The Fractional Diagnosis of Multi-Span Continuous Bridge’s Structural Damage Based on Neural Network and Genetic Algorithm

Abstract

The natural environmental erosion and human factors such as the impact of traffic accidents, crack propagation, concrete carbonation and etc, make the bridge’s damage more serious. Therefore, the bridge damage diagnosis has become a hot field of bridge engineering issues. This paper put forward the fractional diagnosis method of multi-span bridge structure reflecting the structural cracks and carbonation damage. In this paper, adopting the optimization equivalent method, the finite element model of damaged structure is set up according to the damaging characteristic of multi-span continuous bridge structure. A damage index of strain mode with practical meanings is adopted which can reflect local damage. Basing on this index, fractional-step detection method of structural damage is presented. The first step is to identify the damage region, then locate the detailed damage location and degree; Performance of the proposed damage detection approach is demonstrated with analysis of a multi-span continuous bridge. The result turns up trumps.