非等间距改进灰色模型在基坑变形预测中的应用

Application of Improved Non-equidistance Gray Model in Foundation Pits Deformation Prediction

基坑受开挖和人为因素的影响会在短时间内发生较大形变,现有的非等间距灰色模型不能识别处理突变数据,导致其在基坑中适用性低,精度较差。针对这种情况,本文提出了基于改进赋权处理的动态灰色模型和基于小波等距处理的动态灰色模型,结合工程实例,计算结果表明,这两种改进方法克服了现有模型的缺陷,降低了观测序列中含有的较多突变数据的影响,在基坑工程中变形监测的分析预报有较好的结果。

The foundation pits will suffer large deformation in a short time due to the influence of excavation and human factors.Existing non-equidistance gray model is unable to handle mutation data,which leads to the low applicability and poor precision in the foundation pits.In view of this situation,this paper proposes two improved models,Dynamic gray model based on improved weighting and Dynamic gray model based on wavelet equidistance.The calculation results of the engineering examples show that these two improved models can overcome the defects of the existing model,reduce the influence of many mutation data in the observation sequence,and function better in the analysis and prediction of foundation pits engineering deformation monitoring.