Qualitative and quantitative damage detection algorithm for structures using pattern classification and sensitivity analysis

Structural identification for health monitoring involves comparison of changes in structural properties or response, and it can be viewed as pattern classification problems. The fundamental idea of the pattern classification approach is to use training data to determine the classifiers to evaluate the categories of die test data. However, a very large database is required to store training data for complicated damage cases if no technique to reduce the size is used. This paper presents an approach with the purpose of using the least possible sensors. The structural damage location and damage extent are identified respectively by two different methods of pattern classification, that are, the Parzen-window method for the structural damage location firstly and the feed-forward back-propagation neural network used to identify damage extent secondly. The results of numerical simulations show that our proposed approach can indeed identify the structural damage using small number of sensors. Finally, a series of vibration experiments for the 5-story shear frame structure were performed to verify the performance of our proposed approach. The results show that for shear buildings, damage degree and extent can be successfully determined through measuring the frequency change.