| 基坑变形小波去噪及BP神经网络组合预测模型研究 |
| |
| 引用本文： | 梁小龙,齐二恒,王强昆,王建业.基坑变形小波去噪及BP神经网络组合预测模型研究[J].测绘与空间地理信息,2021,44(1):189-192,195. |
| |
| 作者姓名： | 梁小龙 齐二恒 王强昆 王建业 |
| |
| 作者单位： | 机械工业勘察设计研究院有限公司,陕西 西安710043;机械工业勘察设计研究院有限公司,陕西 西安710043;机械工业勘察设计研究院有限公司,陕西 西安710043;机械工业勘察设计研究院有限公司,陕西 西安710043 |
| |
| 摘 要： | 以某市轨道交通1号线地铁站基坑观测数据为例,开展了小波及BP神经网络预测模型的研究。首先采用小波阈值去噪方式对纵向观测线实际观测数据进行去噪处理,依据信噪比最高以及均方根最小判别原则进行判别,实验表明,小波1层分解、软阈值方式、sym4小波基函数、rigrsure阈值原则、scal=sln为最佳组合方式。然后,给出基坑变形小波-BP神经网络组合预测模型。最后对小波去噪前后的数据进行BP神经网络预测模型预测处理并与小波变换神经网络预测模型预测数据进行对比分析,结果表明小波变换神经网络预测模型预测精度最高。
|
| 关 键 词： | 基坑变形监测 小波分析 BP神经网络 组合模型 |
Study on the Combining Prediction Model of Wavelet Denoising and BP Neural Network for Foundation Pit Deformation |
| |
| LIANG Xiaolong,QI Erheng,WANG Qiangkun,WANG Jianye.Study on the Combining Prediction Model of Wavelet Denoising and BP Neural Network for Foundation Pit Deformation[J].Geomatics & Spatial Information Technology,2021,44(1):189-192,195. |
| |
| Authors: | LIANG Xiaolong QI Erheng WANG Qiangkun WANG Jianye |
| |
| Institution: | (Mechanical Industry Survey and Design Institute Co.,Ltd.,Xi′an 710043,China) |
| |
| Abstract: | Taking the foundation pit observation data of a subway station on Hefei metro line 1 as an example,the prediction model of wavelet denoising and BP neural network was studied.Firstly,wavelet threshold denoising method was adopted to denoise the observation data of the longitudinal observation line,and the highest signal-to-noise ratio and the minimum root-mean-square criterion were used to discriminate.The experiment showed that wavelet layer 1 decomposition,soft threshold method,sym4 wavelet basis function,rigrsure threshold principle,and scal=sln were the best combination.Then,the combined wavelet-BP neural network prediction model for foundation pit deformation was presented.Finally,the BP neural network prediction model before and after wavelet de-noising and wavelet transform neural network prediction model were compared and analyzed.The results showed that the wavelet transform neural network prediction model has the highest prediction accuracy. |
| |
| Keywords: | foundation pit deformation monitoring wavelet analysis BP neural network combination model |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
