最优非负变权组合模型在大坝变形中的应用

APPLICATION OF THE OPTIMAL NON-NEGATIVE VARIABLE WEIGHT COMBINATION MODEL FOR MONITORING DAM DEFORMATION

提出基于灰色GM(1,1)、BP神经网络和普通卡尔滤波3种单一模型的最优非负变权组合预测模型。该模型继承单一模型的优点,在一定程度上保证较优的局部预测值和较好的全局预测精度。与最优加权组合模型、各单一模型对比分析表明,该模型预测精度较高,均方根误差为0.11 mm,在大坝变形预测中具有一定的实用价值。

Dam deformation is nonlinear,non-stationary and random which makes it difficult to accurately pre- dict the deformation. Based on three kinds of single model, gray GM （ 1,1 ） , BP neural network and the common Carl filtering, the optimal non-negative variable weight combination forecasting model was proposed. The model inherited the advantages of each single model. It is optimal for local prediction and accuracy is higher for the global predic- tion. The calculation results were compared with the optimal weighted combination model and each single one. The results show that the model prediction accuracy is higher;the root mean square error is 0.11 mm. And it can be ap- plied to dam deformation prediction practically.