Christian Cremona

Optimal extrapolation of traffic load effects

Abstract This paper proposes a method for determining optimal fittings of the Rice formula to the tails of outcrossing rate histograms related to traffic load effects. An optimal fitting is obtained through a series of statistical tests between successive fittings and the histogram tails. Optimality is found when a confidence level β0 for the Kolmogorov test is verified. That provides the optimal number of class intervals to take into account the load-effect extrapolation under the hypothesis that the process is stationary and Gaussian. Extrapolations are given for minimal and maximal load effects for any return period. Some examples highlight the use of the method.

A bridge network maintenance framework for Pareto optimization of stakeholders/users costs

For managing highway bridges, stakeholders require efficient and practical decision making techniques. In a context of limited bridge management budget, it is crucial to determine the most effective breakdown of financial resources over the different structures of a bridge network. Bridge management systems (BMSs) have been developed for such a purpose. However, they generally rely on an individual approach. The influence of the position of bridges in the transportation network, the consequences of inadequate service for the network users, due to maintenance actions or bridge failure, are not taken into consideration. Therefore, maintenance strategies obtained with current BMSs do not necessarily lead to an optimal level of service (LOS) of the bridge network for the users of the transportation network. Besides, the assessment of the structural performance of highway bridges usually requires the access to the geometrical and mechanical properties of its components. Such information might not be available for all structures in a bridge network for which managers try to schedule and prioritize maintenance strategies. On the contrary, visual inspections are performed regularly and information is generally available for all structures of the bridge network. The objective of this paper is threefold (i) propose an advanced network-level bridge management system considering the position of each bridge in the transportation network, (ii) use information obtained at visual inspections to assess the performance of bridges, and (iii) compare optimal maintenance strategies, obtained with a genetic algorithm, when considering interests of users and bridge owner either separately as conflicting criteria, or simultaneously as a common interest for the whole community. In each case, safety and serviceability aspects are taken into account in the model when determining optimal strategies. The theoretical and numerical developments are applied on a French bridge network.

Assessment of European railway bridges for future traffic demands and longer lives – EC project “Sustainable Bridges”

A European Integrated Research Project has recently been started within the 6th Framework Program of the European Commission. The project aims at improved methods for the upgrading of existing railway bridges within the European railway network. The main objectives of the project are to increase the transport capacity by allowing higher axle loads and by increasing the maximum speeds. Other objectives are to increase the residual lifetime of existing bridges and to enhance management, strengthening and repair systems. The overall goal is to enable the delivery of improved capacity without compromising the safety and economy of the working railway. A consortium consisting of railway bridge owners, consultants, contractors, research institutes and universities will carry out the project, having a gross budget of more than 10 million Euros. Funding from the European Commission covers a major portion of the four-year project costs.

Assignment of structural behaviours in long-term monitoring: Application to a strengthened railway bridge

Novelty detection, the identification of data that is unusual or different, is relevant in a wide number of real-world scenarios, ranging from identifying unusual weather conditions to detecting evidence of damage in mechanical systems. Using novelty detection approaches for structural health monitoring presents significant challenges to the non-expert user. In this article, symbolic data analysis is introduced to model variability in tests. Hierarchy-divisive methods and dynamic clouds procedures are then used to discriminate structural changes used as novelty detection approaches for classifying structural behaviours. This article reports the study of experimental tests performed on a railway bridge in France. This bridge has undergone reinforcement works during the summer of 2003. Through the years of 2004–2006, new sets of dynamic tests were recorded. The main objective was to analyse the evolution of the bridge’s dynamic behaviour over time. To this end, the symbolic data analysis–based clustering methods are used for assigning new tests to clusters identified before and after strengthening or to highlight a totally different structural behaviour.

Optimal maintenance strategies for bridge networks using the supply and demand approach

This paper presents a probability-based approach for optimising the management of bridge networks. Most of the Bridge Management Systems are focused on condition features to ensure a minimum safety level for each individual bridge. Their location on the road network, the consequences of inadequate service due to maintenance actions are therefore not taken into consideration. These multiple criteria should be considered when scheduling maintenance activities. To overcome these limitations, a probabilistic supply and demand strategy is proposed for determining the optimal maintenance planning for each interconnected bridge. The problem is solved with genetic algorithms. One objective function is first introduced, corresponding to the summation of all the maintenance, failure, and user costs. Then, two conflicting objective functions are considered, the total user costs and the maintenance and rehabilitation costs. Safety and serviceability aspects are taken into account in the methodology and the theoretical and numerical developments are applied on a part of the French national network.

A simplified numerical approach of global behaviour of RC beams degraded by corrosion

One of major causes responsible for the performance loss of reinforced concrete structures is the corrosion phenomenon. Thereby, taking into account the local effects of the steel/concrete interface is of primary importance to predict properly the response of corroded reinforced concrete structures. A multifiber-based model including the steel/concrete interface is proposed. This interface model allows taking into consideration the bond strength variation due to corrosion. Such an approach leads to reasonable computational costs which a powerful feature of the model. A numerical study of beams already studied in the French project « benchmark des poutres de la Rance » is proposed in order to show the efficiency and reliability of the proposed model: Une des causes majeures pouvant conduire à une perte de performance est due au phénomène de corrosion. De ce fait, considérer le comportement de l’interface acier/béton est de première importance pour prédire la réponse de structures en béton armé corrodées. Un modèle simplifié incluant les effets locaux de l’interface acier/béton est proposé. Ce modèle permet de prendre en compte la variation d’adhérence due à la corrosion. Une telle approche conduit à des coûts de calcul raisonnables ce qui est un des points forts du modèle. Pour ce faire, une étude numérique de poutres issues du projet de recherche « Benchmark des poutres de la Rance » est proposée dans le but de montrer l’efficacité et fiabilité de la méthode proposée.

