCO requires a tuning voltage of 0 V ～ 20 V and DC consumption of 10 V and 15 mA. Designed VCO exhibits an output power of 5.3 dBm $\pm$1.1 dB and a phase noise below -94.8 dBc/Hz ＠ 10 kHz over the entire frequency range. The sweep linearity shows 65 MHz/V with a deviation of $\pm$10 MHz.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2006.05a
• /
• pp.941-945
• /
• 2006

In the development of electricity generating wind turbines for wind farm application, only two types have survived as the methods of power regulation; stall regulation and fun span pitch control. The sound measurement procedures of IEC 61400-11 are applied to field test and evaluation of noise emission from each of 1.5 MW and 660 kW wind turbine generators (WTG) utilizing the stall regulation and the pitch control for the power regulation, respectively. Apparent sound power level, wind speed dependence and third-octave band levels are evaluated for both of WTGs. It is found that while 1.5 MW WTG using the stall control is found to emit lower sound power than 660 kW one using the pitch control at low wind speed (below 8 m/s), sound power from the former becomes greater than that of the latter in the higher wind speed. Equivalent continuous sound pressure levels (ECSPL) of the stall control type of WTG vary more widely with wind speed than those of the pitch control type of WTG These characteristics are believed to be strongly dependent on the basic difference of the airflow around the blade between the stall regulation and the pitch control types of WTG. These characteristics according to the methods of power regulation lead to the very different noise emission characteristics of WTG depending on the seasons because the average wind speed in summer is lower than the critical velocity over which the airflow on the suction side of blade in the stall types of WT are separated. These results propose that, in view of environmental noise regulation, the developer of wind farm should give enough considerations to the choice of power regulation of their WTG based on the weather conditions of potential wind farm locations.

• Journal of KSNVE
• /
• v.4 no.3
• /
• pp.377-384
• /
• 1994

This paper is an investigation of the performance of absorption diffusers to suppress the vent noise emitted when high pressure gas is throttled. First, experiment for the static performance is carried out. When there is no through-flow, the insertion loss has been obtained in terms of 1/3 octave band spectrum and the effect of the number of diffusers and the thickness of the absorption material on the static performance has been obtained. And the similarity in the spectrum of the static insertion loss is confirmed by comparing two similar models with different size. Second, the dynamic performance has been obtained by experiment using blow-down of compressed air from a storage tank through an orifice of diameter 10 mm. The back pressure by the diffuser is measured and compared with that of a single diffuser. It is found that the insertion loss of asingle diffuser is very low around 3 dB at high frequencies with negative value at low frequencies. By absorption material between the diffuser tubes, however, the performance is increased considerably. Without flow the static insertion loss increases by 3 - 4 dB by doubling the thickness or the density of the absorptionmaterial. With flow, however, the dynamic insertion loss increases. While, the back pressure by the diffuser is small enough to be neglected.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2002.05a
• /
• pp.867-872
• /
• 2002

A high intensity acoustic test facility is constructed at Korea Aerospace Research Institute (KARI) by 2003. The reverberant chamber of the facility has a volume of 1,228 cubic meters and shall provide an acoustic environment of 152 dB over the frequency range of 25 Hz to 10,000 Hz. The facility consists of a large scaled reverberant chamber, acoustic power generation systems, gases nitrogen supply systems, and acoustic control systems. This paper describes how the basic parameters of a chamber and power generation systems are controlled to meet the requirement of the test. The volume of a reverberant chamber is controlled by the size of test objects and the reverberant characteristics of a chamber. The capacity of acoustic power generation systems is determined by the energy absorption of a chamber and the efficiency of acoustic modulators. Simple math is employed to calculate the required power of acoustic modulators. Moreover, the paper explains how the distribution of sound pressure level at low frequency is checked by analytical and numerical methods.

