基于预处理共轭梯度法的桥梁移动荷载识别

Moving Load Identification Based on Preconditioned Conjugate Gradient Method

时域法(TDM)是移动荷载识别最有效的方法之一,它通过桥梁时域测量响应,间接识别桥梁移动荷载。针对时域法(TDM)有时识别精度较低、受噪声干扰的影响较大等问题,提出采用预处理共轭梯度法(PCGM)识别桥梁移动荷载。以两轴时变移动荷载识别为例,进行数值仿真研究,得到不同噪声水平下预处理共轭梯度法和时域法的识别结果。通过比较2种方法的识别精度,可以发现预处理共轭梯度法具有识别精度高、抗噪能力强、收敛速度快、初始参数设置少等优点,特别当仅仅采用弯矩响应识别移动荷载时,其识别精度仍然很高,这为移动荷载识别方法的实际应用打下了基础。

The time domain method （TDM） is one of the most effective methods to identify moving vehicle dynamic loads from the measured responses of the bridge. However, some problems sometimes occur in the TDM, such as, lower identification accuracy, sensitivity to the noise involved in the measured responses of bridge, and so on. On the basis of the theory of moving force identification, the preconditioned conjugate gradient method （PCGM） is proposed in this paper. For two-axle time-varying moving loads of a vehicle under various noise levels, some illustrative simulation shows that the identified results by the PCGM have higher identification accuracy, stronger anti-noise capacity, faster convergence and easy selection of initial parameters. It is beneficial to the practical application of the moving force identification method, particularly for the cases of identification of moving loads only from the measured bending moments.