...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Mechanical systems and signal processing >Novel Laplacian Scheme And Multiresolution Modal Curvatures For Structural Damage Identification
【24h】

Novel Laplacian Scheme And Multiresolution Modal Curvatures For Structural Damage Identification

机译：结构损伤识别的新型拉普拉斯方案和多分辨率模态曲率

获取原文
获取原文并翻译 | 示例
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Modal curvature is more sensitive to structural damage than directly measured mode shape, and the standard Laplace operator is commonly used to acquire the modal curvatures from the mode shapes. However, the standard Laplace operator is very prone to noise, which often leads to the degraded modal curvatures incapable of identifying damage. To overcome this problem, a novel Laplacian scheme is proposed, from which an improved damage identification algorithm is developed. The proposed step-by-step procedures in the algorithm include: (1) By progressively upsampling the standard Laplace operator, a new Laplace operator is constructed, from which a Laplace operator array is formed; (2) by applying the Laplace operator array to the retrieved mode shape of a damaged structure, the multiresolution curvature mode shapes are produced, on which the damage trait, previously shadowed under the standard Laplace operator, can be revealed by a ridge of multiresolution modal curvatures; (3) a Gaussian filter is then incorporated into the new Laplace operator to produce a more versatile Laplace operator with properties of both the smoothness and differential capabilities, in which the damage feature is effectively strengthened; and (4) a smoothened nonlinear energy operator is introduced to further enhance the damage feature by eliminating the trend interference of the multiresolution modal curvatures, and it results in a significantly improved damage trait. The proposed algorithm is tested using the data generated by an analytical crack beam model, and its applicability is validated with an experimental program of a delaminated composite beam using scanning laser vibrometer (SLV) to acquire mode shap'es. The results are compared in each step, showing increasing degree of improvement for damage effect. Numerical and experimental results demonstrate that the proposed novel Laplacian scheme provides a promising damage identification algorithm, which exhibits apparent advantages (e.g., high-noise insusceptibility, insightful in damage revealment, and visualized damage presentation) over the standard Laplace operator.
机译：模态曲率比直接测量的模态形状对结构破坏更敏感，通常使用标准的拉普拉斯算子从模态形状获取模态曲率。但是，标准的Laplace算子非常容易产生噪声，这通常会导致模态曲率下降，无法识别损坏。为了解决这个问题，提出了一种新颖的拉普拉斯算子方案，从中提出了一种改进的损伤识别算法。该算法中建议的分步过程包括：（1）通过逐步对标准拉普拉斯算子进行升采样，构造一个新的拉普拉斯算子，由此形成拉普拉斯算子数组； （2）通过将拉普拉斯算子数组应用于受损结构的检索模式形状，可以生成多分辨率曲率模式形状，在该模型上，以前由标准拉普拉斯算子遮蔽的损伤特征可以通过多分辨率模态的脊线显示出来。弯曲（3）然后将高斯滤波器合并到新的拉普拉斯算子中，以产生一种兼具平滑性和微分能力的特性的，用途更广的拉普拉斯算子，其中，损伤特征得到有效增强； （4）引入了平滑的非线性能量算子，通过消除多分辨率模态曲率的趋势干扰来进一步增强损伤特征，从而显着改善了损伤特征。使用解析裂纹束模型生成的数据对提出的算法进行了测试，并通过使用扫描激光振动计（SLV）的分层复合束的实验程序验证了模型的可塑性，验证了其适用性。在每个步骤中对结果进行比较，显示出对损伤效果的改善程度不断提高。数值和实验结果表明，所提出的新颖Laplacian方案提供了一种有希望的损伤识别算法，与标准的Laplace算子相比，该算法具有明显的优势（例如，高噪声敏感性，洞察力和可视化的损伤表现）。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号