The world is urbanizing rapidly with more than half of the global population now living in cities. Improving urban environments for the well-being of the increasing number of urban citizens is becoming one of the most important challenges of the 21st century. Even though it is common that city planners have visions of a ’good urban milieu’, those visions are concerning visual aesthetics or practical matters. The qualitative perspective of sound, such as sonic diversity and acoustic ecology are neglected aspects in architectural design. Urban planners and politicians are therefore largely unaware of the importance of sounds for the intrinsic quality of a place. Whenever environmental acoustics is on the agenda, the topic is noise abatement or noise legislation – a quantitative attenuation of sounds. Some architects may involve acoustical aspects in their work but sound design or acoustic design has yet to develop to a distinct discipline and be incorporated in urban planning.My aim was to investigate to what extent the urban soundscape is likely to improve if modern architectural techniques merge with principles of acoustics. This is an important, yet unexplored, research area. My study explores and analyses the acoustical aspects in urban development and includes interviews with practitioners in the field of urban acoustics, situated in New York City. My conclusion is that to achieve a better understanding of the human living conditions in mega-cities, there is a need to include sonic components into the holistic sense of urban development.

Afin de réduire le bruit dans les villes, il est nécessaire d’avoir une bonne compréhension de la propagation acoustique en milieu urbain. Il existe aujourd’hui des logiciels commerciaux qui permettent de modéliser des champs acoustiques urbains à des coûts de calcul raisonnables. Toutefois, ces outils sont basés principalement sur des approches énergétiques qui ne contiennent pas d’informations sur la phase. Pour cette raison, elles ne permettent pas la prise en compte d’effets d’interférence (par exemple, des résonances), nous offrant ainsi une description physique limitée du champacoustique. Inversement, des méthodes ondulatoires classiques (FEM, BEM, FDTD) permettent de prendre en compte ces effets. Or, en raison de la discrétisation et de la grande extension du domaine de propagation, leur utilisation est généralement limitée aux très basses fréquences.L’objectif principal de cette thèse est de développer des méthodes ondulatoires performants, dans le domaine fréquentiel et temporel, nous permettant de modéliser la propagation acoustique dans des zones urbaines étendues. L’approche proposée est basée sur une formulation mixte modale–éléments finis. L’idée clé de cette méthode estde considérer la rue comme un guide d’ondes ouvert, dont la base modale est composée de modes de fuite (modes qui rayonnent une partie de leur énergie en se propageant). Cette approche combine une description multimodale du champ acoustique dans la direction longitudinale et un calcul par éléments finis des modes propres transverses.L’approche a été mise en oeuvre précédemment à l’échelle d’une seule rue. Dans cette thèse, nous nous intéressons à l’extension de la méthode à l’échelle du quartier, afin de modéliser la propagation dans des milieux contenant un grand nombre de rues interconnectées. Une version simplifiée dans le domaine temporel, contenant uniquement lemode de propagatif le moins fuyant, est également développée.En nous basant sur ces approches, nous étudions des phénomènes ondulatoires qui peuvent apparaître dans des configurations urbaines particulières. Plus précisément, nous nous intéressons à l’interaction des modes de la rue avec des résonances dans une cour intérieure adjacente, ainsi qu’à la formation de bandes de fréquences interditesdans des réseaux périodiques de rues interconnectées. Le résultat principal de cette étude est que, malgré la forte présence de pertes par radiation dans le milieu, des effets de résonance importants peuvent encore se produire. Les résultats présentés dans ce manuscrit mettent en évidence l’importance d’une approche ondulatoire pour décrirecorrectement des champs acoustiques aux basses fréquences, et ils suggèrent l’usage potentiel de ces phénomènes afin de contrôler la propagation acoustique dans le milieu.Enfin, nous présentons une étude sur l’utilisation de métasurfaces (surfaces contenant un réseau de résonateurs) pour améliorer la performance des murs antibruit. Nous démontrons que, grâce à l’excitation des résonances locales sur la métasurface, il est possible d’obtenir des propriétés non conventionnelles, comme par exemple des angles de réflexion négatifs ou de l’absorption acoustique aux basses fréquences. / The