深基坑坑周土体水平位移预测模型的研究与改进  

Research and Improvement of the Horizontal Displacement Prediction Model of the Soil Mass Adjacent to Deep Foundation Pits

作　　者：王铁生[1] 冯康[1] 陈正威 

机构地区：[1]华北水利水电大学,河南郑州450011 [2]中水顾问集团贵阳勘测设计院,贵州贵阳550000

出　　处：《水利与建筑工程学报》2015年第5期185-189,共5页Journal of Water Resources and Architectural Engineering

摘　　要：深基坑工程的变形预测意义重大。为了更好的对其变形进行预测,以某大楼深基坑工程为例,针对深基坑坑周土体水平位移预测进行模型的建立与研究,建立了Verhulst模型,并运用小波理论对原始数据进行降噪处理,建立了基于小波的Verhulst模型,进一步得到降噪前后的预测结果。通过与GM(1,1)模型对比分析,得出了以下结论:降噪前Verhulst模型预测精度为一级,优于GM(1,1)模型;基于小波降噪的Verhulst模型使预测误差进一步减小;小波Verhulst模型可用于深基坑坑周土体水平位移的预测,并体现出了较好的优越性。The deformation prediction of deep foundation pits is of great significance .In order to better forecast the defor-mation ,the Verhulst model of a high-rise building foundation pit was established to predict the horizontal displacement of the soil mass around the pit .And then the original surveying data of the displacement was denoised with the wavelet theo-ry .Based on which ,a new wavelet denoised Verhulst model was constructed .By the comparsion of the prediction data of Verhulst models and the conventional GM （1 ,1 ） model ,it is drawn that the accuracy of the Verhulst model before denois-ing is classified as first grade ,which is better than that of the GM （1 ,1 ） model .Moreover the Verhulst model based on wavelet denoising can furtherly reduce the prediction error .Therefore wavelet denoised Verhulst model performs well in the prediction of the horizontal displacement of the soil mass around the foundation pits .

关 键 词：深基坑 VERHULST模型 GM(1 1)模型 小波降噪 水平位移预测 

分 类 号：TU433[建筑科学—岩土工程]