摘要
为使地铁隧道在施工中沉降监测数据具有一定的预见性,分别采用了BP神经网络改进算法的预测模型、传统BP神经网络预模型以及基于时间序列的三次指数平滑法预测模型对地铁隧道施工中的沉降监测数据进行了预测。对其预测结果进行分析,得出了BP神经网络改进算法模型预测精度优于传统BP神经网络模型以及基于时间序列的三次指数平滑法模型预测精度的结论。
The Subway tunnel subsidence monitoring data were predicted to make it has some predictability in con-struction by the improved algorithm of BP neural network prediction model,the traditional BP neural network prediction model and the cubic exponential smoothing prediction model based on time series respectively. And analysis of theirs prediction results,it gets the conclusion that the improved algorithm of BP neural network model batter than the tradition-al BP neural network model and the cubic exponential smoothing prediction model based on time series about prediction accuracy. It has reference value for the early warning and forecast of subway monitoring.
出处
《城市勘测》
2015年第3期148-150,157,共4页
Urban Geotechnical Investigation & Surveying
关键词
神经网络
时间序列
沉降监测
数据处理
neural networks
time series
subsidence monitoring
data processing
