台大船模水槽之水下聲學性質量測

Abstract

針對目前水下聲學應用的測試場域除了開放水域外，就是室內的水槽或游泳池量測，而於開放水域使用上會耗費許多人力和金費等資源，因此若非具有實海域實驗的必要性，本實驗室人員從過去到現在許多聲學應用通常都於臺灣大學工程科學與海洋工程系上的船模水槽測試，而以非無響水槽測試勢必會有許多環境因素干擾，此研究目的為了解此水槽之聲學性質而作一系列測試和模擬驗證。 量測部分包含環境噪音、吸音係數、餘響時間。環境噪音量測為短、長時間兩部份，分別研究整體水槽之聲場於地域性和時域性上的變化，並同時記錄水溫和氣溫。吸音係數是以水為本體四周的邊界聲強損耗(如:牆壁、與空氣接觸的表面等)作為吸音係數的量測，而水槽內邊界使用脈衝方法(Tone burst)分別量測訊號直接聲壓和反射聲壓，並以此比值去計算吸音係數。餘響時間是觀察單頻正弦波於水槽之聲壓衰減行為，以衰減曲線平均法(Decay curve average method，DCAM)降低窄頻訊號於能量衰減曲線的波動，並使用包絡線平均(Envelope mean)平滑化衰減曲線，得出餘響時間參數早期衰減時間(Early decay time，EDT)、T_20、T_30。模擬部份則是使用軟體COMSOL Multiohysics內的有限分析方法去計算得出餘響時間參數值與實驗量測值比較。 研究結果顯示環境噪音由於水體夠大，水溫短時間內不被氣溫所影響，而水槽內存在一60赫茲倍頻之運轉噪音，吸音係數的牆面吸收能力，小於4kHz之吸音係數為0.974，其餘頻率下大多分布為0.6~0.7，不論是EDT、T_20、T_30，餘響時間在水槽內部都小於0.5秒內，並且顯示水槽內部空間可能存在不只單一空間，導致衰減曲線不具單一線性衰減，使得餘響時間被整體拉長，餘響時間參數實驗值呈現於此水槽使用EDT比T_20和T_30的結果來得好，而T_30目前不適用於此水槽。

In addition to open water, underwater acoustic testing is conduced in a tank or indoor swimming pool, as testing in open water costs considerable manpower, money, and other resources. So, if there is no requirement to test in open water experiments, the Underwater Acoustic Laboratory (UAL) from Department of Engineering Science and Ocean Engineering at National Taiwan University (NTU) usually does underwater acoustic testing in a shipping modal testing tank. As the original use was not for acoustical testing, interference from many environmental factors exists. This research project will help one understand the acoustic properties of the tank, and it gives a series of tests and simulations to verify findings. The items of measurement includes ambient noise, sound absorption coefficient and reverberation time. The ambient noise measurement was separated into two parts: short time and long time. The measurement in short time is spatially varying, and the measurement in long time is time varying while concurrently recording temperature in air and water. The sound absorption coefficient measures the sound intensity of the boundaries around the tank(such as walls, surfaces in contact with air, etc.), while measuring both the direct sound pressure and the reflected sound pressure of the signal by the Tone burst method. The reverberation time shows the attenuation behavior of pure tone energy in the tank, where the fluctuating signal on the energy attenuation curve is reduced using the decay curve average method (DCAM), and the envelope mean is used to smooth the decay curve, which eventually results in reverberation time parameters like early decay time (EDT), T_20and T_30. The simulation utilizes room acoustic concepts in the water, and the calculation module is based on statistical acoustics theory, and is called the diffusion equation. Its boundary condition is found by adding the measured sound absorption coefficient to the Eyring absorption model. Finally, verifications is through comparing the theoretical reverberation time with the measured value. The results showed that, due to the body of water being big enough, the air temperature does not affect the water temperature in short time, and the tank exists within the 60 Hz octave band. The sound absorption coefficient of the side walls at frequencies less than 4 kHz is up to 0.974, with other frequencies being 0.6 ~ 0.7. All of the reverberation time parameter EDT, T_20, T_30 is within 0.5 seconds, The results display that it may have coupled room effect caused by multiple spaces existing in the tank, which makes the decay curve not have linear attenuation, resulting in extension of the reverberation time. The experimental data show that EDT is better than T_20, T_30 in 4kHz ~ 10kHz.