Baseline-free real-time assessment of structural changes

This article addresses the subject of data-driven structural health monitoring and proposes a real-time strategy to conduct structural assessment without the need to define a baseline period, in which the monitored structure is assumed healthy and unchanged. Independence from baseline references is achieved using unsupervised discrimination machine-learning methods, widely known as clustering algorithms, which are able to find groups in data relying only on their intrinsic features and without requiring prior knowledge as input. Real-time capability is based on the definition of symbolic data, which allows describing large amounts of information without loss of generality or structural-related information. The efficiency of the proposed methodology is illustrated using an experimental case study in which structural changes were imposed to a suspended bridge during an extensive rehabilitation programme. A single-value novelty index capable of describing multi-sensor data is proposed, and its effectiveness in identifying structural changes in real time, using outlier analysis, is discussed.

Application of the reliability theory to the assessment of carbonation-induced corrosion risk of rebars

Deterministic approaches for the prediction of the carbonation penetration have limitations related to the high variability of the input data. This paper deals with two carbonation models which are developed in the framework of a probabilistic approach. The random input data are presented, i.e. materials, environment and engineering design properties. The evaluation of the time-dependent probability of carbonation-induced reinforcement corrosion is assessed by calculating the reliability index for three studied concretes. This probabilistic approach provides a framework for a sensitivity analysis which enables one to identify the most influencing model parameters which have to be determined carefully in order to improve the accuracy of the prediction and to determine the parameters whose variability should be reduced to enhance the durability. Furthermore, a probabilistic approach provides a tool to design the thickness of the concrete cover in order to avoid oversizing. Les modélisations déterministes de la carbonatation présentent des limites, essentiellement liées à la forte variabilité des données d’entrée des modèles. Le présent article porte sur deux exemples de modèle de carbonatation qui sont assortis d’un cadre probabiliste. Les données d’entrée des modèles sont présentées et assorties de lois de probabilité. La sécurité, vis-à-vis du risque de carbonatation, est donneé avec un niveau de confiance. L’approche probabiliste offre aussi un cadre pour une analyse de sensibilité de ces deux modèles permettant de dégager les paramètres les plus influents qui devront être déterminés avec soin pour améliorer la justesse de la prédiction et ceux pour lesquels la variabilité devra être réduite afin d’améliorer la durée de vie. De plus, les modèles probabilistes constituent des outils d’aide au dimensionnement de l’enrobage, notamment lorsque les approches normatives conduisent à un surdimension-nement.

Influence of accelerated corrosion on the reinforced cover concrete cracking behavior: experimental and numerical study

This article aims to present both experimental and numerical studies of the corrosion induced cracking pattern evolution of a reinforced concrete sample subjected to accelerated corrosion. The beam was not mechanically loaded. Rebars were intentiostatically corroded using a current density of 100μA/cm² of steel, in a chloride pond and for a 30 day period. Electrochemical tests and visual inspections were carried out in order to characterise the rebar state. Width evolution of the main longitudinal cover crack was measured thanks to ball-extensometer. A numerical modeling of the corroded RC beam has also been realised. The experimental and numerical results are in good agreement confirming the efficiency of the proposed modelling Cet article présente les principaux résultats d’une étude combinant approches expérimentale et numérique appliquées à un essai de corrosion accélérée sur un corps d’épreuve en béton armé. Ces résultats portent sur le suivi et la prédiction de l’évolution de la fissuration du béton sous l’effet de l’expansion des produits de corrosion. L’éprouvette n’a pas été chargée mécaniquement. Les armatures ont été corrodées en imposant un courant de densité de 100μA/cm² d’acier dans une solution saline pendant 30 jours. Des essais de caractérisation électrochimique ont été réalisés puis complétés par des inspections visuelles en vue de décrire l’état de dégradation de l’interface acier/béton. Les résultats de la simulation numérique conduisent à une évaluation satisfaisante de l’évolution de la fissure principale démontrant ainsi la pertinence de la modélisation proposée

Improved Assessment Methods for Static and Fatigue Resistance of Old Metallic Railway Bridges

A large number of railway structures in Europe are metallic bridges. The increasing volume of traffic and axle weight of trains mean that, for many structures, the loads today are much higher than those envisaged when they were designed. This paper presents a summary of the different recommendations and advice proposed in European guidelines for assessing load and resistance of railway bridges issued from a research project. The knowledge of the material properties of existing metallic bridges is essential for the resistance assessment and the determination of the remaining lifetime of old metallic bridges. Furthermore, old bridges require more exact and efficient assessment methods that call for a precise description of the material. Among the problems met in metallic structures and material properties estimation, fatigue is the most common cause of failure. To be able to make accurate assessments of existing bridges, it is important to know the behavior of bridges exposed to fatigue and how the old materials behave owing to cyclic exposure. The main question answered in this paper is how to make a safe estimation concerning the remaining life in service. The possible traffic load on steel rail bridges is usually limited by the fatigue resistance, but for certain situations the static resistance also has to be checked. Most design rules for steel structures are applicable also to riveted structures. However, some information is missing on how to deal with the special case when elements are intermittently connected in contrast to welded structures that are connected continuously. Because the traditional methods for assessing the resistance of steel bridges are based on elastic analysis, a method for utilizing a limited redistribution of bending moments based on beam theory is proposed.