Research On Inversion Of Sound Speed Profile In The Ocean

The study about inversion of interested ocean environmental parameters is an important problem in the field of underwater sound. The sound speed profile in the ocean has important influence on sound propagation in the medium. Propagation of sound is distinct with different sound speed profiles. Since ocean environment is so complex, the speed of sound may vary greatly with time and location. Researches on inversion of sound speed profile have practical meaning to improve capability of underwater sound equipments and exploitation of ocean resources. In this dissertation, the inversion of sound speed profile in the ocean is studied. The main contents are as follows:Firstly, numerical method for calculating sound field in the ocean is studied. The sound propagation model is important to the problem of inversion. Ray theory is practical for its clear physical interpretation and capability of calculating three-dimensional sound field. A three-dimensional ray-tracing program has been developed.In the dissertation, the methods for fast computation of sound field are developed in order to improve efficiency of inversion algorithm. The sound field is calculated based on ray theory. Traditional algorithm for finding eigen-rays is analyzed, and modification of the algorithm is made. Simulations show that the efficiency of the new algorithm for finding eigen-rays is improved greatly with these modifications.The inversion of sound speed profile is an optimization problem. A genetic algorithm is used to search the optimum solution. The cost function is constructed. In this dissertation, the sound speed profile in the ocean is inverted by using travel time of sound signal.The experimental data are used to invert sound speed profiles in the experimental sea area. The results of inversion are given in the dissertation. In the process of calculation, the sound speed profile is described by empirical orthogonal functions. The empirical orthogonal functions are calculated by using previously measured sound speed data. The problem of mismatch is also studied. The errors in measurement of sound signal travel time have more influence on the result of inversion. The sound speed profile inverted by using experimental data is compared to the sound speed profile measured in the experiment. The calculated result is consistent to the sound speed profile measured.