improvement of the urban sound environment requires a good understanding of the acoustic propagation in urban areas. Available commercial softwares give the possibility to simulate urban acoustic fields at relatively low computational costs. However, these tools are mainly based on energy methods that do not contain information on the phase. Therefore, these tools are unable to capture interference effects (e.g., resonances), providing a limited physical description of the acoustic field. Conversely, classical wave methods such as FEM, BEM or FDTD give the possibility to model interference effects, but their use is often restricted to very low frequencies due to discretisation and the huge extension of the propagation domain.The main goal of this thesis is to develop efficient wave methods for the acoustic propagation modelling in extended urban areas, both in the frequency and time domain. The proposed approach is based on a coupled modal–finite elements formulation. The key idea is to consider the urban canyon as an open waveguide with a modal basis composedof leaky modes, i.e., modes that radiate part of their energy into the atmosphere as they propagate. The approach combines a multimodal description of the acoustic field in the longitudinal direction and a finite elements computation of the transverseeigenmodes. This coupled approach, which has been successfully implemented at the scale of a single street, is extended in the present manuscript at a larger scale (the neighbourhood scale), in order to model problems arising in propagation domains containing many interconnected streets. A time domain version of the method, containing only the least damped mode, is also proposed.Using these methods, we investigate wave phenomena arising in specific urban configurations, as forbidden frequency bands in periodic networks of interconnected streets, and resonances in inner yards. It is found that, despite the presence of significant radiative losses in the propagation medium, strong interference effects are still observed. Not only this result highlights the relevance of a wave approach to describe accurately urban acoustic fields at low frequencies, but it suggest the potential use of these phenomena to control the acoustic propagation in urban environments.The last part of this dissertation presents a preliminary study on the use of metasurfaces (surfaces decorated with an array of resonators) to improve the performance of noise barriers. It is shown that, exciting resonances in these structures, it is possible to achieve some unconventional behaviours, including negative angles of reflection and low frequency sound absorption.

Em espaços urbanos abertos, particularmente nas grandes cidades de climas tropicais, os pedestres estão expostos não somente a níveis sonoros elevados como também a elevadas cargas térmicas, situação que pode gerar tanto o desconforto acústico quanto o térmico. Entretanto, na maioria das vezes, a relação entre a exposição a condições acústicas e térmicas adversas e a percepção humana dessas condições são estudadas separadamente. Neste trabalho propõe-se, portanto, uma abordagem integrada para a avaliação do conforto acústico e térmico urbano e para o estudo de seus prováveis efeitos combinados. Esta pesquisa foi realizada em Belo Horizonte - MG, Brasil, cidade localizada em região de clima tropical de altitude, com verões quentes e úmidos e invernos frios e secos. Utilizou-se o método indutivo experimental na condução dos trabalhos. Dados acústicos e climáticos foram medidos simultaneamente à aplicação de formulários em dois dias representativos do verão (março/2013) e do inverno (agosto/2013), e em duas praças contrastantes em relação ao seu ambiente acústico e térmico bem como aos seus parâmetros morfológicos como o fator de visão do céu, o altura dos edifícios, o tipo de pavimento, a presença de fontes de água e vegetação. Os índices Nível de Pressão Sonora Equivalente Contínuo, ponderado na curva A (LAeq) e Temperatura Equivalente Fisiológica (PET) foram usados para representar, respectivamente, as condições acústicas e microclimáticas. Foram coletados por meio dos formulários variáveis subjetivas (percepção do volume do ambiente sonoro, avaliação de incômodo relacionado ao ambiente sonoro, avaliação de conforto acústico, percepção de sensações térmicas, preferência de sensações térmicas e avaliação de conforto térmico), variáveis individuais (vestimenta, atividade física, idade, peso, altura, sexo) e dadoscontrole, relacionados aos aspectos psicológicos, sociais e culturais que podem interferir na percepção acústica e térmica do ambiente. A amostra compreendeu aproximadamente 1.700 entrevistados. O tratamento estatístico dos dados coletados abarcou