基于小波去噪的灰色Verhulst模型在高铁路基沉降预测的应用  

Forecast Application in High Speed Railway Roadbed Based on Wavelet Denoising for Grey Verhulst Models

作　　者：洪雪倩 陈冠宇[1] 周超 周吕[2,3,4] 徐骏平 HONG Xueqian;CHEN Guanyu;ZHOU Chao;ZHOU Lyu;XU Junping(Zhejiang Brigade of Surveying and Mapping,Hangzhou 310030,China,2.School of Geodesy and Geomatics,Wuhan University,Wuhan 430079,China;3.Beijing Key Laboratory of Urban Spatial Information Engineering,Beijing Institute of Surveying and Mapping,Beijing 100038,Chin;4.Key Laboratory for Digital Land andResources of Jiangxi Province,East China University of Technology,Nanchang 330013,China;5.Hubei Tianshang Land Surveying and Planning Design Co.,Ltd.,Wuhan 430079,China)

机构地区：[1]浙江省测绘大队,浙江杭州310030 [2]武汉大学测绘学院,湖北武汉430079 [3]北京市测绘设计研究院城市空间信息工程北京市重点实验室,北京100038 [4]东华理工大学江西省数字国土重点实验室,江西南昌330013 [5]湖北天尚土地勘测规划设计有限公司,湖北武汉430079

出　　处：《测绘与空间地理信息》2018年第8期123-126,共4页Geomatics & Spatial Information Technology

基　　金：城市空间信息工程北京市重点实验室经费项目(2016204);东华理工大学江西省数字国土重点实验室开放研究项目(DLLJ201711)资助

摘　　要：高铁路基需严格控制工后不均匀沉降。鉴于高铁路基沉降预测值精度受观测噪声和预测拟合函数的影响,本文提出了基于小波函数去噪,对去噪数据进行灰色Verhulst模型预测的方法,并阐述了高铁路基沉降预测评价方法。通过工程实例对比分析了去噪灰色Verhulst模型、GM(1,1)模型、双曲线模型在沉降数据处理中的拟合精度和预测精度。结果表明:GM(1,1)模型拟合精度高,预测精度低,不适用于长期预测;双曲线法预测精度最低,预测曲线不包含路基饱和发展过程;小波去噪灰色Verhulst模型符合高铁路基沉降规律,预测精度高,可以广泛用于路基沉降预测。The High speed railway roadbed need to control the uneven settlement strictly after work. In view of the high speed railway subsidence prediction accuracy is affected by the observation noise and the prediction function,in this paper,we proposed prediction a method that based on wavelet denoising and grey Verhulst model prediction. We include the high speed railway foundation subsidence prediction evaluation method. Through engineering examples,we compared the fitting precision and prediction precision in subsidence data processing between the denoising grey Verhulst model,GM（ 1,1） model and hyperbolic model,and found that the GM（ 1,1）model fitting precision is high,but precision is low,which can not apply to long-term forecast. hyperbolic forecasting curve does not include the subgrade saturation development process,which has the minimum accuracy. Wavelet denoising grey Verhulst,whose shape changes conforms to high railway foundation settlement rules has the highest prediction accuracy,can be widely used in High speed railway roadbed prediction.

关 键 词：高铁路基沉降预测 小波去噪 灰色VERHULST模型 沉降变形评估 精度对比 

分 类 号：P25[天文地球—测绘科学与技术] TB22[天文地球—大地测量学与测量工程]