基于狼群算法优化极限学习机的北斗卫星钟差预报  

Optimization of Extreme Learning Machine Based on Wolf Optimization Algorithm for Predicting Beidou Satellite Clock Errors

作　　者：史红金 SHI Hongjin(Guangzhou Southern Surveying and Mapping Technology Co.,Ltd.,Guangzhou 510655,China)

机构地区：[1]广州南方测绘科技股份有限公司,广东广州510655

出　　处：《测绘与空间地理信息》2025年第2期73-76,共4页Geomatics & Spatial Information Technology

摘　　要：为了提高北斗卫星钟差预报精度,提出了一种基于狼群算法优化极限学习机的北斗卫星钟差预报方法。利用WPA算法对ELM进行参数优化,建立了基于WPA-ELM的北斗卫星钟差预报模型,采用卫星实际钟差数据进行仿真分析,并与其他方法对比,结果表明,WPA-ELM模型的均方根误差和平均相对误差分别为3.75%和1.236,预测误差低于另外两种方法,验证了本文提出的卫星钟差预报方法的正确性和实用性。In order to improve the accuracy of Beidou satellite clock error prediction,an optimized extreme learning machine method based on wolf swarm algorithm for Beidou satellite clock error prediction is proposed.Using the WPA algorithm to optimize the parameters of ELM,a Beidou satellite clock error prediction model based on WPA-ELM was established.Actual satellite clock error data was used for simulation analysis and compared with other methods.The results showed that the root mean square error and average relative error of WPA-ELM model were 3.75%and 1.236,respectively.The prediction error was lower than the other two methods,verifying the correctness and practicality of the satellite clock error prediction method proposed in this paper.

关 键 词：北斗卫星 钟差预报 狼群优化算法 极限学习机 

分 类 号：P228.49[天文地球—大地测量学与测量工程]