论文部分内容阅读
在对深基坑墙体位移时序规律分析的基础上,提出了基于MATLAB 5.3平台的神经网络多步预测模型,可以一次预测多步施工的变形,达到及早调整基坑施工参数,完成变形控制的目的。对润扬大桥北锚碇深大基坑工程的施工实例分析显示,预测值和实测值基本吻合,表明该预测模型具有较高的精度。
Based on the analysis of the displacement regularity of deep foundation pit, a multi-step neural network prediction model based on MATLAB 5.3 platform is proposed, which can predict the deformation of multi-step construction at one time, and adjust the parameters of foundation pit as early as possible to complete the deformation control purpose. The analysis of the construction example of the deep foundation pit project of North Anchorage of Runyang Bridge shows that the predicted value is in good agreement with the measured value, indicating that the prediction model has high accuracy.