摘要
为了提高变形监测数据预测的精度与可靠性,以及提高人工神经网络预测方法的稳定性,尝试将小波分析与BP神经网络相结合的小波神经网络应用于高层建筑物沉降监测数据处理中。综合小波分析与神经网络算法的优点,将良好的时频局域化特性和神经网络理论的自学习功能相结合,建立高层建筑物的小波神经网络变形预测分析模型。通过实验数据对比分析,小波神经网络用于高层建筑物沉降预测数据处理中可以得到更好的预测效果,预测稳定性及预测精度较高。
In order to improve the accuracy and reliability of prediction of deformation monitoring data,and improve the stability of artificial neural network prediction method,we try to combine the wavelet analysis with the BP neural network to apply the wavelet neural network in the data processing of high rise building subsidence monitoring.We integrated the advantages of wavelet analysis and neural network algorithm,and we will be good time-frequency localization characteristics,and combining the self-learning function of neural network theory,and then building high-rise buildings deformation of wavelet neural network forecasting analysis model.Through the comparison of experimental data analysis,wavelet neural network used in high-rise building subsidence prediction data processing can get better prediction effect,and the stability prediction and the prediction precision is higher.
出处
《北京测绘》
2016年第1期47-51,共5页
Beijing Surveying and Mapping
关键词
小波分析
神经网络
小波神经网络
沉降监测
the wavelet analysis
the neural network
wavelet neural network
subsidence monitoring
