Parameter Identification and Damage Detection Using Structural Optimization and Static Response Data

Abstract

A new parameter identification method and algorithm will be presented that utilizes structural optimization concepts to correctly identify the stiffness in linear elastic models of civil structures. Displacement measurements resulting from applied static point loads are used as constraints in an optimization algorithm that employs optimality criterion methods to extract the cross sectional properties of elements within a mathematical model of a structure. Through simulated examples the algorithm will correctly identify stiffness parameters in a 10-bar truss structure and a two bay, two story moment frame. Once the parameter identification process has been illustrated in each example, the same algorithm and procedure is used to detect areas of simulated damage. Changes in the cross sectional properties of elements in the mathematical model of the structure are used to locate and quantify damage. Changes in cross section are used to simulate damage such as brittle fracture of steel members, spalling of concrete structures, and decay of wood structures. Lastly, the effects of measurement error on the damage detection results will be explored.