EM - Escola de Minas
URI permanente desta comunidadehttp://www.hml.repositorio.ufop.br/handle/123456789/6
Notícias
A Escola de Minas de Ouro Preto foi fundada pelo cientista Claude Henri Gorceix e inaugurada em 12 de outubro de 1876.
Navegar
6 resultados
Resultados da Pesquisa
Item Model-based damage identification of railway bridges using genetic algorithms.(2020) Alves, Vinicius Nicchio; Oliveira, Matheus Miranda de; Ribeiro, Diogo; Calçada, Rui; Cury, Alexandre AbrahãoThe assessment of structural integrity via numerical model updating has been drawing attention in several areas of engineering over the last years. Basically, it consists in an optimization process based on the minimization of the residuals between measured and estimated numerical re- sponses. In such methodologies, several factors influence the success of both localization and quantification of structural damage, such as: the damage features used in the formulation of the objective function, the optimization algorithm and the adopted updating parameters. Many ex- isting studies using these methods are applied to simple structural systems, e.g., beams, frames and trusses. However, few studies applied to large and complex structures are found in the lit- erature. In this context, this work proposes to assess the performance of a genetic algorithm- based approach applied to two case studies. The first case refers to a two-dimensional model of a hypothetical railway bridge, where the efficiency and robustness of five different indicators are assessed considering three damage scenarios. In the second case, a real railway bridge is con- sidered. The results obtained show that the proposed approach is able to detect, locate and quantify multiple damage with several updating parameters and few target responses.Item Comparison of machine learning techniques to predict the compressive strength of concrete and considerations on model generalization.(2022) Paixão, Rafael Christian Fonseca da; Penido, Rúben El-Katib; Cury, Alexandre Abrahão; Mendes, Júlia CastroThe compressive strength of concrete is an essential property to ensure the safety of a concrete structure. However, estimating this value is usually a laborious and uncertain process since the mix design is based on empirical methods and its confirmation in the laboratory demands time and resources. In this context, this work aims to evaluate Machine Learning (ML) models to predict the compressive strength of concrete from its constituents. For this purpose, a dataset from the literature was used as input to four ML models: Extreme Gradient Boosting (XGBoost), Support Vector Regression (SVR), Artificial Neural Networks (ANN) and Gaussian Process Regression (GPR). The accuracy of the models was evaluated through 10-fold cross-validation, and quantified by R2 , Mean Absolute Error (MAE), and Root-Mean-Square Error (RMSE) metrics. Subsequently, a new dataset was put together with mixtures from the literature and used to validate the previous models. In the model creation step, all algorithms obtained similar and positive results, with MAE between 1.96-2.26 MPa and R2 varying from 0.79 to 0.83. However, in the validation step, the accuracy of the models dropped sharply, with MAE growing to 3.04-4.04 MPa and R2 decreasing to 0.37-0.59. ANN and GPR showed the best results, while SVR had the worst predictions. This work showed that ML tools are promising techniques to predict the compressive strength of concrete. However, care must be taken with the input data to guarantee that models are not overfitted to a given region, set of materials, or type of concrete.Item Symbolic data analysis and supervised/ non supervised learning algorithms for bridge health monitoring.(2011) Crémona, Christian; Cury, Alexandre Abrahão; Orcesi, André; Dieleman, LucIn the past few years, numerous methods for damage assessment in connection with structural health monitoring were proposed in the literature. Several problems are raised for making these approaches practical for the engineer. The first concern is to determine whether a structure presents an abnormal behavior or not. Statistical inference is concerned with the implementation of algorithms that analyze the distribution of extracted features in an effort to make decisions on damage diagnosis. Learning algorithms have extensively been applied to classification and pattern recognition problems in the past years and deserve to be used for structural health monitoring. Two approaches are nevertheless available depending on the ability to perform supervised or unsupervised learning. The first group of methods forms the family of classification methods whereas the second group is referred to clustering techniques. In addition, data acquisition campaigns of civil engineering structures can last from several minutes to years. Dealing with large amounts of data is not an easy task and suitable tools are required to correctly extract important features from them. To deal with this issue, symbolic data analysis (SDA) is introduced for managing