论文部分内容阅读
根据轨道几何不平顺的发展特性,在灰色预测理论的基础上,考虑模型参数随时间的变化,并优化背景值,建立以轨道几何不平顺检测数据为时间序列的非等时距灰色时变参数模型。为更好地描述轨道几何不平顺影响因素间复杂的函数关系,提高模型拟合和预测精度,基于残差分析引入周期性函数,对模型进行组合修正。应用此模型对轨道质量指数TQI数据进行分析预测,并对其精度进行检验。结果表明:模型能较好地反映轨道质量随时间发展的随机波动特征,拟合、预测精度高,适合进行中长期预测,可为了解和掌握轨道质量状态的发展规律提供新的方法。
Based on the gray prediction theory, considering the variation of the model parameters over time and optimizing the background values, the gray time-varying parameters with non-isochronal interval based on the irregular track detection data are established model. In order to better describe the complex function of the influential factors of orbital irregularities and improve the accuracy of model fitting and prediction, the periodic function is introduced based on the residual analysis to modify the model. This model is used to analyze or predict the track quality index (TQI) data and verify its accuracy. The results show that the model can reflect the stochastic fluctuation characteristics of orbital quality over time. The fitting and prediction precision is high, which is suitable for medium and long term prediction. It can provide a new method for understanding and mastering the law of the development of orbital mass state.