动态灰色预测模型在大坝变形监测及预报中的应用研究

Application of dynamic gray forecast model in dam deformation monitoring and forecast

详细地讨论了灰色预测模型GM(1,1)和动态灰色预测模型的基本内容及建模过程,并成功地将等维新息和等维灰数递补两种动态灰色预测模型应用于大坝变形的预测预报。实践证明,等维新息动态预测由于实时地加入了新的信息,提高了灰区间的白色度,预测效果最好;等维灰数递补动态预测利用了序列建模的结果,淡化了灰平面的灰度,使预测结果有所改善;GM(1,1)模型由于是静态地反映系统的变化趋势,预测的精度最低。因此,动态灰色预测模型在大坝变形的预测预报中比静态预测模型具有更高的应用价值。

This paper discussed the fundamental contents and the establishment courm of the GM （1, 1） forecast model and the dynamic gray forecast model. Two kinds of dynamic gray forecast model were successfully applied in the dam deformation forecast. They are the equal dimension and new information and the equal dimension and recruiting gray data dynamic GM（ 1,1 ） forecast model. It is proved through the example that the equal dimension and new information dynamic gray forecast model had the best effect in those three methods because it added the new information and increased the white grade of gray interzone. The equal dimension and recruiting gray data dynamic gray forecast model improved the forecast precision because it used the results of sequence modeling and the gray grade of gray plane was faded, and the GM（ 1,1 ） forecast model had the lowest forecast precision because it was the static model and could not represent the dynamic trend of system. Therefore, the dynamic gray forecast model had higher value than static model in dam deformation forecast.