Mirko Čudina’s research while affiliated with University of Ljubljana and other places

In some cases an impulsive noise source such as a gunshot can be a preferred alternative when investigating building acoustics, including sound insulation measurements, when compared to conventional steady state noise sources. A gun equipped with blank cartridges is an impulsive noise source that is lightweight and small enough to be easily transported. The differences in the noise characteristics between individual cartridges for the same gun are usually small, so the impulsive source can be replicated to a high degree. This paper is focused on the practical application of the sound exposure levels produced by a gunshot with a known sound energy level in the rooms under investigation. In this way, the equipment and methods required by the conventional method are simplified significantly. Furthermore, reverberation times need not be measured, since the equivalent absorption area can be directly obtained from the measured sound exposure levels. Using Green’s theorem, the roles of the sound source and measuring microphone were exchanged, which simplified the determination of sound insulation as it was easier to change the position of the gun than the microphone. The results obtained using the impulsive noise source were in good agreement with those obtained using the conventional method. Above 100 Hz, their difference in any frequency band of interest was less than 1 dB.

A necessity for the suitable recognition of different machinery and equipment based on the sound they generate is constantly present and will increase in the future. The main motivation for the discrimination between different types of machinery sounds is to develop algorithms that can be used not only for final quality inspection but for the monitoring of the whole production line. The objective of our study is to recognize the operation of the individual machine in a production hall, where background noise level is high and constantly changing. An experimental plan was designed and performed in order to confirm the hypothesis proposing that automatic speech recognition algorithms can be applied to automatic machine recognition. The design of the automatic machine recognition procedure used in our study was divided into three stages: feature extraction, training, and recognition (classification). Additionally, a traditional mel-frequency cepstral coefficient (MFCC) procedure was adjusted for machinery noise by using different filter compositions. Finally, two classifiers were compared, the k-NN classifier and the multivariate Gaussian distribution. The results of the experiment show that machinery noise features frequency cepstral coefficients (FCC) should be extracted by using linear filter compositions and processed with recognition algorithm based on the multivariate Gaussian distribution.

Adaptive muffler upgrades the conventional muffler with a variable geometry to extend its working range. Adaptive muffler systems, described in the available literature, operate on the principle of controlling only single property of the muffler to tune its working range. During the adaptation process of these systems the basic principle of operation is not changed. Based on the literature review, the question arises whether it would be possible to control the principle of operation. At one instant it would work as a Helmholtz resonator, in the other as an expansion chamber or side branch resonator. Analytical calculations provided a positive answer and a muffler with four active valves was designed and constructed to confirm the theoretical expectations. Properties of proposed muffler were simulated with Finite Element Method. Results of simulations were validated with measurements. Finally, simulations of proposed muffler, working at different operating conditions, were performed. Simulations have shown that the proposed system allows tuning of the muffler to a wide working frequency range. Such approach to muffler design eliminates the need for additional mufflers used in IC exhaust systems. Therefore a smaller volume of the proposed muffler enables the same effect as multiple non-adaptive conventional mufflers.

Information about vibration modes is needed during the planning of noise control measures on different parts of machinery. A visualization of vibration modes is a starting point and different methods can be used to visualize vibration modes. Some methods which incorporate an inverse calculation of surface velocity from the sound pressure on some boundary have already been proposed, among others. A direct inverse method based on the discretized Rayleigh integral was used in our work, to demonstrate that microphones should be placed close to the vibrating structure to provide an acoustic transfer matrix with a low condition number. It is demonstrated, that there is practically no need for the calculation of the inverse matrix if microphones are placed in a very near field of the vibrating structure. A single microphone placed in a very near field together with a reference vibration sensor provides sufficient information for producing the images of vibration modes. Analytical results, numerical results, FEM simulations and measurement results were used to prove that properties of the sound pressure in a very near field permit a cost effective visualization of the vibration modes.

Impulse noise is one of the most hazardous and annoying types of noise, which is present in the working and community environment. A-weighted equivalent sound pressure level, describing most types of noise, appears not to be the appropriate descriptor, especially not for high-impulsive and high-energy impulsive noise. However, this descriptor is often used as a basis for impulse noise evaluation, when combined with appropriate adjustment terms. But despite of objective character of such evaluation, care should be taken regarding certain facts, especially the source of impulse noise, its environment and time of measurement. In this article, the relationships between all these influential parameters have been investigated in detail. Today, on the other hand, a sophisticated sound level meter offers the possibility of simultaneous measurement of many acoustical descriptors. By combining some of these descriptors with some analytical investigations, as shown in this article, more useful information concerning impulse noise can be obtained. In this article some of them are used for more detailed analysis of impulse correction, according to some international standards.

To succeed in strong competition market the reduction of sound levels and even sound color of products is increasingly more important. Automotive industry has a lot of experiences with psychoacoustics and how to make sound quality of their products more pleasant and customer appealing. A transfer of experiences from the automotive industry to the household appliances is presented in this paper. A sound quality estimation of seven suction unit and five different vacuum cleaners who had built the same suction unit, was performed. Two methods were used to determine the sound quality of these suction units; a jury based listening tests and a quantitative sound quality metrics. Details of these methods are presented and discussed. Quantitative sound quality metrics results are presented and compared with jury based listening tests. Suggesting on results, sound quality analysis can be used for a sound pleasantness evaluation of vacuum cleaner and suction units and give satisfactory indication of customer preference.

During the reconstruction process of the concert hall acoustic it is important to have some information about the sound absorption coefficient of already installed and applied materials. A new in-situ method for measuring the sound absorption was developed. Presented method utilizes a dipole characteristic of the source, which generates the test signals. A dipole source in a perfect free field conditions generates a specific shape of the sound field. If the dipole source is placed in the vicinity of arbitrary acoustic obstacle, sound field is deformed and it depends only on the acoustic properties of the obstacle. A difference between the sound pressure field of the dipole source in a perfectly free field conditions and the sound pressure field just above the acoustic obstacle gives sufficient information for the calculation of the sound absorption of the obstacle itself. A Rayleigh integral was used for numerical calculations of the sound pressure in a Very Near Field. Results indicate that the size of the dipole source should be as small as possible, but it should provide sufficiently high sound pressure levels. Numerical results indicated that it should be placed as close as possible to the surface of the sample to achieve good signal to noise ratio. Based on the numerical results, a measuring setup was build. A 8 cm dipole source was used and it was placed 2 cm above the acoustic obstacle. Measurement results were compared with results obtained with the Impedance tube.

Measurement of noise and vibration signal in audible frequency range to detect cavitation in centrifugal pumps is rather unknown technique. There were already some studies performed on this technique and they showed quite good results. Due to many factors that influence the quality of the measurement, an uncertainty analysis should be performed. This paper deals with estimation of a measurement uncertainty for different kinds of measurement ways to detect the cavitation in a centrifugal pump with noise and vibration signal in audible frequency range from 20Hz to 20kHz. Especially the measurement uncertainties for cavitation detection in broad frequency range and at a discrete frequency were analyzed. Results showed that this technique is reliable despite many possible influences on uncertainty.