Structural Damage Detection Through Finite Element Model Updating Using Evolutionary Algorithm

Abstract

Structural damage can be caused by a variety of factors, which include internal factors such as design flaws, construction errors, material deficiencies, and external factors such as earthquakes, overloading, environmental factors, and terms of service violations. As such damage can compromise the safety, functionality, and financial viability of a structure, it is essential to identify and address any issues quickly. The engineering community, including civil, mechanical, and aerospace engineers, is greatly interested in structural damage identification. Numerous methods, such as experiential and simulation-based techniques, have been developed to achieve this goal. However, accurately modeling a structure and developing a robust inverse algorithm for damage detection is a significant challenge. Finite element modeling is one of the widely used methods for detecting structural damage. Here, Model updating is a crucial process that involves adjusting a structural model to improve its accuracy and reliability. Various methods for updating Finite Element Models have been proposed, including sensitivity-based, direct, statistical, probabilistic, and iterative methods. However, traditional methods can be complex and time-consuming, leading researchers to explore the use of evolutionary algorithms to streamline the process. Evolutionary algorithms are a type of computational optimization technique that mimics the process of natural selection to find the best solution to a given problem. In the context of model updating, evolutionary algorithms can identify the optimal combination of model parameters that best matches measured data. There has been a growing interest in using evolutionary algorithms to update Finite Element Models for structural damage identification in recent years. This paper presents a case study on damage identification, specifically sectional loss and boundary condition rigidity, as an optimization problem. It demonstrates the use of an evolutionary algorithm to update a Finite Element model for structural damage detection and discusses the advantages of using this algorithm over traditional methods. The abstract also highlights some of the key challenges associated with using evolutionary algorithms for model updating, including the need for accurate and reliable data and the need to carefully tune algorithm parameters.