Application Research Of Semi-parametric Model In Airborne Gravity Downward Continuation

The airborne gravity survey is a new gravity measurement technology, which basing on aircraft, including gravimeter, GPS, altimeter and attitude sensor and other equipment to determination of near-Earth space gravity.Airborne gravity measurement is a breakthrough change in gravity measurements. Traditional gravity measurements and satellite gravity measurements complement the wide range of applications in geophysics, geodesy, and other disciplines. Airborne gravity measurement of the maturity of the fast high-precision low-level gravity anomalies become a reality, but in geodesy used the ground gravity anomaly, how to route a high degree of gravity anomalies downward continuation to the ground known as the airborne gravity data deal with key issues.The Downward Continuation of the measured gravitational field of non-stationary amplification process, if the observations contain small errors, the extension will have a great error, resulting in the distortion of the extension is ill-posed problems. So, how to eliminate or weaken the Downward Continuation of instability, to find the right solution method more accurate down extension is necessary. Due to the presence of systematic errors or model errors, the classical least squares adjustment model based on the linear model can not fully adapt to the need for downward continuation, you need to study the applicability of a broader, more general adjustment model.Semi-parametric model function model component of a parameter, a non-parametric component and random error component, the parameters component describes the observations in the function composition, non-parametric component specifically said that a function of the unknown systematic errors or model errors, thus the model has good ability to adapt. On the basis of the summary of the traditional airborne gravity downward continuation of the successful application of the semi-parametric model to the airborne gravity downward continuation of airborne gravity data processing provides a new solution, has great practical value.The first chapter is an introduction.lt provides an overview of the Airborne Gravity Data downward continuation of the study of the meaning and development in the world.It cited the downward continuation of the existing problems.lt describes the basic theory of airborne gravimetry system structure, data processing and accuracy assessment methods, abstract write the content of this study.The second chapter analyzes traditional of several airborne gravity data downward continuation of the basic methods:Inverse Poisson method, the virtual point-mass method, least squares collocation method, gradient method.Choose the most effective traditional method-the least-squares method based on the inverse Poisson equation.lt carried downward continuation simulation downward continuation test.In the two examples:First simulate different height data downward continuation of the simulation;And the second compare the estimated value and the true value of the difference,and in conjunction with the vertical and horizontal coordinates,preparing visual programming,and draw a better difference simulation diagram.We analyed of the formation of its disadvantages,then find out the reason..The third chapter analyzes the airborne gravity downward continuation of the causes of problems:On the one hand the non-stationary signal amplification caused by the normal equation morbid;On the other hand, the high-frequency noise caused by the impact of topography. Address these two issues, to find out the more successful ways to correct them,such as regularization method, ridge estimation method, remove-recovery method.Select the data the downward continuation calculated,at first we regularizate method direct simulation.Then we re-simulation, terrain correction regularization method.Finally, the difference between the estimated value and the true value is obtained and draw the difference figue.In the forth Chapter we study semi-parametric model estimation model and make reasonable summary.After experimental Comparation,we choose to apply to the airborne gravity downward continuation of the semi-parameter ridge estimation method,cross-verification method introduces a unique dual regularization parameters.In the numerical example we use semi-parameter ridge estimation model downward continuation of different heights gravity anomalies,and draw data dimensional visual simulation figue. We compare the simulation as well as the difference between the effect of least squares method, regularization method, application of semi-parametric model. We analyze the reason of the advantages of the semi-parametric model.The fifth chapter summed up the main research content and some of the deficiencies and limitations,then prospect directions for future research and work.This is the basic theory of applied research. We supply land improve on the basis of existing research results,studying new theories and methods.The focus is to find problems in the downward continuation of airborne gravity. We summary of the success of the previous deal with the problem, choose specific method developed semi-parametric model which is applied to such problems.Viable theories and methods for practical applications is applied to resolve the difficulties in the application and promote development of the airborne gravity downward continuation.