complex, aggregated, relational, and higher-level data. SDA is then coupled with supervised and non supervised learning algorithms to form a new family of hybrid techniques. From the non supervised learning side, dynamic clouds and hierarchy-divisive method have been used. From the supervised learning side, neural networks and support vector machines have been introduced. All these techniques have been developed within the concept of symbolic data analysis in order to compress data without losing its inherent variability. To highlight the different features of these techniques for structural health monitoring, this paper focuses attention on the monitoring of a railway bridge belonging to the high speed track between Paris and Lyon. During the month of June 2003, a strengthening procedure was carried out in this bridge. In so doing, vibration measurements were recorded under three different structural conditions: before, during and strengthening. In the following years (2004, 2005 and 2006), new tests were performed to observe how the dynamic behavior of the bridge evolved, especially for the case of frequency changes. The objective was to verify whether the strengthening procedure was still effective or not, in order terms if the new data could be still assigned to the condition “after strengthening”. This paper reports the major results obtained and shows how the techniques can be applied to cluster structural behaviors and classify new data.Item Application of symbolic data analysis for structural modification assessment.(2010) Cury, Alexandre Abrahão; Crémona, Christian; Diday, EdwinStructural health monitoring is a problem which can be addressed at many levels. One of the more promising approaches used in damage assessment problems is based on pattern recognition. The idea is to extract features from the data that characterize only the normal condition and to use them as a template or reference. During structural monitoring, data are measured and the appropriate features are extracted as well as compared (in some sense) to the reference. Any significant deviations from the reference are considered as signal novelty or damage. In this paper, the corpus of symbolic data analysis (SDA) is applied on the one hand for classifying different structural behaviors and on the other hand for comparing any structural behavior to the previous classification when new data become available. For this purpose, raw information (acceleration measurements) and also processed information (modal data) are used for feature extraction. Some SDA techniques are applied for data classification: hierarchy divisive methods, dynamic clustering and hierarchy agglomeratives chemes. Resultsregarding experimental performed onarail way bridge in France are presented in order to show the efficiency of the described methodology. The results show that the SDA methods are efficient to classify and to discriminate structural modifications either considering the vibration data orthe modal parameters. In general, both hierarchy divisive and dynamic cloud methods produce better results compared to those obtained by using the hierarchy agglomerative method. More robust results are given by modal data than By measurement dataItem Long-term monitoring of a PSC box girder bridge : operational modal analysis, data normalization and structural modification assessment.(2012) Cury, Alexandre Abrahão; Crémona, Christian; Dumoulin, JohnFor reliable performance of vibration-based damage detection algorithms, it is impor-tant to distinguish abnormal changes in modal parameters caused by structural damage from normal changes due to environmental fluctuations. This paper firstly addresses the modeling of temperature effects on modal frequencies of a PSC box girder bridge located on the A1 motorway in France. Based on a six-month monitoring experimental program, modal frequencies of the first seven mode shapes and temperatures have been measured at three hour intervals. Neural networks are then introduced to formulate regression models for quantifying the effect of temperature on modal parameters (frequencies and mode shapes). In 2009, this bridge underwent a strengthening procedure. In order to assess the effect of strengthening on the vibration characteristics of the bridge, modal properties had to be corrected from temperature influence. Thus, the first goal is to assess the changes on the vibration signature of this bridge induced by the strengthening. For this purpose, classical statistical analysis and clustering methods are applied to the data recorded over the period after strengthening. The second goal is to evaluate the influence of temperature effects on the clustering results. It comes that the temperature correction significantly improves the confidence in the novelty detection and in the strengthening efficiency.Item Assignment of structural behaviours in long-term monitoring : application to a strengthened railway bridge.(2012) Cury, Alexandre Abrahão; Crémona, ChristianNovelty 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