基于BP神经网络学习模型对桥梁变形数据处理研究

Study on Data Processing of Bridge Deformation Based on BP Neural Network Learning Model

桥梁的变形预测对桥梁的安全性能研究具有重大意义。利用神经网络的自主学习能力,建立了BP神经网络学习模型,对已测得实际数据进行学习处理,修正权值,预测未来数据变化。结合MATLAB8.0软件进行了程序流程图设计与程序实现。以杭州湾跨海大桥为例,选取前10期检测数据作为学习样本,选取后5期检测数据作为预测期望值,与传统GM模型进行对比,证明了BP学习模型的可靠性与高精度,可以用来对大桥的变形量进行良好的预测。

Bridge deformation prediction is of great significance to the safety performance of the bridge. By using the neural network autonomous learning ability, the BP neural network learning model is established, and the measured data are processed to correct the weight. The program flow chart design and program implementation are combined with MATLABS. 0 software. In Hangzhou Bay Cross Sea Bridge as an example, select 10 test data as learning samples and selected 5 stage of testing data as a predictor of expectations, is compared with the traditional GM （ 1, 1） model, to prove the BP learning model of reliability and high precision and Can be used to good prediction of the deformation of the bridge.