The shielding effect of the front pile-row on the ice force acting on the back pile-row is studied by ice force model tests. In the tests, the front pile-row is designed to model jacket legs and the back pile-row to m...The shielding effect of the front pile-row on the ice force acting on the back pile-row is studied by ice force model tests. In the tests, the front pile-row is designed to model jacket legs and the back pile-row to model the water resisting pipe-phalanx within the jacket. The shielding factor for ice force corresponding to different conditions are given in this paper. The research indicates that there are many factors, including the longitudinal and lateral spacing between the front and back pile-row, ice attacking angle and the ratio of pile diameter to ice thickness, that influence the shielding effect on ice force.展开更多
为更加准确有效地进行桥梁桩基托换工程的墩台沉降数值预测,针对传统BP(Back Propagation)神经网络随机赋值、收敛速度慢等问题,提出了多步滚动算法(Multi-step Rolling Algorithm,MRA)优化的BP神经网络预测模型。以南京市某隧道穿越高...为更加准确有效地进行桥梁桩基托换工程的墩台沉降数值预测,针对传统BP(Back Propagation)神经网络随机赋值、收敛速度慢等问题,提出了多步滚动算法(Multi-step Rolling Algorithm,MRA)优化的BP神经网络预测模型。以南京市某隧道穿越高架桥桩基础工程沉降监测为研究背景,采用MRA-BP神经网络针对桥墩西侧匝道下JC10-13、JC10-14最大沉降监测点进行沉降预测,并与传统BP预测模型对比,对预测结果进行准确度分析。结果表明:MRA-BP神经网络预测模型在JC10-13和JC10-14监测点拟合优度R^(2)的数值均在0.85左右,相比传统BP预测模型提高了0.33,均方误差MSE控制在0.04左右,预测5#桥墩西侧匝道的沉降最终将稳定在25.5mm。MRA-BP神经网络预测模型适用于桩基托换桥墩沉降预测,能够为桥梁墩台施工建设提供更可靠的预测值,为既有建筑和托换结构进行有效变形控制提供依据。展开更多
文摘The shielding effect of the front pile-row on the ice force acting on the back pile-row is studied by ice force model tests. In the tests, the front pile-row is designed to model jacket legs and the back pile-row to model the water resisting pipe-phalanx within the jacket. The shielding factor for ice force corresponding to different conditions are given in this paper. The research indicates that there are many factors, including the longitudinal and lateral spacing between the front and back pile-row, ice attacking angle and the ratio of pile diameter to ice thickness, that influence the shielding effect on ice force.
文摘为更加准确有效地进行桥梁桩基托换工程的墩台沉降数值预测,针对传统BP(Back Propagation)神经网络随机赋值、收敛速度慢等问题,提出了多步滚动算法(Multi-step Rolling Algorithm,MRA)优化的BP神经网络预测模型。以南京市某隧道穿越高架桥桩基础工程沉降监测为研究背景,采用MRA-BP神经网络针对桥墩西侧匝道下JC10-13、JC10-14最大沉降监测点进行沉降预测,并与传统BP预测模型对比,对预测结果进行准确度分析。结果表明:MRA-BP神经网络预测模型在JC10-13和JC10-14监测点拟合优度R^(2)的数值均在0.85左右,相比传统BP预测模型提高了0.33,均方误差MSE控制在0.04左右,预测5#桥墩西侧匝道的沉降最终将稳定在25.5mm。MRA-BP神经网络预测模型适用于桩基托换桥墩沉降预测,能够为桥梁墩台施工建设提供更可靠的预测值,为既有建筑和托换结构进行有效变形控制提供依据。
