优化GM(1,1)模型在桥梁静载试验中应用

摘要:

设计一种适用于桥梁静载荷试验中极不完备数据条件下的优化GM(1,1)模型。使用3个前置神经网络模块替换传统模型中的微分拟合过程,利用神经网络的回归能力对其进行优化,使用1个后置神经网络模块结合1个数据解模糊模块替换传统模型的数据还原过程,其他数据处理方式沿用传统模型的数据处理方式。经过仿真计算,发现不同桥梁设计规模下,优化模型的标准差显著低于传统模型,证实优化模型具有较显著的算力提升。

Abstract:

An optimized GM (1,1) model is designed for the bridge static load test under the condition of extremely incomplete data. Three pre neural network modules are used to replace the differential fitting process in the traditional model, and the regression ability of neural network is used to optimize it. One post neural network module is used to replace the data restoration process of traditional model with one data solving fuzzy module. Other data processing methods are based on the traditional model. Through simulation, it is found that the standard deviation of the optimization model is significantly lower than that of the traditional model under different bridge design scale, which proves that the optimization model has a significant calculation force improvement.