BP Neural Network And It's Using In Elevation Transmission Of The Cross-ocean Bridge

GPS measurements provide the three-dimensional coordinates, in other words, it can accurately determine plane location and ellipsoidal height of ground observation stations, but we use normal height in practical. How we directly use the GPS ellipsoidal height to obtain normal height in order to save manpower, resources and time, has been a focus of concern and research in surveying industry.In this paper,some Elevation fitting related concept has been introduced.such as geodetic level surface, quasigeoid, geodetic elevation, orthometric height, normal height ect; Analysis some fundamental theory and method of the geodetic level surface (or quasigeoid); And then have analysed the current situation that GPS elevation changes. At presren, the method of fitting is still their first choice to convert GPS height currently. Aceording to the difference of mathe matical models, there are many methods of GPS height conversion, such as plane fitting method, the ploynomials fitting method, the multi-function fitting method, the spline-function fitting method and the neural network method etc.Common methods, such as two surface function and three surface function, inevitably exist model error. On the contrary, the method of GPS height conversion based on neural network is a method of self-adaptive mapping. It makes no assumption, can avert from the influence made by unknown factors, can reduce model error, the oretieally is more reasonable and should raise the accuraey of GPS height conversion. The thesis is detailed to learn the BP algorithnl of neural network and makes many items of experiment and researeh to BP network model. Finally the thesis summarizes a structure of BP network model for GPS height fitting. According to the fact of Xiamen,Jimei bridge. We suggest that along bridge axis respectively setting up the high-accuracy GPS control net, Regularity using the second-order leveling to measure the normal height of the control points. And then using the BP neural network method to carry out the elevation fitting, in progress with the methods such as two surface function, Shepard interpolation, Fitting results have reached expected effect.Using COM builder, on the basis of VB and Matlab programming,we developed GPS Height conversion system of Xiamen Ji Mei bridge. Good to meet the need that GPS heihgt convert to Normal height of the construcion company in their practical work.