摘要
提出一种基于小波分析和回归估计的变形预测模型,基本思想是根据小波分析的优点,对监测数据进行小波分解,通过对小波系数进行估计得到回归估计函数。由于在小波分解过程中对小波系数作用于阈值,即对监测过程受到的干扰进行了消除或降低,从而使建立的预测模型更接近于变形体规律,预测更准确。实例计算表明,这种模型是切实可行的。
A deformation prediction model based on wavelet transform and regression estimation is proposed in this paper.As the wavelet coefficient is affected by threshold during wavelet decomposition,that is,noise is restrained,the model is closed to the deformation rule and prediction is more exact.The results show that the model is effective and suited to the prediction of deformation monitoring.
出处
《测绘科学》
CSCD
北大核心
2006年第6期125-126,84,共3页
Science of Surveying and Mapping
基金
淮海工学院引进人才科研启动基金资助项目
关键词
预测
小波
回归
阈值
prediction
wavelet
regression
threshold
