考虑到土石坝沉降变形因素的复杂性以及GM (1,1)灰色模型基于贫信息数据所表现出来的优势,由大坝实测数据拟合构造GM (1,1)模型。为了进一步提高大坝沉降变形的预测精度,对原始GM (1,1)模型分别进行了考虑不同数据对预测结果有不同影响的加权改进和利用指数函数变换来提高原始数据光滑度的改进。实例应用表明,加权改进的GM (1,1)模型和函数变换改进的GM (1,1)模型的预测精度均优于原始GM (1,1)模型。
Considering the complexity of earth and rockfill dam and the application of the gray model in the prediction of the dam deformation with poor information ,we constructed a GM (1 ,1) model based on the actual observation data of the dam .To increase the accuracy of prediction ,a weighted GM (1 ,1) model was put forward considering the effect of different data on the predic-ting results .Another improved GM (1 ,1) model was built with exponent function transformation to improve the smoothness of original datas .The pratical case showed that the prediction preci-sions of the two improved models are higher than that of the original one .