基于滑动多项式拟合模型的建筑物垂直沉降预测

多项式拟合是预测建筑物垂直沉降时应用最广泛的模型之一，但模型受外界影响较大。为了更加充分的利用观测信息，减少外界因素影响，提出了滑动多项式拟合模型。该模型要求参与建模的数据个数保持不变，以新观测的数据取代旧有的数据，从而获得新的拟合参数，预测沉降量，并利用秩和检验对模型进行了检验。通过实例验证，得出滑动模型预测效果明显优于常规模型预测效果的结论。

Prediction of Building Vertical Settlement Based on Slide Polynomial Fitting Model

Multinomial fitting model is one of the most widely used models in the prediction of building settlement,but the external factor has a serious effect on the model.To curtail the extra effect,and uses the measured sufficiently,slide multinomial fitting model is put forward.The model requests the number of data for the model unchanged with new data replaced the old data,so the new parameter is gained and the model has been rank tested.The conclusion of slide model prediction effect being excelled general model prediction effect is gotten validated through the instance.