Optimal Sensor Placement for Structural Health Monitoring Under Budget Constraints

Sensor placement for structural health monitoring is fundamentally a budget-constrained optimization problem. This paper proposes a bi-objective formulation that simultaneously maximizes modal observability and minimizes information redundancy, solved by an improved greedy algorithm with theoretical submodularity guarantees.

研究背景

传感器布置需在有限的预算下最大化监测系统的信息获取能力，同时避免传感器之间的信息冗余。

方法

提出双目标优化模型：

• 目标一：最大化模态可观测性（基于 Fisher 信息矩阵）；
• 目标二：最小化传感器间的信息冗余（基于互信息度量）。

证明该目标函数具有子模性，并据此设计具有近似保证的改进贪心算法。

结论

在大跨空间结构与连续梁桥两个案例上，所提方法在相同传感器数量下模态置信准则指标优于传统有效独立法，且在传感器失效场景下表现出更好的鲁棒性。

@article{H.Zhao20252025-sen,
  title   = {Optimal Sensor Placement for Structural Health Monitoring Under Budget Constraints},
  author  = {H. Zhao and J. Wu},
  journal = {Mechanical Systems and Signal Processing},
  year    = {2025},
  doi     = {10.1016/j.ymssp.2025.100003},
}