análise descritiva, correlações e regressões. Modelos de regressão logística ordinal foram utilizados para predizer as faixas de percepção acústica e térmica; e modelos de regressão logística, para predizer as faixas de conforto e desconforto acústico e térmico. Os resultados do estudo incluem, dentre outros: 1) a calibração do índice LAeq para percepção do volume sonoro - faixas: \"Baixo\", <35dB(A), \"Normal\", de 36 a 67dB(A), e \"Alto\", >68dB(A); e para avaliação de conforto acústico - faixas: \"Confortável\", <67dB(A), e \"Desconfortável\", >68dB(A); 2) a calibração do índice PET para percepção de sensações térmicas - faixas: \"Frio\", <18,9°C, \"Bem\", de 19 a 27°C, e \"Calor\", >27,1; e para avaliação de conforto térmico - faixas: \"Confortável\", de 23 a 31°C, e \"Desconfortável\", <22,9 e >31,1°C; 3) a definição das temperaturas neutra e preferida para verão (27,7 e 14,9°C) e inverno (15,9°C e 20,9°C), respectivamente, demonstrando a influência da expectativa na avaliação das condições térmicas; e 4) a comprovação de que o aumento do desconforto acústico pode acarretar (ainda que em pequena escala) o aumento do desconforto térmico e vice-versa. Estes resultados podem nortear o esclarecimento de questões referentes à percepção e ao conforto acústico e térmico em espaços urbanos, orientando as políticas públicas em projetos urbanísticos relacionados a esses temas. / In urban open public spaces, particularly in big cities of tropical climate, city-users are often exposed not only to high sound levels but also to high thermal loads, a situation that can cause both acoustic and thermal discomfort. Nevertheless, in most cases, the relationship between the exposure to each of these adverse conditions and human perceptions towards each of them are studied separately. In order to address the lack of a combined analysis of these conditions, this research has adopted an integrated approach to evaluate urban acoustic and thermal comfort and their likely combined effects. This study was carried out in in the Brazilian city of Belo Horizonte, in the state of Minas Gerais, a city located in a region of tropical of altitude climate, with hot wet summers and cold dry winters. Acoustic and climatic data were measured simultaneously with the administration of questionnaires in two representative days of summer (March/2013) and winter (August/2013), in two squares that noticeably differ in relation to their acoustic and thermal environment and their morphological parameters such as the sky view factor, the height of the buildings, the type of pavement, the presence of water sources and the vegetation. The LAeq,T and the PET index were used to represent acoustic and microclimatic conditions respectively. Subjective variables (perceived volume of the environmental sound, assessment of annoyance caused by environmental sound, acoustic comfort evaluation, perception of thermal sensation, thermal sensation preference and evaluation of thermal comfort), personal variables (clothing, physical activity, age, weight, height, gender) and control data related to psychological, social and cultural issues that might interfere with acoustic and thermal perception of the environment were collected through the questionnaires. The sample consisted of approximately 1,700 respondents. The statistical treatment of the data collected was comprised of descriptive analysis as well as analysis using correlations and regressions. Ordinal logistic regression models were used to predict the ranges of acoustic and thermal perception and logistic regression models were used to predict the ranges of acoustic and thermal comfort and discomfort. Some of the results of this study are: 1) the calibration of the LAeq index for perceived loudness - ranges: \"Low\", <35dB(A), \"Normal\", between 36 and 67dB(A), and \"High\", >68dB(A); and for evaluation of acoustic comfort - ranges: \"Comfortable\" <67dB(A), and \"Uncomfortable\", >68dB(A); 2) the calibration of the PET index for perceived thermal sensations - ranges, \"Cold\", <18.9°C, \"Well\", 19-27°C, and \"Hot\", >27.1°C; and for evaluation of thermal comfort - ranges: \"Comfortable\", 23-31°C, and \"Uncomfortable\", <22.9 and >31.1°C; 3) the definition of neutral and preferred temperatures for Summer (27.7 and 14.9°C) and Winter (15.9°C and 20.9°C), respectively, showing the influence of expectation on evaluation of thermal conditions; and 4) the confirmation that an increase of the acoustic discomfort may cause (albeit on a small scale) an increase in the thermal discomfort and vice versa. These results might shed light on the issues of acoustic and thermal perception and comfort in urban spaces, helping to guide public policies on urban projects related to these topics.

