Abstract:Due to ageing of the bridge stock, continuously increasing load requirements, the need for assessing existing bridges is becoming increasingly important. Reinforced concrete (RC) bridge deck slabs are among the most critical parts to the load carrying capacity of bridges. Previous research indicated that the assessment method used in current practice largely underestimate the load-carrying capacity. The objective of the study reported in this licentiate thesis is to develop and calibrate improved methods for assessment of load-carrying capacity and response of existing bridge deck slabs. This study proposes an enhanced assessment through improved structural analysis and resistance evaluation in order to achieve higher detectable load-carrying capacity. To achieve this objective, the scientific approaches including literature study, laboratory tests, analytical analyses and finite element analyses were adopted. Two major studies and a series with supporting tests were conducted as following: A large test series, in which three specimens containing traditional steel bar reinforcement in ordinary concrete, were carried out to study the load carrying capacity and structural behavior of two-way slab. A proposed multi-level assessment method was validated through two case studies: a two-way slab and a cantilever slab. The multi-level assessment method includes simplified analysis, 3D linear FE analysis, 3D nonlinear analysis with shell elements, 3D nonlinear FE analysis with continuum elements include fully bonded reinforcement as well as 3D nonlinear FE analysis with continuum elements including bond-slip for the reinforcement. To develop the modeling strategies for two-way RC slabs with continuum elements, parameter studies has been carried out to investigate the influence of different modeling choices. Parameters including finite element properties, modeling of concrete, modeling of reinforcement and stiffness of supports were varied in different comparative models to investigate the influence. The study shows that existing methods currently used in assessment are not able to accurately reflect the response of RC bridge deck slabs, and underestimate their capacity. With enhanced assessment methods, such as the nonlinear finite element method, it is possible to better reflect the structural behavior, and more accurately determine the load carrying capacity. Two case studies have proven that the multi-level assessment method is able to reflect the structural behavior at different levels of approximation, which is beneficial in evaluating existing RC slabs. The parameter studies indicate that modeling choices such as geometric nonlinearity, crack bandwidth and Poisson’s ratio have significant impact on determination of the load-carrying capacity. Finite element properties, reinforcement model and stiffness of supports influence crack pattern and load distribution respectively.

Artificial Neural Network and Kriging Surrogate Model for Embodied Energy Optimization of Prestressed Slab Bridges

The Application of a Damage Detection Method Using Artificial Neural Network and Train-Induced Vibrations on a Simplified Railway Bridge Model

Neighborhood Component Analysis-Based Feature Selection in Machine Learning to Predict Tendon Ultimate Stress of Unbonded Prestressed Concrete Beams

Structural Analysis of Existing RC Bridge Deck Slabs Structural Analysis of Existing RC Bridge Deck Slabs

Structural Assessment of Bridge Deck Slabs

An ML-based framework for predicting prestressing force reduction in reinforced concrete box-girder bridges with unbonded tendons

Numerical Study of Active Mass Damper Application on Cable-Stayed Bridge Structure Using Artificial Neural Network Algorithm

Examination of load-deformation characteristics of long-span bridges in harsh natural environments based on real-time updating artificial neural network

Machine-Learning-Based Consumption Estimation of Prestressed Steel for Prestressed Concrete Bridge Construction

Artificial neural network surrogate modeling for uncertainty quantification and structural optimization of reinforced concrete structures

Reliability based design optimization of bridges considering bridge-vehicle interaction by Kriging surrogate model

Optimization of under-deck cable-stayed concrete bridges

A study on the application of an improved adaptive neural network in prestressed bridge engineering inspection

Applications of Machine Learning to Predict the Flexural Bearing Capacity of Hollow Core Slabs After Fire Exposure

Research on Intelligent Inspection Method of Prestressed Bridge Engineering Based on Machine Learning

Active-learning Kriging-assisted robust design optimization of tuned mass dampers: Vibration mitigation of a steel-arch footbridge

Nonlinear global design resistance: Case studies of post‐tensioned concrete bridges made of I‐73 and KA‐61 girders

High-Efficiency Finite Element Model Updating of Bridge Structure Using a Novel Physics-Guided Neural Network

Machine Learning Prediction and Evaluation for Structural Damage Comfort of Suspension Footbridge

Determination of Bridge Prestress Loss under Fatigue Load Based on PSO-BP Neural Network

Prediction of steel plate-based damper for improving the behavior of concentrically braced frames based on RSM and ML approaches for sustainable structures