In this paper we describe a new algorithm for the detection, location, and estimation of damage on a space truss using parameter identification techniques. The concept is to start with a model of the truss which is known to accurately model the healthy truss. This healthy model of the truss is periodically updated using dynamic response data. When damage has occurred, a damage model of the truss is constructed from dynamic response data using an identification algorithm. By comparing the parameters in the truth model with the parameters of the damage model, the damage is detected, located, and the extent of the damage accessed. The algorithm proposed here uses an explicit model of the damage to the truss. This description of the damage allows the algorithm search over a smaller set of possible damage models with attendant increase in performance. In addition to locating the damage, an estimate of the damage is also given. The performance of this algorithm for locating and estimating damage is presented in the form of three example problems including a 44 dof planar truss.
[1]
Scott Hendricks,et al.
Damage detection and location in large space trusses
,
1988
.
[2]
David C. Zimmerman,et al.
Structural damage detection using a subspace rotation algorithm
,
1992
.
[3]
Suzanne Weaver Smith,et al.
Experiments for locating damaged truss members in a truss structure
,
1991
.
[4]
D. C. Zimmerman,et al.
An inverse problem approach for structural damage detection - Finite element model refinement
,
1992
.
[5]
David C. Zimmerman,et al.
Model Refinement and Damage Location for Intelligent Structures
,
1992
.
[6]
A. Kabe.
Stiffness matrix adjustment using mode data
,
1985
.
[7]
E.Y. Shapiro,et al.
Eigenstructure Assignment for Linear Systems
,
1983,
IEEE Transactions on Aerospace and Electronic Systems.