Research On The Treatment Methods Of Un-uniqueness In Geodetic Inversion

Because of the equality of geophysical field and localization of geodetic observations, the un-uniqueness of estimated parameter is one of the most important and unavoidable problems in geodetic and geophysical inversion. At the sametime, getting the optimization is a critical problem to the explanation to relative geoscience. So how to get the optimization of inversion problems is the hotspot and nodus in geodetic and geophysical inversion. From how to get the optimization of non-linear inversion, external detection of geodetic inversion model, and the establishment of inversion model (application of inequality constraint), the methds of disposing un-uniqueness of estimate parameter are discussed in this paper. This thesis involves the aspects as following:1. At the base of R.Parker's theory, the research of geodetic inversion of thirty years is unified at four aspects: the existence of settlement; the foundation of inverse model; the non-uniqueness of settlement; the appraisement of settlement. The definitions of existence of settlement, foundation of inverse model, non-linear geodetic inversion and appraisement of settlement are given.2. For the first time, regression analysis, survey adjustment and geodetic inversion are compared with each other in definition, theory composing, resolving methods, applied areas and so on. Their differences and affiliation are summarized.3. Studies are concentred on the methods of non-linear inversion problems. Quantum Annealing is introduced in geodetic inversion. Hemi-linear inversion method of geodesy based on cubic spline is put forward.4. Hemi-outer testing method is given for geodetic inversion model. The divide method of observation and the standard of Hemi-outer testing method of geodetic inversion model are put forward.5. The Inequality Constraints Geodetic Inversion Theory is given to solve the un-uniqueness problem with the prior information. At the same time, the advantages, key point and difficulty of this theory are discussed. At last, the realization, classification method, and applied foregrounds of the theory are given.