The Application of Various Methods in Geoid Fitting
Liu Qing-yua · Gnss World of China · 2014
This paper studies the application of the quadratic surface,multi-faceted function,BP neural network,and improved BP neural network in geoid fitting.Because of the shortcomings of BP neural network,such as slow convergence rate and falling into the minimum value easily,this paper proposes that the global optimization features of particle swarm can be used to overcome the shortcomings.By fitting the experimental data,this paper analyzes and compares the advantages and disadvantages of different algorithms.The experimental results shows that the improved neural network has considerably reduced the number of training and short the time of convergence and increasing accuracy,verify the effectiveness of the improved algorithm.