摘要
当信号中混有噪声时,使用小波阈值法去噪可以很好地抑制噪声。灰色模型是常用的沉降预测模型,但预测结果与实际工况存在差异,可采用恒正凹化和相邻均值法对原始数据进行预处理能较好地弥补这一缺点。以某建设项目建筑物形变监测为研究对象,基于小波去噪理论和灰色GM(1,1)理论,构建GM(1,1)模型和优化GM(1,1)模型,对重构去噪前后预测模型的预测结果和实测数据进行对比分析。结果表明,与传统GM(1,1)模型相比,优化GM(1,1)模型的原始数据与预测曲线拟合度较好,精度高于传统GM(1,1)模型,预测效果有较大提升;在小波去噪前后优化GM(1,1)模型预测效果方面,对于数据变化较为平稳的,小波去噪效果不明显,而对于数据变化有差异的,去噪后优化GM(1,1)模型的精度有较大提高。
When the signal is mixed with noise,it can be well denoised the noise by using wavelet threshold method.The gray model is used commonly,but the prediction results are different from the actual working conditions.The constant positive concave treatment and the adjacent mean method can preprocess the original data,which will make up for this shortcoming much better.Taking building deformation monitoring of a construction project as the research object,this paper constructs GM(1,1)model and optimized GM(1,1)model based on wavelet denoising theory and grey GM(1,1)theory to compare the prediction results of the prediction model before and after reconstruction and denoising with the measured data.The results show that compared with the traditional GM(1,1)model,optimized GM(1,1)model has a better fit between the original data and the predicted curve,the accuracy of optimized GM(1,1)model is greatly improved,and it is higher than that of traditional GM(1,1)model;in terms of the prediction effect of the optimized GM(1,1)model before and after wavelet denoising,the wavelet denoising effect is not obvious for data changes that are relatively stable,and for different data changes,the accuracy of the optimized GM(1,1)model after denoising is greatly improved.
作者
范飞玲
黄基峻
FAN Feiling;HUANG Jijun(Qingyuan County Planning,Surveying and Mapping Design Institute,Lishui,Zhejiang,323800,China;Guangxi Mechanical and Electrical Industry School,Nanning,Guangxi,530023,China)
出处
《测绘标准化》
2022年第2期85-90,共6页
Standardization of Surveying and Mapping
作者简介
第一作者:范飞玲,工程师,注册测绘师,现主要从事工程测量。
