Aimied at the low smoothness of the original discrete data of grey GM (1, 1) model and the unusual background frequency value in the concrete construction, based on the traditional model, the original data were transformed by a function and the background value was adjusted according to the construction conditions, in order to improve the accuracy of the model.The improved model was applied in deep excavation construction of Metro Line 3 in Nanchang, to predict the settlement of railway subgrade that is close to the deep excavation.The results show that the original data that been transformed by the function to adjust the smoothness of the data, can meet the level requirements.Predicting the settlement by construction conditions, the background value can be adjusted, the effect of abnormal background value on accuracy was reduced.The predicted value and the actual monitoring value fit well.%针对灰色GM(1,1)模型原始离散数据光滑度低以及在施工中背景值频现异常的问题,在传统模型的基础上,对原始数据进行函数变换,并按施工工况调整背景值,提高了模型精度.将改进的模型应用到南昌地铁3号线深基坑施工中,对临近铁路路基沉降进行建模预测.结果表明,利用函数对原始数据进行变换,调整数据光滑度,可使其符合级比要求;按照分工况预测的方法根据施工进度调整背景值,减小了异常背景值对精度的影响,预测值与实际监测值拟合度较好.
展开▼
