Development of a bridge management system incorporating a newly developed model for element condition evaluation based on damage effects

Abstract

This article presents a comprehensive research project undertaken for the development of a bridge management system (BMS) for the General Directorate of Highways in Turkey, establishment of a prototype inspection program and the development of a new model, implemented within the BMS, for element condition state evaluation based on element damage types. As an introduction, a summary of the BMS development process is presented. Then, a detailed explanation of the newly developed model for element condition evaluation based on the effects of different damage types is given. The model presented herein extends the percentage-based condition state distribution for a bridge element to percentage and damage-based condition state distribution. This is achieved by considering all damage types a given bridge element may be subjected to. The model is already implemented into the newly developed BMS. It has been found that the newly implemented model will enable the inspector to more accurately define the condition state distribution of a bridge element taking into account both quantitative and qualitative characteristics of damage. Using the developed methodology and damage weight factors, it is possible to calculate a weighted condition state percentage distribution for a bridge element by taking into consideration the importance and urgency of different damage types.

A method for time‐varying reliability calculation of RC bridges considering random deterioration process

“…Then, a neural network 𝐵 1 is established according to Equation (15) to implement the indefinite integral of u A1 with respect to x 1 . The upper and lower limits of variable x 1 are then substituted into u B1 according to Equation (20), giving an approximation of the first definite integral of f 1 with respect to x 1 . These steps are repeated for each variable, with a new integrand function f i generated using Equation (20), and neural networks 𝐴 𝑖 and 𝐵 𝑖 established to approximate the function and implement the indefinite integral, respectively.…”

“…The upper and lower limits of variable x 1 are then substituted into u B1 according to Equation (20), giving an approximation of the first definite integral of f 1 with respect to x 1 . These steps are repeated for each variable, with a new integrand function f i generated using Equation (20), and neural networks 𝐴 𝑖 and 𝐵 𝑖 established to approximate the function and implement the indefinite integral, respectively. Finally, a neural network is established to learn the integrand function f n and approximate it as u An .…”

“…The upper and lower limits of 𝑡 were then substituted into 𝑢 𝐵1 according to Equation (18) to obtain an approximation of the definite integral of 𝑓 1 with respect to 𝑡. The third neural network, 𝐴 2 , was constructed with the same active functions as 𝐴 1 and trained to learn the integrand 𝑓 2 using Equation (20). The function 𝑢 𝐵1 was used to generate 𝑓 2 , which was then used to train 𝐴 2 .…”

“…9 Akgul proposed a new model for structural reliability calculation and a degradation model for steel-reinforced concrete bridge structures. 20 Czarnecki and Nowak found that the resistance of steel-reinforced concrete bridge structures changes significantly over time in a corrosive environment and proposed a time-varying reliability assessment model for steelreinforced concrete structures, which was used to predict the ultimate service life of the structure. 21 In summary, reliability assessment and ultimate life prediction of bridge structures based on reliability theory is a complex issue.…”

A Novel Runtime Algorithm for the Real-Time Analysis and Detection of Unexpected Changes in a Real-Size SHM Network with Quasi-Distributed FBG Sensors

“…To cope with the challenges of managing existing bridge structures, the concept of bridge management systems (BMS) emerged to support engineers and bridge managers to provide cost-effective decisions for the planning of maintenance, rehabilitation, and replacement (MRR) [ 14 ]. A BMS is defined as the rational and systematic approach to organizing and carrying out all activities related to managing individual bridges within an infrastructure network [ 15 ]. It became well established in 1993 when the US government issued legislation outlining the obligatory requirements for a BMS model, as summarized by Tran (2018) [ 14 ]: Database (inventory, inspection data, maintenance data).…”