The Research On Prediction And Control Of The Noise In The Sound Field In Tunnels

In recent one or two decades, China has been building dozens of highwaytunnels and several undersea tunnels. Tunnel enclosure is semi-enclosed andconsequently the direct sound and reverberation sound mix together. The soundpressure level is usually relatively high and the reverberation time lasts up to20s.The long reverberation time and high sound pressure level reduce the speechintelligibility in tunnels, where the acoustic quality is important for speech. In sum,Noise in the tunnels of China has always been a problem which needs urgentsolutions. There is a recognised need for a method of predicting and improvingacoustic indices and in long enclosures such as tunnels.The noise pollution in Qingdao Jiaozhou Bay undersea tunnel is investigated.The sound pressure level has been measured and the main frequencies of noise hasbeen analysed. The results indicate that the noise inside the tunnel is much higherthan that outside the tunnel, and the noise inside a car in the tunnel is lower than thatinside the tunnel.Computer simulation is an effective tool in the project of tunnel noise reduction.A number of innovative sound absorptive structures and low noise pavements couldsignificantly cut down the reverberation time and noise level in tunnels. A theoreticalmodel called image walls method had been developed for the prediction of soundpressure level and other acoustic indices in long enclosures, such as tunnels and tubestations. To start with, at the aim of predicting and improving acoustic quality intunnels, this paper models the finite-length tunnel enclosures based on the imagewalls method by using Odeon software. The model established by Odeon software isbased on the image source method and Ray-tracing method. Furthermore, this paperpresents that image walls boundaries should not be assumed to be hard boundaries incomputer simulation. Meanwhile this paper presents how to determine thereasonable length of the image walls computer model of tunnel enclosures. To validate the computer simulation model in practical situations, the simulationpredictions on reverberation time are found to agree reasonably well with themeasured data in a real tunnel at all frequencies. An application of the image wallcomputer simulation model is suggested. The image wall computer simulation modelserves as the first attempt to optimize the pattern of sound absorption materials in along enclsure, like tunnels, for the reduction of noise and reverberation time andimprovement of sound quality, with particular attention to the speech intelligibility.In conclusion, the finite length tunnel enclosure computer simulation model thatthis paper presents is an efficient and accurate method to predict and optimizeacoustic quality in tunnels.