Soft Computing Based Multilevel Strategy for Bridge Integrity Monitoring

Abstract

In recent years, structural integrity monitoring has become increasingly important in structural engineering and construction management. It represents an important tool for the assessment of the dependability of existing complex structural systems as it integrates, in a unified perspective, advanced engineering analyses and experimental data processing. In the first part of this work the concepts of dependability and structural integrity are discussed and it is shown that an effective integrity assessment needs advanced computational methods. For this purpose, soft computing methods have shown to be very useful. In particular, in this work the neural networks model is chosen and successfully improved by applying the Bayesian inference at four hierarchical levels: for training, optimization of the regularization terms, data-based model selection, and evaluation of the relative importance of different inputs. In the second part of the article, Bayesian neural networks are..

Physical Parameter Identification of Structural Systems with Hysteretic Pinching

“…The nonparametric methods maps the inputs and outputs to the structure by a set of equations that may not have any explicit physical meaning (Lozano Galant et al, 2013). Artificial neural networks (ANN) are one of the commonly used nonparametric methods for system identification (Adeli and Jiang, 2006;Graf et al, 2012) and damage detection (Jiang and Adeli, 2007;Arangio and Bontempi, 2010;Osornio-Rios et al, 2012;Story and Fry, 2014), ANN methods can approximate arbitrary continuous function and provide an efficient mechanism for structural identification using the functional mapping between the inputs and outputs of the structure without any knowledge about the internal structural model (Wu et al, 2002). However, a poorly trained model leads to inaccurate results when the training data is incomplete or corrupted (Sirca and Adeli, 2012), and the local minima and convergent efficiency are also the problems of using ANNs (Hung et al, 2003).…”

A web application system for structural modal identification

“…Furthermore, through structural dynamic analysis it is possible to perform a quite extensive structural evaluation concerning reliability, vulnerability or even life cycle (Torres and Ruiz, 2007). Novel techniques are being developed in order to better identify and assess the integrity of structural systems by means of clustering analyses (Arangio and Bontempi, 2010;Carden and Brownjohn, 2008), neural network schemes (Jiang and Adeli, 2005), etc. However, the necessity of performing a modal identification procedure still remains very important.…”

Genetic Algorithms for the Dependability Assurance in the Design of a Long‐Span Suspension Bridge

“…They are characterized by their robustness and flexibility and by their ability to generalize. In recent years, ANNs have been applied in the civil engineering field (an overview can be found in Adeli, 2001), structural design (Adeli and Park, 1995; Hajela, 1999), reliability engineering and structural identification (Ceravolo et al, 1995; De Stefano et al, 1999; Kim et al, 2000; Jiang and Adeli, 2005; Adeli and Jiang, 2006; Arangio and Bontempi, 2010; Arangio and Beck, 2011), transportation engineering (Vlahogianni et al, 2007; Boto‐Giralda et al, 2010; Graf et al, 2010), foundation engineering (Reuter and Moeller, 2010), and for pattern recognition (Adeli and Samant, 2000).…”