内容：
针对导航卫星钟差预报精度不高的问题,该文引入了GM-LSSVM钟差预报模型,采用全局寻优能力较强的遗传算法对模型的参数选取过程进行优化,避免模型陷入局部最优,从而改善了组合模型中惩罚因子和核函数参数选择的盲目性。最后选取国际GPS服务组织提供的卫星钟差数据,分别建立GM(1,1)模型、LSSVM模型、GM-LSSVM模型和遗传算法优化的GM-LSSVM模型进行短期钟差预报分析和仿真实验。仿真结果表明,优化后的模型预报精度小于1.3ns,精度比前3种模型提高了45%~60%,符合钟差预报的要求。Aiming at the problem that the accuracy of navigation satellite clock error is not high, this paper introduces the GM-LSSVM clock error prediction model, and adopts the genetic algorithm with strong global optimization ability to optimize the parameter selection process of the model to avoid the model's local optimal, Thus improving the blindness of the penalty factor and kernel function parameter selection in the combined model. Finally, GM (1) model, LSSVM model, GM-LSSVM model and GM-LSSVM model optimized by genetic algorithm are used to carry out short-term time difference forecast analysis and simulation experiment respectively by selecting the satellite clock error data provided by international GPS service organization. The simulation results show that the accuracy of the model is less than 1.3ns, and the accuracy is 45% ~ 60% higher than that of the first three models, which accords with the requirement of the clock error forecast.

文件列表（点击上边下载按钮，如果是垃圾文件请在下面评价差评或者投诉）: