| This paper discusses the important significance of the acoustic temperature fieldmeasurement in the industrial and civil measurement area. Briefly introduces the researchof the technology at home and abroad, and describes the principle of temperaturemeasurement. The hardware constitution and software design of the acousticthree-dimensional temperature field measurement are expatiated systematically. And dataprocessing, temperature field reconstruction and other related problems are studied in thispaper.A time-delay estimation method based on the combination of wavelet transform andcross-correlation is proposed by the paper. This method preprocesses the signal withcompromise threshold function and threshold estimation strategy of Birge-Massart, thendetermines delay between the signals by calculating the cross-correlation of the twode-noised signals. The method proposed is comparatively studied with cross-correlationmethod and second correlation method. The results show that these three methods can giveaccurate time-delay estimation results if the SNR is high; however only the methodproposed in the paper can give accurate time-delay values if the SNR is very low and thenoises are strong correlated. Thus the new method has strong antinoise ability.The20groups of speakers and microphones are decorated around the1m1m1mcube support. Cooperating with transmitting, receiving, signal conditioning, dataacquisition module circuit, it composes the hardware system of the acousticthree-dimensional temperature field measurement. Through the LabVIEW software, theacoustic signal is controlled to transmit and receive, and the acoustic data are analyzed andacquired. The acoustic data acquired are processed to estimate the time delay value in theMATLAB software environment, then the travel time of sound waves through the validpath in measurement area are determined, after loading the acoustic data to temperaturefield reconstruction algorithm, the three-dimensional temperature field measurement isrealized. The principle of positive and reverse path acoustic travel time complementary is takento improve the number of available path used in temperature field reconstruction. Themodified factor is employed to overcome the angle of the near and far microphones and themeasurement error between the near and far microphones. From the results of severaltemperature field reconstructions, it can be seen that the modified factor is effective. Theeffect of speaker voice interval to the number of effective sound path and the accuracy ofthe temperature field reconstruction is researched. It can be considered to reduce thespeaker voice interval to improve the system efficiency. However, the speaker voiceinterval should be longer than acoustic signal duration in case of the acoustic echoinfluence to the time delay estimation. |