基于Takagi-Sugeno模糊神经网络模型的卫星钟差预报方法

发布时间：2018-04-13 03:16

本文选题：天体测量学 + 时间　；参考：《天文学报》2017年03期

【摘要】：卫星钟差预报精度的不断提升是精密导航的关键问题.为了进一步提高钟差的预报精度和更好地反映钟差的变化特性,提出一种基于Takagi-Sugeno模糊神经网络(Fuzzy Neural Network,FNN)的钟差预报方法.该方法首先根据钟差数据的特点对钟差进行预处理,然后以预处理后的数据建立一种高精度预报钟差的Takagi-Sugeno模糊神经网络算法.采用IGS(International Global Navigation Satellite System Service)不同采样间隔的精密钟差数据进行了短期预报试验,并与ARIMA(Auto-Regressive Integrated Moving Average)模型、GM(1,1)模型及QP(Quadratic Polynomial)模型进行了对比试验,分析结果表明:对不同类型原子钟,该方法用于钟差短期预报是可行的、有效的,其获得的卫星钟差预报结果明显优于常规方法.
[Abstract]:The key problem of precision navigation is to improve the accuracy of satellite clock error prediction.In order to improve the accuracy of clock error prediction and better reflect the variation of clock error, a clock error prediction method based on Takagi-Sugeno fuzzy Neural network is proposed.In this method, the clock error is preprocessed according to the characteristics of clock error data, and then a Takagi-Sugeno fuzzy neural network algorithm with high precision for clock error prediction is established.The short term prediction experiment was carried out by using IGS(International Global Navigation Satellite System Service data with different sampling intervals, and compared with the ARIMA(Auto-Regressive Integrated Moving average model and QP(Quadratic Polynomial model. The results showed that different types of atomic clocks were tested.It is feasible and effective to use this method in the short-term prediction of clock difference, and the result of satellite clock error prediction is obviously better than that of conventional method.
【作者单位】：桂林电子科技大学信息与通信学院;广西精密导航技术与应用重点实验室;
【基金】：国家自然科学基金项目(61263028) 桂林电子科技大学研究生教育创新计划资助项目(2016YJCX14)资助
【分类号】：P228

【相似文献】