Effects of Inter-Coaching Spacing on Aerodynamic Noise Generation Inside High-speed Trains

초록

Experiments were performed in order to investigate the effects of inter-coach spacing on the aerodynamic noise generation inside high-speed trains. The inter-coach spacing is an important aerodynamic noise source. The cavity in the inter-coach space induces the flow separation and self-sustained oscillation which results in significant increase in the wall pressure spectra on the structure of the high speed train. From the propagation of the wall pressure along the structure, it results in flow induced vibration of side glass windows and consequent interior noise generation. As there has been continued success in increasing the speed of the trains, the aerodynamic noise sources become more important, recently. To insure safe and comport ride of passengers and to further increase the operating speeds, the noise generation mechanism of the aerodynamic sources should be understood. From the identified mechanism, effective methods of noise control can be proposed. The primary goal of this investigation was to assess the influence of the geometries of the inter-coach spacing on the noise generated from side glass windows of high-speed trains. The modal properties of the glass windows were determined through experimental modal analysis. Measurements of surface wall pressure fields were performed using condenser microphones to understand excitation characteristics. For experiments using actual high speed trains in operation, there is a limitation on the investigating the effects of inter-coach spacing by varying parameters due to existence of other noise sources such as rolling noise. Also, it is difficult and expansive to vary the design parameters for actual trains. To conduct the experiments in the controlled environment, the wind tunnel tests were performed. The model of the cavity which has same dimensions as the real cavity of the train was custom-built and installed in the wind tunnel test section. The actual mud-flap was used in the model construction. The purpose of the mud flap in actual high-speed trains is to protect various instruments installed in the inter-coaching space. The model was designed so that the length between two mud-flap can be varied. This length represents that of the inter-coach spacing. A microphone was installed inside the cavity to measure the acoustic pressure. An array of Microphones was installed flush with the wind tunnel test section. Those were installed to measure characteristics of flow grazing over the cavity. Using the microphone, the wall pressure spectra and its convection speed was measured. The experiments were performed for various length and different geometries of the cavity. From the measurement the design parameters that yield minimum wall pressure excitation was investigated. To investigate the effects of cavity spacing, the length between the mud flap was varied and its effects were measured using the microphone arrays. The flow velocity monitored using manometer was varied from 5 m/s to 70 m/s. Velocity fluctuations within the turbulent boundary layers contributed to the wall pressures on the structure. Due to the complex nature of the turbulent flow, statistical descriptions of the wall pressure fields were used. When the surfaces of the structure are irregular, the turbulence intensity of the grazing flows is significantly increased relative to flow grazing over smooth and flat surfaces. For the current study, the cavity acted as the irregularity. From the measurements, increased wall pressure spectra were measured for larger cavity length. In the measured wall pressure, several pure tones were measured which is not present for turbulent flows over smooth surface. These tonal components in the measured pressure occurred due to flow feedback phenomena. These frequencies were predicted by Rossiter’s equation and agreed well to the measured values. The tonal components at specific frequencies were significantly larger than those of the wall pressure contribution from turbulent flows alone. Also, more significant variation was observed for the magnitude of these pure-tones from the variation of the length of the cavity. At these frequencies, the convection speed was much higher than the components from the turbulent boundary layer alone. Large convection speed results in significant vibration. Also, the sound transmission loss of the window panel when excited by moving pressure of high convection speed is smaller compared to those of the small convection speed. As the length of the spacing decreases, the tonal frequencies increased, but its magnitude decreased. Since there is a limitation in minimizing the length, methods to reduce wall pressure excitation was investigated. First approach was to corrugate the mud-flap. This corrugation was intended to reduce resonances due to the flow feedback. The mud-flaps were made using different sizes and arrangements (mirror or parallel) of corrugation. This corrugation resulted in reduction of the wall pressure measured at specific frequencies. However it yielded larger excitation in overall frequency ranges. As an alternative approach, the acoustic treatment in the inter-couch spacing was considered. When the cavity was treated by the acoustic materials, there was significant reduction in the pressure inside the cavity and also on the wall pressure spectra. Finally the modal properties of the glass windows of the high speed train were measured. Fundamental frequency ranged from 30 Hz to 55 Hz depending on the mechanical properties. These modal properties have impact on the sound radiation characteristics and transmission loss of the panel. Using the measured wall pressure spectra and the modal properties sound radiation characteristics is analyzed. These models can be used to minimize sound radiation by maximizing the energy dissipation at boundaries of the window or by adjusting the natural frequencies to avoid resonance due to the flow-feedback at the cavity of the inter-coach spacing.