摘要
为了对建筑物变形数据进行有效处理,预测变形量,为建筑物的安全运营保驾护航,本文将GM(1,1)模型、小波分析与神经网络结合,提出了一种灰色—小波神经网络模型,并借用Matlab软件实现了该预测模型。将构建的变形预测模型应用到建筑物变形监测数据处理中,首先使用GM(1,1)预处理数据,并且将预处理后的数据作为小波神经网络模型的输入数据进行变形数据预测。试验结果表明,相比于灰色预测模型与小波神经网络模型,组合模型的预测准确度与稳定性更高,并且随着训练样本的增加,预测结果与实际观测值越接近。
In order to effectively process the deformation data of buildings,predict the deformation and ensure the safe operation of buildings,a grey wavelet neural network model is proposed by combining GM(1,1)model,wavelet analysis and neural network,and the prediction model is realized by Matlab.The deformation prediction model is applied to the data processing of building deformation monitoring.GM(1,1)is used to preprocess the data and the preprocessed data is used as the input data of wavelet neural network model to predict the deformation data.The experimental results show that compared with the grey prediction model and wavelet neural network model,the combined model has higher accuracy and better stability,and with the increase of training samples,the prediction results are more closer to the actual observation values.
作者
王坤
WANG Kun(Ningbo Branch of CCCC Third Harbor Engineering Co.,Ltd.,Ningbo,Zhejiang,315200,China)
出处
《测绘技术装备》
2022年第1期28-32,共5页
Geomatics Technology and Equipment
关键词
灰色模型
小波分析
神经网络
建筑物变形监测
grey model
wavelet analysis
neural network
building deformation monitoring
