Simulation test of continuous rigid frame railway bridge under flood scouring

Structures, Год журнала: 2024, Номер 66, С. 106893 - 106893

Опубликована: Июль 17, 2024

Язык: Английский

---

A novel extreme gradient boosting algorithm based model for predicting the scour risk around bridge piers: application to French railway bridges

European Journal of Environmental and Civil engineering, Год журнала: 2022, Номер 27(3), С. 1104 - 1122

Опубликована: Май 11, 2022

Bridge failures inevitably cause severe social consequences in terms of the economic losses or even human fatalities. Scour is one leading causes for bridge failures. Due to its complexity and multidisciplinary nature, mechanism scour has not been completely understood yet. Inspired by rapid development Machine Learning (ML) techniques, this paper aims construct a novel data-driven extreme gradient boosting (XGBoost) algorithm based model predict risk around piers. Data used present study provided French National Railway Company (SNCF). The performance XGBoost-based compared with three commonly algorithms: support vector machine, random forest multilayer perceptron. Results show that XGBoost classifier achieves high accuracy (0.959/0.938), precision (0.970/0.961), recall (0.974/0.956) low false positive rate (0.085/0.114) training test set respectively. Moreover, obtains an area under ROC curve (AUC) score equal 0.974 (a perfect AUC 1). This presents cutting-edge application maintenance railway bridges. proposed methodology may allow engineers determine piers accurate way.

Язык: Английский

---

Integration, harmonization, and processing of geomatic data for bridge health assessment: the Lastra a Signa case study

Applied Geomatics, Год журнала: 2023, Номер 15(3), С. 533 - 550

Опубликована: Май 3, 2023

Abstract A visual inspection, which entails field surveying, such as photodocumentation and footage, is the first step of a multi-level approach to bridge health assessment. Furthermore, use surface models, CAD drawings, orthophotos, guarantees complete accurate documentation, thus allowing for better understanding environment, anthropic structures, their relationships. All georeferenced surveys’ outputs are especially advisable within prespective periodical monitoring, forseen by national legislation. The work aimed study two adjacent bridges over Arno River in Lastra Signa, Italy. One bridges, Ponte nuovo sull’Arno, an overpass motor vehicles. other one, called Passarella pedestrian viaduct. topographic reference network has been settled using GNSS survey technique. 3D point clouds have acquired performing Laser Scanning survey. bathymetric carried out acquire cloud submerged bridges’ parts riverbed. Through Photogrammetric from RPAS, orthophoto area built. Finally, evidence historical structures identified thanks multi beam work’s objective integrate surveying geomatics techniques create reliable surrounding area, riverbed, support most common structural assessment methods.

Язык: Английский

---

An interpretable model for bridge scour risk assessment using explainable artificial intelligence and engineers’ expertise

Structure and Infrastructure Engineering, Год журнала: 2023, Номер unknown, С. 1 - 13

Опубликована: Июль 4, 2023

A machine learning (ML) model would be difficult to apply in practice without knowing how the prediction is made. To address this issue, paper uses XAI and engineers’ expertise interpret an ML for bridge scour risk prediction. By using data from French National Railway Company (SNCF), extreme gradient boosting (XGBoost) algorithm-based was constructed at first. Later, approaches were employed have global local explanations as well explicit expressions interpretation. Meanwhile, a group of engineers SNCF asked rank input parameters based on their engineering judgment. In end, feature importance obtained survey compared. It found that both interpretations, observation around foundation most important decision-making. The differences between interpretations human emphasize knowledge hydrology hydromorphology assessment, since currently make decisions primarily observed damages (e.g., hole, crack). results could trustworthy by understanding made provide valuable guidance improving current inspection procedures.

Язык: Английский

---

Prioritizing Riverine Bridges Interventions: A Hydrological Multidimensional Approach

Опубликована: Авг. 29, 2023

Globally, bridges fail mostly due to hydrological causes such as scour or flooding. Therefore, under a approach, this study proposes methodology that contributes prioritize the intervention of prevent their collapse. Through an exhaustive literature review, evaluation matrix subdivided into 4 dimensions was developed and total 18 parameters were considered, distributed follows: environmental, 6 technical, social economic. This applied eight with history problems in same river validated through semi-structured interviews specialists. Data collected field visits, journalistic information, review gauged basin historical flow rates consultations population. Then, modeling, which considered influence gullies discharge additional flow, carried out using HEC-HMS HEC-RAS subsequently calibrated. The application revealed five have high vulnerability, three medium vulnerability. multidimensional approach can be adapted for studies other riverine bridges.

