论文部分内容阅读
针对模型修正方法进行结构损伤时识别精度易受实测模态噪声干扰的问题,提出一种基于小波包降噪与有限元模型修正的结构损伤识别方法。在损伤识别的每一迭代步中,以结构有限元模型为基准,考虑真实模态与模态噪声在时频域内的差异,以实测模态与有限元模型计算模态的差值为对象进行小波包降噪处理;并利用降噪处理后的模态参数构造模型修正目标函数,进行有限元模型修正,经过修正迭代计算后可识别结构的损伤位置与程度。数值算例表明本文方法可明显提高有限元模型修正方法进行结构损伤识别的鲁棒性。
Aimed at the problem that the recognition accuracy of the model correction method is easily disturbed by the measured modal noise, a method of structural damage identification based on wavelet packet noise reduction and finite element model correction is proposed. In each iteration of the damage identification, taking the finite element model as a reference, the difference between the real modal and the modal noise in the time-frequency domain is considered, and the difference between the calculated modal of the measured modal and the finite element model is taken as the object Wavelet packet denoising; and using the modal parameters after denoising, the model is modified to correct the objective function, and the finite element model is modified. After corrected and iteratively calculated, the location and degree of the damage can be identified. Numerical examples show that this method can obviously improve the robustness of the structural damage identification by the finite element model updating method.