Structural health monitoring based on the hybrid ant colony algorithm by using Hooke–Jeeves pattern search

Abstract

Structural health monitoring is crucial for the timely damage diagnosis of civil infrastructure. This paper explores the damage detection method based on the ant colony algorithm (ACO) by using Hooke–Jeeves (HJ) pattern search for intensification. The HJ is incorporated into the ACO to improve its performance in detecting damages. The damage is simulated by reducing the stiffness of the structural members, via elastic modulus reduction factor. Four civil engineering structures of varying complexity are analysed for low- and high-level damage scenarios to test the efficacy of the proposed approach. An inverse problem is formulated to minimise the objective function based on the frequency response function rather than using the frequency and mode-shape-based approach. The analysis results indicate that the proposed method can locate damages and identify their severity with higher precision than previously used GA, SPSO, and UPSO can.