Kiyoshi NAGAKURA Environmental Engineering Division Makoto IWASAKI

Based on vortex theory, we experimentally and directly predict sound sources distributing in the flow field and determine the sound pressure level as a result of the spatial integration of sound sources. In employing this direct evaluation method for the aeroacoustic sound, the problem is that a large integration area is required to minimize errors caused by the sudden truncation of the integration area; we overcome it by adopting and applying a modified formula that neglects the quadrupole sound under the condition that the dipole sound is dominant at a low Mach number. Through the flow field measurement using a time-resolved particle image velocimetry (TR-PIV) technique, we will clearly demonstrate the feasibility of our method and the distribution of dipole sound sources in the vicinity of a body even if a comparatively small integration area must be taken. In this basic study, a circular cylinder with a diameter of 6.0 mm is used; the spatially integrated sound pressure is compared with the actual sound pressure which is measured with a microphone. Further, the sound sources evaluated using only the flow field are determined, which give us detailed information about the amplitude and phase of the sound source structure. This direct evaluation method for the dipole sound is applicable to a more complex body.

East Japan Railway Company Aerodynamic and bridge noise originate from pressure fluctuations generated by highspeed trains in open sections without tunnels. In this study, field tests using a linear microphone array were conducted in order to clarify low-frequency aerodynamic sources of noise below 100 Hz. In addition, a scale-model experiment using a launching facility for a model train was carried out to simulate actual aerodynamic noise and investigate low-noise bogie cavity designs. Through these tests, the bogie cavities under the train body were identified as being one of the major sources of aerodynamic noise. It was also found out that rounding cavity edges could be an effective measure to reduce low-frequency aerodynamic noise.

Yusuke WAKABAYASHI East Japan Railway CompanyAerodynamic bogie noise generated from Shinkansen trains is the main source of the noise when they are running at above 300 km/h and noise reduction is important for preserving the quality of the environment around railway lines. In order to reduce bogie noise effectively, it is important to evaluate the contribution of the bogie components to the aerodynamic bogie noise. The purpose of this paper is to estimate their contributions at the measurement point close to the track by a wind tunnel test. Both the noise contribution of each component and the measures to reduce the aerodynamic noise are investigated by arranging the component in the bogie model.

It is known that pressure fluctuations are generated from intermediate vehicles of Shinkansen trains. Those pressure fluctuations involve the infra-sound phenomenon with a wavelength longer than several meters. Past studies showed that the infra-sound consists of the aerodynamic and structure-borne components, and the aerodynamic component is originated in the high-speed airflow around vehicles. The generating mechanism of the aerodynamic component, however, had been remained unclear. In this paper, a new methodology applicable to railway field tests is proposed to evaluate low-frequency aerodynamic sound less than 100Hz. A field test was conducted at a flat land without noise barriers, in which fifteen ultra-low frequency microphones constituting a linear microphone array were arranged in line along rails. In order to determine the frequency spectrum and distance decay rate, another five ultra-low frequency microphones were deployed at measurement points 8.7m to 50m apart from the nearest track center. The reason why the flat measurement site was selected was to suppress the bridge noise and focus on the aerodynamic noise during a train passage. On processing the microphone array signal to identify sound sources in field tests, much attention should be paid for the number of averaging; the number of averaging were limited so that sound source identification was difficult in practice. In this analysis, therefore, more than 100 trains were measured and ensemble-averaged to clarify the relationship between the number of averaging and precision of determining sound sources. Our field test campaign showed that low-frequency aerodynamic sound was locally distributed around all sections between two neighbouring bogies and pantograph sections.