This study approaches the problem for damage diagnosis and defect localization in structures applying a pattern recognition procedure combined with substructuring. The problems for damage detection and localization are first defined as pattern recognition (PR) problems for differentiating between two possible categories of damaged and non-damaged structures (substructures). The problem for feature extraction from frequency response functions is discussed. A stochastic PR procedure is introduced in the context of defect detection and localization problem. The method is demonstrated on a test case of a cantilevered beam. It shows rather accurate performance and low error with simulated and noise contaminated data.
|Title of host publication||Damage Assessment of Structures|
|Number of pages||11|
|Publication status||Published - 2001|
|Name||Key Engineering Materials|
|Publisher||Trans Tech Publications Ltd|
- pattern recognition
- mechanical engineering
- signal processing
Trendafilova, I. (2001). Pattern recognition methods for damage diagnosis in structures from vibration measurements. In Damage Assessment of Structures (Vol. 204-2, pp. 85-96). (Key Engineering Materials).