Damage Detection in Structures Using Precursor Transformation Method

Abstract

In this paper, a new concept for damage detection and long-term health monitoring of structures is presented. The Precursor Transformation Method (PTM) is based on determining the causes (precursors) of change in the measured state of the structure under non-variable loading conditions (e.g., dead loads in bridges). The PTM concept addresses the inability of the current structural monitoring methods to discriminate, in structural behavior terms, the meaning of voluminous measured sensor data on a timely and cost effective basis. This method offers advantages in sensitivity and cost efficiency when compared to conventional vibration-based or parameter estimation methods. PTM was developed as part of a research project sponsored by the Federal Highway Administration on bridge stay cable condition assessment. Measured changes in the state of a structure (displacements, strains, internal forces) can be related to precursors through a transformation matrix. This matrix is formed by determining the patterns of change in the state of structure associated with externally imposed strains (temperatures) or displacements representing possible damage scenarios. A finite element model of the undamaged structure is used to calculate these patterns. The use of an undamaged model of the structure in determining damage patterns simplifies the calculation process significantly, while introducing some approximation in results. Theoretical derivations and special case studies indicate that these approximations are limited to second order effects, and in many cases well within measurement and calculation accuracies. Examples using simulated damages on two truss structures and a cable-stayed bridge are also presented.