Research On The Improving Algorithm Of A-weighted Wavelet Transform And The Analyzing Method Of Acoustic Mode For I.C.Engine Sound Source Identification

With the rapid development of the automobile industry of our country and the great increase of automobile quantity, the traffic noise caused by vehicles has become the main noise source of city environment now. As internal combustion (I.C.) engine is the main sound source of vehicle, the noise level of I.C. engine should be firstly controlled in order to reduce the vehicles' noise level. Emission and noise are two main projects for I.C. engine development in the new century, and they determine the development future of I.C. engine. According to the acoustic characteristic of noise, sound source identification plays an important role in I.C. engine noise control. To control I.C. engine noise level, the main sound sources must be found out according to the noise characteristic. The expected control results can be obtained only after the corresponding measures being taken.The thesis focused on sound source identification of I.C. engine. On the base of a large quantity of literature and comparing various sound source identification methods, the effective methods of sound intensity and wavelet analysis were selected out for deep research, and some new conclusions and analysis methods worthy of engineering application were obtained. When the new methods were used to investigate an actual engine's noise and identify the noise sources, ideal results of sound source identification were achieved.The corresponding knowledge of I.C. engine measurement was introduced firstly. The knowledge was the theoretic base and necessary data for the selection and improvement of I.C. engine sound source identification method. The characteristic of I.C. engine main sound sources and the measurements to control the sound level of the sound sources were systematically analyzed by searching literature. They were necessary for exact identification of the sound source and the effective control of the sound level.Fourier transform(FT) is a bridge between time domain and frequency domain, and it's the foundation of some modern signal analysis methods such as short time Fourier transform(STFT), sound intensity analysis and wavelet transform(WT) etc. The theory and algorithm of FT were introduced in the thesis and an example showed that FT has some shortages when being used to process non-stationary signal. Because I.C. engine signal is a periodic non-stationary signal, time-frequency analysis...