Em espaços urbanos abertos, particularmente nas grandes cidades de climas tropicais, os pedestres estão expostos não somente a níveis sonoros elevados como também a elevadas cargas térmicas, situação que pode gerar tanto o desconforto acústico quanto o térmico. Entretanto, na maioria das vezes, a relação entre a exposição a condições acústicas e térmicas adversas e a percepção humana dessas condições são estudadas separadamente. Neste trabalho propõe-se, portanto, uma abordagem integrada para a avaliação do conforto acústico e térmico urbano e para o estudo de seus prováveis efeitos combinados. Esta pesquisa foi realizada em Belo Horizonte - MG, Brasil, cidade localizada em região de clima tropical de altitude, com verões quentes e úmidos e invernos frios e secos. Utilizou-se o método indutivo experimental na condução dos trabalhos. Dados acústicos e climáticos foram medidos simultaneamente à aplicação de formulários em dois dias representativos do verão (março/2013) e do inverno (agosto/2013), e em duas praças contrastantes em relação ao seu ambiente acústico e térmico bem como aos seus parâmetros morfológicos como o fator de visão do céu, o altura dos edifícios, o tipo de pavimento, a presença de fontes de água e vegetação. Os índices Nível de Pressão Sonora Equivalente Contínuo, ponderado na curva A (LAeq) e Temperatura Equivalente Fisiológica (PET) foram usados para representar, respectivamente, as condições acústicas e microclimáticas. Foram coletados por meio dos formulários variáveis subjetivas (percepção do volume do ambiente sonoro, avaliação de incômodo relacionado ao ambiente sonoro, avaliação de conforto acústico, percepção de sensações térmicas, preferência de sensações térmicas e avaliação de conforto térmico), variáveis individuais (vestimenta, atividade física, idade, peso, altura, sexo) e dadoscontrole, relacionados aos aspectos psicológicos, sociais e culturais que podem interferir na percepção acústica e térmica do ambiente. A amostra compreendeu aproximadamente 1.700 entrevistados. O tratamento estatístico dos dados coletados abarcou análise descritiva, correlações e regressões. Modelos de regressão logística ordinal foram utilizados para predizer as faixas de percepção acústica e térmica; e modelos de regressão logística, para predizer as faixas de conforto e desconforto acústico e térmico. Os resultados do estudo incluem, dentre outros: 1) a calibração do índice LAeq para percepção do volume sonoro - faixas: \"Baixo\", <35dB(A), \"Normal\", de 36 a 67dB(A), e \"Alto\", >68dB(A); e para avaliação de conforto acústico - faixas: \"Confortável\", <67dB(A), e \"Desconfortável\", >68dB(A); 2) a calibração do índice PET para percepção de sensações térmicas - faixas: \"Frio\", <18,9°C, \"Bem\", de 19 a 27°C, e \"Calor\", >27,1; e para avaliação de conforto térmico - faixas: \"Confortável\", de 23 a 31°C, e \"Desconfortável\", <22,9 e >31,1°C; 3) a definição das temperaturas neutra e preferida para verão (27,7 e 14,9°C) e inverno (15,9°C e 20,9°C), respectivamente, demonstrando a influência da expectativa na avaliação das condições térmicas; e 4) a comprovação de que o aumento do desconforto acústico pode acarretar (ainda que em pequena escala) o aumento do desconforto térmico e vice-versa. Estes resultados podem nortear o esclarecimento de questões referentes à percepção e ao conforto acústico e térmico em espaços urbanos, orientando as políticas públicas em projetos urbanísticos relacionados a esses temas. / In urban open public spaces, particularly in big cities of tropical climate, city-users are often exposed not only to high sound levels but also to high thermal loads, a situation that can cause both acoustic and thermal discomfort. Nevertheless, in most cases, the relationship between the exposure to each of these adverse conditions and human perceptions towards each of them are studied separately. In order to address the lack of a combined analysis of these conditions, this research has adopted an integrated approach to evaluate urban acoustic and thermal comfort and their likely combined effects. This study was carried out in in the Brazilian city of Belo Horizonte, in the state of Minas Gerais, a city located in a region of tropical of altitude climate, with hot wet summers and cold dry winters. Acoustic and climatic data were measured simultaneously with the administration of questionnaires in two representative days of summer (March/2013) and winter (August/2013), in two squares that noticeably differ in relation to their acoustic and thermal environment and their morphological parameters such as the sky view factor, the height of the buildings, the type of pavement, the presence of water sources and the vegetation. The LAeq,T and the PET index were used to represent acoustic and microclimatic conditions respectively. Subjective variables (perceived volume of the environmental sound, assessment of annoyance caused by environmental sound, acoustic comfort evaluation, perception of thermal sensation, thermal sensation preference and evaluation of thermal comfort), personal variables (clothing, physical activity, age, weight, height, gender) and control data related to psychological, social and cultural issues that might interfere with acoustic and thermal perception of the environment were collected through the questionnaires. The sample consisted of approximately 1,700 respondents. The statistical treatment of the data collected was comprised of descriptive analysis as well as analysis using correlations and regressions. Ordinal logistic regression models were used to predict the ranges of acoustic and thermal perception and logistic regression models were used to predict the ranges of acoustic and thermal comfort and discomfort. Some of the results of this study are: 1) the calibration of the LAeq index for perceived loudness - ranges: \"Low\", <35dB(A), \"Normal\", between 36 and 67dB(A), and \"High\", >68dB(A); and for evaluation of acoustic comfort - ranges: \"Comfortable\" <67dB(A), and \"Uncomfortable\", >68dB(A); 2) the calibration of the PET index for perceived thermal sensations - ranges, \"Cold\", <18.9°C, \"Well\", 19-27°C, and \"Hot\", >27.1°C; and for evaluation of thermal comfort - ranges: \"Comfortable\", 23-31°C, and \"Uncomfortable\", <22.9 and >31.1°C; 3) the definition of neutral and preferred temperatures for Summer (27.7 and 14.9°C) and Winter (15.9°C and 20.9°C), respectively, showing the influence of expectation on evaluation of thermal conditions; and 4) the confirmation that an increase of the acoustic discomfort may cause (albeit on a small scale) an increase in the thermal discomfort and vice versa. These results might shed light on the issues of acoustic and thermal perception and comfort in urban spaces, helping to guide public policies on urban projects related to these topics.

