模糊神经网络在GPS高程转换中的应用

介绍了模糊神经网络基本原理和GPS高程转换方法,采用模糊神经网络算法,实现了GPS高程转换.在用模糊神经网络进行GPS高程转换时,输入变量模糊分割数对计算结果的精度起着关键作用,模糊分割数过少,计算精度较低,过多,模糊规则数会急剧增加,计算精度不一定高,同时又会影响系统计算速度.为了进一步提高GPS高程转换的精度,模糊神经网络模型参数的选择以及网络结构的确定等问题,都需要进一步的研究和实践.

On Application of Fuzzy Neural Network in GPS Height Conversion

This paper discusses the basic principles of fuzzy neural network and the method of GPS height conversion, and implements of the GPS height conversion by fuzzy neural network method. The variable parameters play an inpertant role in calculation accuracy. The less paramters, the poorer acctmacy; while, however, the more parameters do not mean high accuracy, and may result in innormal cal- culating speed. In order to improve the accuracy of GPS height conversion, the selection of model parameters in fuzzy neural network and network structure should be paid more attention to based on the research and practice.