摘要
结合白果渡嘉陵江大桥施工线形控制的具体实践,采用BP神经网络进行连续刚构桥施工线形控制中的参数识别及预测工作。基于影响桥梁线形主要参数的截面尺寸、距离及标高建立神经网络系统,并对其进行计算训练样本、训练神经网络和网络仿真分析。运用神经网络仿真分析进行连续刚构桥施工线形的具体方法是,先计算当前施工状态的标高,再预测下一节段的标高。经过往复循环,逐一进行节段预测调整,从而指导连续刚构桥顺利施工。网络学习及仿真预测结果表明:该法对数据的处理及预测,在操作简单的基础上,分析结果具有较高的精度。该结论可推广到采用悬臂法施工的连续梁桥、拱桥、斜拉桥等桥型的施工线形控制工作及研究。
Combining the engineering practice of Baiguodu Jialing river bridge construction geometry control,this paper attempts to make a study of parameter prediction in construction geometry control of a continuous rigid frame bridge by BP neural network method.The establishment of neural network system model,the calculation of training swatch,the training of neural network and the simulation of network,studied by neural networks for the cross-sectional size of the girder,distance and elevation of the main parameters.The specific method for continuous rigid frame bridge construction geometry control by neural network simulating calculation is to first calculate the elevation under current construction state and then estimate the elevation of the next section.Through such repetitive cycles,each section is estimated and adjusted one by one to successfully guide the construction of continuous rigid frame bridge.The results obtained from the training and simulation of the neural network,indicate that the data processing with the method is easy for operation and has satisfactory accuracy for parameter prediction.
出处
《四川建筑科学研究》
北大核心
2011年第1期263-266,共4页
Sichuan Building Science
关键词
桥梁工程
连续刚构
神经网络
施工控制
参数预测
bridge engineering
continuous rigid frame bridge
neural network
construction control
parameter prediction
