A Rough Set-Based Revised Counter-Propagation Network Model for Structural Damage Identification

摘要

In order to make full use of redundant, complementary and uncertain informationrnand thus assess the structural health states from a structural health monitoringrnsystem, a new damage identification method by integrating with rough set andrnrevised counter-propagation network (RCPN) model is proposed in this paper. Inrnthis method, rough set is used to deal with data so as to reduce the uncertainties andrnthe spatial dimensions of data firstly; then the current CPN model is revised so as tornimprove the capabilities of processing uncertainties and classification, and thernRCPN model is used to identify damage. To validate the method proposed, sixrnpatterns from a steel frame are identified finally, and the effect of measurementrnnoise, of network models and of data without processing by rough set on damagernidentification results are also investigated. The results show that the proposedrnmethod not only reduces the spatial dimension of data, but also has preferablerndamage identification capability and robustness.