RESUMO
We live in the era of large data analysis, where processing vast datasets has become essential for uncovering valuable insights across various domains of our lives. Machine learning (ML) algorithms offer powerful tools for processing and analyzing this abundance of information. However, the considerable time and computational resources needed for training ML models pose significant challenges, especially within cascade schemes, due to the iterative nature of training algorithms, the complexity of feature extraction and transformation processes, and the large sizes of the datasets involved. This paper proposes a modification to the existing ML-based cascade scheme for analyzing large biomedical datasets by incorporating principal component analysis (PCA) at each level of the cascade. We selected the number of principal components to replace the initial inputs so that it ensured 95% variance retention. Furthermore, we enhanced the training and application algorithms and demonstrated the effectiveness of the modified cascade scheme through comparative analysis, which showcased a significant reduction in training time while improving the generalization properties of the method and the accuracy of the large data analysis. The improved enhanced generalization properties of the scheme stemmed from the reduction in nonsignificant independent attributes in the dataset, which further enhanced its performance in intelligent large data analysis.
RESUMO
This dataset is related to the research paper entitled "Bridge-specific flood risk assessment of transport networks using GIS and remotely sensed data" published in the Science of the Total Environment. It provides the information necessary for the reproduction of the case study that was used for the demonstration and validation of the proposed risk assessment framework. The latter integrates indicators for the assessment of hydraulic hazards and bridge vulnerability with a simple and operationally flexible protocol for the interpretation of bridge damage consequences on the serviceability of the transport network and on the affected socio-economic environment. The dataset encompasses (i) inventory data for the 117 bridges of the Karditsa Prefecture, in Central Greece, which were affected by a historic flood that followed the Mediterranean Hurricane (Medicane) Ianos, in September 2020; (ii) results of the risk assessment analysis, including the geospatial distribution of hazard, vulnerability, bridge damage, and associated consequences for the area's transport network; (iii) an extensive damage inspection record, compiled shortly after the Medicane, involving a sample of 16 (out of the 117) bridges of varying characteristics and damage levels, ranging from minimal damage to complete failure, which was used as a reference for validation of the proposed framework. The dataset is complemented by photos of the inspected bridges which facilitate the understanding of the observed bridge damage patterns. This information is intended to provide insights into the response of riverine bridges to severe floods and a thorough base for comparison and validation of flood hazard and risk mapping tools, potentially useful for engineers, asset managers, network operators and stakeholders involved in decision-making for climate adaptation of the road sector.
RESUMO
Flash floods are common manifestations of extreme weather events and one of the most severe natural hazards. In Europe, they have been responsible for 359 fatalities and an economic loss totalling 67 million USD in the past decade (EM-DAT), while their increasing severity is linked to climate change. Nevertheless, flash floods remain a poorly documented natural phenomenon due to the lack of flow intensity data in many of the affected watersheds. Based on a thorough field investigation, including UAV-based 3D mapping and material characterization with on-site testing, we carry out a numerical study of a notable flood that caused the collapse of bridges and buildings in Central Greece, following a recent Mediterranean hurricane. Focusing on a carefully selected case study, we combine 3D modelling of flow-structure interaction with detailed mechanical modelling of the nonlinear structural response to reproduce the flood-induced fracture of a bridge abutment. Back-analysis of this failure responds to the fundamental problem of estimating the undocumented magnitude of this extreme event. The paper estimates a lower bound value of the flow velocity at the studied location. This can be valuable input for the interpretation of the extensive damage that took place downstream and for the re-assessment of flood risk in a region where similar events are expected to become more frequent because of climate change. The approach, where disaster forensics and engineering analysis are used to fill the gap of missing real-time measurements, can be implemented for the a posteriori estimation of flood intensity in similar events. The well-documented case study of a bridge failure due to extreme flooding can also be used for validation of future numerical and experimental methods and motivate investigations of the mechanisms governing flow-soil-structure interaction in river crossings.
Assuntos
Desastres , Colapso Estrutural , Mudança Climática , Inundações , RiosRESUMO
The purpose of this survey is to define the restoration tasks after hydraulic-induced damage and/or loss of functionality of bridges. This includes the duration and sequence of restoration tasks, idle times, cost, and traffic/functionality loss for specified damage levels of given bridge components. The potential use of this data is the generation of sets of restoration and reinstatements functions for quantifying the resilience of bridges exposed to hydraulic hazards, i.e. scour, debris accumulation and hydraulic forces (Mitoulis et al. 2021). The data are expected to inform boroughs, county councils, road and rail owners and stakeholders by providing valuable information for managing efficiently their assets prior to and after catastrophic events on the basis of resilience. The survey was based on a questionnaire answered by experts on bridge and infrastructure engineering.