| So far,many theories about sound propagation in seabed sediments have been put forward,but these theories have limitations,and the theories suitable for most cases have not been put forward.Although particle size is one of the important physical parameters of sediments,it is still not directly introduced into many acoustic models.Although previous scholars have also studied the influence of particle size on acoustic characteristics,most of the research results are obtained by measuring sediments in different sea areas.As we all know,although the data obtained from seabed in-situ measurement is closer to the real data of sediments in nature,there are too many factors affecting seabed in-situ measurement,so it is impossible to effectively carry out the study of control variables in seabed in-situ measurement.Compared with in-situ measurement,laboratory measurement has the advantages of less environmental impact and conducive to controlling variables.This paper will study the effect of particle size on the acoustic characteristics of water saturated sandy sediments in the laboratory.The experimental research covers the frequency band of 90 k Hz-200 k Hz.Through this experiment,the influence of particle size on sediment acoustic characteristics and the variation of acoustic parameters with frequency in this frequency band can be obtained.Due to the small interference of laboratory measurement data,the obtained sound velocity and attenuation data can be used to test the acoustic model.This paper will study four kinds of water saturated sandy sediments with different particle sizes.In order to ensure the accuracy of the experimental measurement results,broadband measurement and narrowband measurement are used to measure the sound velocity and attenuation of sediments respectively.In narrowband measurement,the performances of various time delay estimation methods are compared,and the time delay estimation algorithms suitable for calculating phase velocity and group velocity are selected respectively.Considering the uneven transmission voltage response of the sound source used in the experiment,the driving signal of the sound source is added with "Blackman" window and compensated in advance.After the pretreatment of the transmitted signal,the acoustic pulse distortion caused by the uneven voltage response of the transducer is effectively reduced.Comparing the sound velocity and attenuation data obtained by the above two methods,the results show that the two methods agree well in the main frequency band.1.Effect of particle size on acoustic characteristics of sedimentsFor the four kinds of sandy sediments with different particle sizes studied in this paper,the sound velocity of sediments increases with the increase of average particle size;The sediment with the largest particle size has the smallest acoustic attenuation coefficient,and the sediment with the smallest particle size has the largest acoustic attenuation coefficient;The sound attenuation coefficients of the other two sediments are close.2.Relationship between acoustic parameters and frequencyFor the four kinds of sediments with different particle sizes studied in this paper,the sound attenuation coefficient increases with the increase of frequency.In the studied frequency band,due to the weak dispersion of sediment sound velocity,in order to test the effectiveness of the phase velocity data obtained by the experiment,the sediment group velocity data are obtained in this paper.By comparing the group velocity and phase velocity,the positive dispersion characteristics of phase velocity can be determined.3.Comparison between acoustic parameter measurement results and model prediction resultsAfter measuring the acoustic parameters in water saturated sandy sediments,some physical parameters of sediments are obtained experimentally.Combined with genetic algorithm,some physical parameters that are difficult to be measured experimentally are obtained.Based on the experimental data and fitting data of sediment physical parameters,the sediment acoustic model is used to predict the acoustic parameters.It is found that the model results of EDFM are in good agreement with the experimental data,while the model results of Buckingham model are quite different from the experimental data. |