Язык: Английский

---

The Socio-Economic Effects of Floods and Ways to Prevent Them: A Case Study of the Khazir River Basin, Northern Iraq

Water, Год журнала: 2023, Номер 15(24), С. 4271 - 4271

Опубликована: Дек. 13, 2023

The current study deals with how floods affect the social and economic lives of villagers in lower basin Khazir River (northern Iraq), where villages Kazkan, Tal al-Laban, Wardak are located within active floodplains close to riverbed. aim this paper is perform both spatial temporal analysis affecting society activities river assess ways preventing them. In study, qualitative quantitative methods were used analyze results, we relied on hydrological data field studies that included oral interviews personal observations phenomenon floods. first main findings indicate through a statistical sample 100 people area an expected recurring phenomenon; they occur every year, damage losses varying from year one region another. second frequent flooding has negative impact growth all sectors. These agricultural sector more than other sectors as result land being flat plain areas little slope. Based effects floods, improved engineering solutions have been suggested better control handle emergencies. This done using GIS models (HEC-GeoRAS) HEC-RAS build number hydraulic designs riverbed according different scenarios. should serve inspiration for water policy makers make effort implement feasible effective preventive measures before well allow rapid reaction, recovery, reconstruction after flood.

Язык: Английский

---

A study of stone arch bridge’s flood reliability through a surrogate model approach

Structure and Infrastructure Engineering, Год журнала: 2024, Номер unknown, С. 1 - 16

Опубликована: Фев. 26, 2024

Europe's historic masonry arch bridges are culturally and economically significant, but their long-term safety must be ensured. Scour effects the most common cause of collapse, so it is necessary to carry out structural assessments mitigate risk prevent potential failures. In this study, a metamodel-based method was used determine probability failure an existing stone bridge in Portugal due local contraction scour on abutments. Non-linear finite element analysis supported calculation reliability index, which took into account soil-structure interaction mechanism. The variables with greatest influence load-carrying capacity structure were identified surrogate model implemented. Fragility curves then derived based model, using depth as measure intensity load factor engineering requirement parameter. results study indicate that numerical compromised when 1.5 m reaches base foundation. As result, stability problems settlements observed model. At 2.5 m, soil its ultimate bearing capacity.

Язык: Английский

---

Deciphering seasonal bridge scour effects: a 1D sediment transport model study in India's tropical basin

Modeling Earth Systems and Environment, Год журнала: 2024, Номер 11(1)

Опубликована: Дек. 26, 2024

Язык: Английский

---

Image-Based Techniques for the Advanced Characterization of Scour around Bridge Piers in Laboratory

Journal of Hydraulic Engineering, Год журнала: 2022, Номер 148(6)

Опубликована: Март 16, 2022

The study of local scour in bridge foundations is often restricted to quantifying maximum depth on the upstream side. This limiting understanding resulting by numerical methods, for which shape hole a mandatory input. In this work, two nonintrusive three-dimensional (3D) image-based measuring techniques, consisting close-range photogrammetry and Kinect V2 sensor, were employed characterize initial final development holes around piers, under steady flow laboratory flume. A novel two-dimensional (2D) methodology was also applied continuous measurement side using submersible cameras. Models geometries vicinity oblong piers obtained at high accuracy levels. database calibration validation models thus provided.

Язык: Английский

---

Analysis of a three-dimensional numerical modeling approach for predicting scour processes in longitudinal walls of granular bedding rivers

EUREKA Physics and Engineering, Год журнала: 2023, Номер 4, С. 168 - 179

Опубликована: Июль 27, 2023

In rivers, longitudinal walls are protective structures that directly supported on the riverbank frequently used as hydraulic prevent current from eroding bank and causing scouring. However, these have potential to block flow produce erosive processes progressively worsen scour in their area, leading faulting other problems. The study Flow-3D software understand process at base of rivers with a well-graded granular bed. Experimental data physical model replicating river wall bed were validate model. investigation examined average velocity its effects behavior along using program. findings showed could improve evaluation debugging processes, because it provided useful answer closely matched experimental derived Validation 0.07 m mesh demonstrated faithfully simulate wall. Overall, this suggest can be tool for predicting scouring beds walls. This is particularly important engineers researchers who interested designing optimizing mitigate scouring,

Язык: Английский

---