Validation of the Rotation Algorithm for Earthquake Damage Estimation

Abstract

A rotation algorithm was previously developed by the authors to estimate residual displacements in single columns subjected to earthquake loading. The objective of this paper is to determine the optimal number of sensors and their placement. For that purpose, numerical models for single columns deflecting in single and double curvature were developed where the nonlinear behavior was verified using laboratory test data. The algorithm is applied to the numerical simulations and its capability to capture single and double curvature deformations is verified. The accuracy of the algorithm is tested and it is found that the predicted displacements are within 2% of the true displacements with 50% confidence and within 4% of the true displacements with 90% confidence. Guidelines are provided for the optimal number of sensors, their placement and the functional form to be used for the estimation of the displacements. In particular, it is found that four sensors are sufficient for this level of accuracy. In addition, at least one sensor should be placed in the vicinity of expected plastic hinge locations, such as the base of all columns and near the top for double curvature columns. It is also shown that a fourth order polynomial as an estimate for displacements results in the aforementioned level of accuracy. The presented algorithm serves as a simple, accurate and reliable option for rapid damage assessment after a major event due to its ability to be embedded in the microprocessors of the sensors and the low cost of the sensors involved.