• Journal of the Korean Society of Fisheries and Ocean Technology
• /
• v.8 no.1
• /
• pp.14-22
• /
• 1972

For the development of acoustic fishing method, the noises of fishes have been recorded and analy/'ed by many scientists. Some specimens of fishes were selected as such Cyprinus carpio, Ctenopharyngodon idellus Carassius carassius, and pagrosol1ms major in this experiment. The noises such as feeding noise, driving away noise, jumping noise and fi llip noise were recorded by the tape recorder, Sony Model 262, through the underwa te r microph I one, Oki ST 6582, and analyzed in frequencies bv octave band analyzer, Rion SA-55, and sound pressure level of source by sound level meter, Rion NA-opNN The supplied feed was placed within 5em apart from the hydrophone. The result of analyzed noises were as follow. Cyprinus carjJio; Feeding noise 250- 500 cps, 92- 99 dB Driving away noise 125-2, 000 eps, 101-112 dB Jumping noise 125-2, 000 eps, 99-116.5 dB Ctenopharyngodon idcllus; Driving away noise 125-1, 000 cps, 96-109 dB Carassius carassius; Feeding noise 250- 500 cps, 91. 5- 99.5 dB Driving away noise 125-1, 000 eps, 99-108 dB Carassius auratus Feeding noise 250 eps, 94-101 dB Driving away noise 125-1, 000 cps, 98-110 dB Pagrosomus major Feeding noise 230-500 cps, 90-101 dB Fillip noise 500 cps, 98-108 dB (1) Feeding noise was produced as like as snap noise of twig and gulping down saliva noise in human and dominant frequency range of the noise is 250-500 cps and noise level 90-101 dB. (2) It was found that feeding noise were not a monotonic but a complex tones though fish took the same food. (3) Driving away noise was produced not so keen and the wave form of the noise is rising very sharp and big amplitude in the oscillograph. Dominant frequency range of this noise was about 150-1, 000 cps and noise level 96-112 dB except thut of carp. (4) The frequency of snapper's fillip noise, when it produced by caudal fin in swimming at the surface of water, was 500 cps and noise level 93-108 dB snd that of jumping noise of carp about 150-2, 000 cps and noise level 99-116.5 dB.

• International Journal of Highway Engineering
• /
• v.15 no.2
• /
• pp.11-18
• /
• 2013

PURPOSES : This study presents the noise level and frequency characteristics investigated in the national highways with the consideration of various measuring conditions and/or methods. METHODS : The noise levels on the asphalt concrete pavement(ACP) and the jointed plain concrete pavement(JPCP) of the national highway were measured and analysed with respect to three variables, i.e., pavement type, surface condition, and measurement distance. The PASS-By method is utilized for the noise measurement and then using CPB spectrum analysis method with 1/3 octave bandwidth, the noise levels and frequency characteristics were calculated for two-second periods before and after the peak noise. RESULTS : Depending on the pavement type, the noise level was changed as the average noise levels are 73.3dB(A) and 78.3dB(A) for ACP and JPCP, respectively. With respect to the effect of surface condition, the average noise levels for crack H(high), M(medium), and L (low) sections are 77.4dB(A), 77.4dB(A), and 78.1dB(A), respectively. Regarding the measurement distance, 1.2meter difference in measuring location reduces 1.6dB(A) of noise level; the average noise levels at 5.3m and 7.5m from the centerline of outer lane are 72.8dB(A) and 71.2dB(A), respectively. It should be noted that the noise levels are slightly different as a function of vehicle speed and type. However, the overall trends for each case was similar. It was found that the domain frequency bands for ACP and JPCP were 400Hz~2000Hz and 500Hz~2000Hz, respectively. CONCLUSIONS : Based on the analysis with the measured noise date from national highway, it was concluded that the noise level and frequency band vary depending on the various conditions. It was also found that some variables significantly affect the noise level while others do not. With further systematic investigation, the comprehensive noise characteristics on the national highway can be achieved. Using such database, it is possible to develop the fundamental noise reduction technology.