改进的GM-AR组合模型在地铁沉降预测中的应用  

Application of improved GM-AR combined model to the prediction of subway settlement

作　　者：成枢[1] 郭祥琳 冯东恒 CHENG Shu;GUO Xianglin;FENG Dongheng(College of Geomatics, Shandong University of Science and Technology, Qingdao 266590)

机构地区：[1]山东科技大学测绘科学与工程学院,山东青岛266590

出　　处：《测绘工程》2018年第5期59-63,共5页Engineering of Surveying and Mapping

摘　　要：灰色预测模型对沉降的整体趋势有很好地预测结果,时序模型针对随机性的数据进行预测,二者结合一定程度上可提高预测的精度。文中通过改进GM-AR模型并将其应用于地铁沉降预测中,同时与灰色预测模型和GM-AR组合模型的预测精度进行对比分析。结果表明,改进后GM-AR模型可以有较好的预测效果。The gray prediction model has a good prediction result for the overall trend of settlement,while the time series model makes a prediction in views of the random data,and the combination of the two models can improve the accuracy of the prediction to some extent.In this paper,the GM-AR model is improved and applied to the prediction of subway settlement,and furthermore compared with the gray prediction model and the GM-AR combined model in the prediction accuracy.The result shows that the improved GM-AR model has a higher accuracy in the prediction of subway settlement.

关 键 词：灰色模型 时间序列 GM-AR组合模型 地铁沉降 残差 

分 类 号：TU196[建筑科学—建筑理论]