...
  • 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 >The effect of prediction error correlation on optimal sensor placement in structural dynamics
【24h】

The effect of prediction error correlation on optimal sensor placement in structural dynamics

机译：预测误差相关性对结构动力学中最佳传感器放置的影响

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

The problem of estimating the optima! sensor locations for parameter estimation in structural dynamics is re-visited. The effect of spatially correlated prediction errors on the optimal sensor placement is investigated. The information entropy is used as a performance measure of the sensor configuration. The optimal sensor location is formulated as an optimization problem involving discrete-valued variables, which is solved using computationally efficient sequential sensor placement algorithms. Asymptotic estimates for the information entropy are used to develop useful properties that provide insight into the dependence of the information entropy on the number and location of sensors. A theoretical analysis shows that the spatial correlation length of the prediction errors controls the minimum distance between the sensors and should be taken into account when designing optimal sensor locations with potential sensor distances up to the order of the characteristic length of the dynamic problem considered. Implementation issues for modal identification and structural-related model parameter estimation are addressed. Theoretical and computational developments are illustrated by designing the optimal sensor configurations for a continuous beam model, a discrete chain-like stiffness-mass model and a finite element model of a footbridge in Wetteren (Belgium). Results point out the crucial effect the spatial correlation of the prediction errors have on the design of optimal sensor locations for structural dynamics applications, revealing simultaneously potential inadequacies of spatially uncorrelated prediction errors models.
机译：估计最优值的问题！重新访问用于结构动力学参数估计的传感器位置。研究了空间相关的预测误差对最佳传感器位置的影响。信息熵用作传感器配置的性能度量。传感器的最佳位置被公式化为涉及离散值变量的优化问题，可使用计算效率高的顺序传感器放置算法解决该问题。信息熵的渐近估计用于开发有用的属性，这些属性可洞悉信息熵对传感器的数量和位置的依赖性。理论分析表明，预测误差的空间相关长度控制着传感器之间的最小距离，在设计最佳传感器位置时，应将其考虑在内，而传感器位置的潜在距离可以达到所考虑的动态问题的特征长度的数量级。解决了模态识别和与结构相关的模型参数估计的实现问题。通过设计连续梁模型，离散链状刚度质量模型和比利时Wetteren人行桥的有限元模型的最佳传感器配置，说明了理论和计算方面的发展。结果指出了预测误差的空间相关性对结构动力学应用的最佳传感器位置设计的关键影响，同时揭示了空间不相关的预测误差模型的潜在不足之处。

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