利用某岩质边坡2002年—2011年的变形监测资料，结合区域地质与气象背景，运用莱茵达准则选定有效的监测数据，分别建立了GM（1,1）模型和趋势曲线预测模型；最后，基于最优加权组合原理，建立了边坡变形的最优加权组合模型。运用组合模型对该岩质边坡的变形进行了拟合和变形预测，模拟变形趋势与实际变形趋势对比结果表明，组合模型的预测精度高于任何单一模型的拟合精度，证明该组合模型合理、可靠。

Based on the deformation monitoring data of a rock slope from 2002 to 2011 and the geological and meteorological conditions of the area, "PauTa" rule was used to select the effective monitoring data, and the GM (1,1) model and trend curve prediction model were developed. The optimal weighted combination model of slope deformation was established based on the optimal weighted combination principle. The combination model was used to predict the slope deformation. The comparison between the predicted and actual deformation trend suggested that the prediction accuracy of the combination model is higher than that of any single model, and therefore the combination model is more reasonable and reliable.

国家自然科学基金（51279091）；三峡大学研究生科研创新基金（2012CX019）；三峡大学硕士学位论文培优基金项目（2013PY015）；三峡大学土木与建筑学院硕士研究生培优基金（PY201303）