Research On The Construction Of Ocean Sound Velocity Field And The Adjustment Method Of Seafloor Geodetic Network

The ocean contains abundant resources.Due to the progress of science and technology as well as the needs of production and living,the development and utilization of the ocean has become a common strategy of all countries in the world.The demand for underwater high-precision acoustic positioning technology is increasing day by day.The construction of China's marine geodetic datum is still in its infancy,and high-precision acoustic positioning technology has become an important part of the construction of marine space-time datum and a key issue that needs to be solved urgently.Supported by the national key research and development program of "Marine Geodetic Datum and Marine Navigation's New Technologies(2016YFB0501700)",this paper carried out an in-depth study on the construction of ocean sound velocity field and the adjustment of seafloor geodetic network,and achieved the following research results:1.This paper summarizes the research status of underwater acoustic positioning technology,ocean sound velocity field and seafloor geodetic network,compares the advantages and disadvantages of four common positioning systems,analyzes various errors in the positioning process,and presents two correction algorithms for sound ray bending error:ray-tracing method and equivalent sound velocity profile method.2.Based on the research of constructing the traditional sound velocity space field,aiming at the problem that the sound velocity profiles(SVPs)cannot be accurately obtained in time,this paper proposes a method of SVPs inversion and prediction based on radial basis function(RBF)neural network.The method is based on the nonlinear function approximation capability of neural network,by using the measured temperature,salinity of the sea area and "Array for Real-time Geostrophic Oceanography"(Argo)data to build the sound velocity profile prediction model.The proposed SVPs prediction method was verified with the Argo data.The results show that the SVPs prediction method based on RBF neural network is more suitable for real-time or near real-time prediction of marine SVPs.It not only has a shorter time for training and modeling,but also can improve the prediction accuracy of shallow water layer by adding the temperature and salt information of sea surface,so as to obtain SVPs closer to the actual situation.3.Usually,an acoustic positioning method based on the ship with a surveying trajectory is used to perform absolute calibration on the seafloor geodetic points,which can transmit the land-based measurement datum to the seafloor.Due to the influence of unreasonable observation structure and various errors including sound velocity error,sound ray bending error,and observation gross error,the calibration accuracy of will be reduced.Aiming at the above problems,a calibration method of seafloor geodetic points based on improved acoustic ray-tracing method and robust estimation theory is proposed in this paper.This method raises the calculation efficiency by continuously improving the accuracy of the average sound velocity in the process of ray-tracing,and carries out robust estimation,which greatly weakens the influence of sound velocity error and observation gross error.By resolving the measured data,the new method can effectively improve the positioning accuracy of the seafloor geodetic points.And by comparing the results of different surveying trajectories,it is verified that the trajectory is uniformly distributed and symmetrical can effectively improve the accuracy and reliability of the positioning results.4.Since the conventional adjustment methods of seafloor geodetic network cannot weaken the influence of the ranging system error,this paper proposes a three-dimensional network adjustment method based on semi-parameter estimation theory and additional depth values as the constraints.The proposed method is verified by using the measured data of seafloor geodetic network.The results show that the new method can effectively reduce the influence of the system error of sound velocity by introducing semi-parameter components,and through the additional depth values to constrain,effectively improve the estimation accuracy of seafloor geodetic points in the vertical direction.Therefore,the overall calibration results of the seafloor geodetic points based on the new method are closer to the reference coordinate values.The adjustment method provides a good research idea and method for the data processing of the seafloor geodetic network.