钢箱梁悬索桥颤振导数和颤振临界风速预测  

Prediction on Critical Flutter Wind Speed and Flutter Derivatives of Slotted Steel Box Girder Suspension Bridge

作　　者：蒋亚琼[1] 邵亚会[2] 

机构地区：[1]安徽新华学院土木工程系 [2]合肥工业大学土木与水利工程学院,安徽合肥230009

出　　处：《阴山学刊（自然科学版）》2016年第1期30-34,共5页Yinshan Academic Journal(Natural Science Edition)

基　　金：国家自然科学基金委重大研究计划重点研究项目(90715039);安徽省高校自然科学研究项目(KJ2015A336)

摘　　要：21世纪是跨海大桥建造的新时代,钢箱梁悬索桥作为首选桥型,其空气动力稳定性的研究至关重要。本文以西堠门跨海大桥钢箱梁悬索桥为工程依托,基于风洞试验数据,对不同槽宽比和不同稳定板高度的多个风洞试验工况的颤振临界风速进行了研究:鉴于BP神经网络的强鲁棒性,实现了颤振临界风速的人工神经网络建立、训练和预测;完成了BP神经网络在颤振导数预测中的应用。最终研究认为:预测的颤振临界风速和颤振导数经与已有风洞试验数据进行对比,吻合较好;本文方法可以用于开槽箱梁桥的颤振临界风速和颤振导数的预测,进而为箱梁的选型提供帮助,节省风洞试验的成本。21 Century is a new era of building oversea bridge. As the preferred type of oversea bridge, it is im- portant to study the aerodynamic stability of steel box girder suspension bridge. Relied on the project of the steel box girder suspension bridge in Xihoumen Bridge and the multiple working conditions of wind tunnel test with differ- ent slot width ratio and different stable plate height, the flutter critical wind speeds were studied. With strong ro- bustness of BP neural network, the establishment, training and forecast of the artificial neural network for the flutter critical wind speed was realized. The application of BP neural network in prediction of flutter derivatives is accom- plished. The results prove that the predicted critical flutter wind speed and flutter derivatives are well agree with the wind tunnel test data. This method can be used for the prediction of critical flutter and flutter derivatives of the slot- ted box girder bridge, and it can also provide the help for the selection of box girder, thus saving the cost of wind tunnel test.

关 键 词：钢箱梁悬索桥 颤振导数 颤振临界风速 BP神经网络 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] U441[自动化与计算机技术—控制科学与工程]