论文部分内容阅读
在时间序列模型中,等时间序列模型建模过程简单,预报准确,但观测时间是不等间隔的;三次样条插值可以对一定的区间进行有效内插,得到等时间间隔的拟合数据。针对两种模型的优点,该文结合等时间序列模型和三次样条插值进行高铁桥墩的沉降预测。首先通过三次样条插值将数据进行等时间间隔处理,然后利用得到的数据建立时间序列模型,最后通过三次样条插值求得相应观测时间的预测值。以某高铁桥墩沉降的观测点为例,分别进行原始数据时间序列建模以及三次样条插值和等时间序列组合模型建模。实验结果证明,组合模型的预测精度更高。
In the time series model, the modeling process of the isochronous model is simple and the forecasting is accurate, but the observation time is not equal; the cubic spline interpolation can effectively interpolate a certain interval and get the fitting data with the same time interval. In view of the advantages of the two models, this paper combines equal-time series model and cubic spline interpolation to predict the settlement of high-speed railway piers. Firstly, the data is processed at equal time intervals by cubic spline interpolation. Then the time series model is established by using the obtained data. Finally, the predicted value of the corresponding observation time is obtained by cubic spline interpolation. Taking the observation points of settlement of a high-speed railway bridge piers as an example, the time series modeling of raw data and the modeling of the combination of cubic spline interpolation and isochronous time series are respectively carried out. Experimental results show that the combined model has higher prediction accuracy.