摘要
提出了一种基于量子粒子群神经网络(QPSO-BP)模型的GNSS高程转换方法,通过建立GNSS点平面坐标与正常高之间的三层QPSO-BP数学模型而实现GNSS高程转换。试验分析结果表明,该方法全局迭代进化搜索能力高、稳健性强、拟合及预测精度高,在GNSS高程转换方面具有良好的有效性与先进性。
In this paper,a GNSS elevation transformation method based on quantum particle swarm neural network (QPSO-BP)model is proposed.By establishing the three level QPSO-BP mathematical model between GNSS point plane coordinates and normal height, the GNSS elevation transformation is realized.The experimental results show that the method has high global search ability,strong robustness,high fitting and prediction accuracy.This method has good effectiveness and advancement in GNSS elevation transformation.
作者
韩红超
HAN Hongchao(Mapping Design Academy of Ningbo,Ningbo 315100,China)
出处
《测绘通报》
CSCD
北大核心
2019年第1期85-88,共4页
Bulletin of Surveying and Mapping
关键词
量子粒子群
BP神经网络
GNSS高程转换
quantum particle swarm
BP neural network
GNSS elevation transformation