Des réseaux de capteurs acoustiques sont actuellement mis en place dans plusieurs grandes villes afin d’obtenir une description plus fine de l’environnement sonore urbain. Un des défis à relever est celui de réussir,à partir d’enregistrements sonores, à estimer des indicateurs utiles tels que le niveau sonore du trafic routier. Cette tâche n’est en rien triviale en raison de la multitude de sources sonores qui composent cet environnement. Pour cela, la Factorisation en Matrices Non-négatives (NMF) est considérée et appliquée sur deux corpus de mixtures sonores urbaines simulés. L’intérêt de simuler de tels mélanges est la possibilité de connaitre toutes les caractéristiques de chaque classe de son dont le niveau sonore exact du trafic routier. Le premier corpus consiste en 750 scènes de 30 secondes mélangeant une composante de trafic routier dont le niveau sonore est calibré et une classe de son plus générique. Les différents résultats ont notamment permis de proposer une nouvelle approche, appelée « NMF initialisée seuillée », qui se révèle être la plus performante. Le deuxième corpus créé permet de simuler des mixtures sonores plus représentatives des enregistrements effectués en villes, dont leur réalisme a été validé par un test perceptif. Avec une erreur moyenne d’estimation du niveau sonore inférieure à 1,2 dB, la NMF initialisée seuillée se révèle, là encore, la méthode la plus adaptée aux différents environnements sonores urbains. Ces résultats ouvrent alors la voie vers l’utilisation de cette méthode à d’autres sources sonores, celles que les voix et les sifflements d’oiseaux, qui pourront mener, à terme, à la réalisation de cartes de bruits multi-sources. / Acoustic sensor networks are being set up in several major cities in order to obtain a more detailed description of the urban sound environment. One challenge is to estimate useful indicators such as the road traffic noise level on the basis of sound recordings. This task is by no means trivial because of the multitude of sound sources that composed this environment. For this, Non-negative Matrix Factorization (NMF) is considered and applied on two corpuses of simulated urban sound mixtures. The interest of simulating such mixtures is the possibility of knowing all the characteristics of each sound class including the exact road traffic noise level. The first corpus consists of 750 30-second scenes mixing a road traffic component with a calibrated sound level and a more generic sound class. The various results have notably made it possible to propose a new approach, called ‘Thresholded Initialized NMF', which is proving to be the most effective. The second corpus created makes it possible to simulate sound mixtures more representatives of recordings made in cities whose realism has been validated by a perceptual test. With an average noise level estimation error of less than 1.3 dB, the Thresholded Initialized NMF stays the most suitable method for the different urban noise environments. These results open the way to the use of this method for other sound sources, such as birds' whistling and voices, which can eventually lead to the creation of multi-source noise maps.

Submitted by Geandra Rodrigues (geandrar@gmail.com) on 2018-09-04T12:58:08Z No. of bitstreams: 1 guilhermevallelouresbrandao.pdf: 12401220 bytes, checksum: 9bc0af060b36cbc30fae487ed61aa6cd (MD5) / Approved for entry into archive by Adriana Oliveira (adriana.oliveira@ufjf.edu.br) on 2018-09-04T13:24:55Z (GMT) No. of bitstreams: 1 guilhermevallelouresbrandao.pdf: 12401220 bytes, checksum: 9bc0af060b36cbc30fae487ed61aa6cd (MD5) / Made available in DSpace on 2018-09-04T13:24:55Z (GMT). No. of bitstreams: 1 guilhermevallelouresbrandao.pdf: 12401220 bytes, checksum: 9bc0af060b36cbc30fae487ed61aa6cd (MD5) Previous issue date: 2018-03-28 / A ambiência acústica dos lugares é fator preponderante para o adequado desempenho laboral, intelectual ou simplesmente vivencial do ser humano. O local de vivência, se possuir fatores que potencializam a percepção da ambiência acústica, tais como ruído de tráfego, pode causar estresse fisiológico, contribuindo para uma avaliação qualitativa desfavorável do espaço urbano. O objeto de estudo desta pesquisa é o conjunto de características acústicas inerentes às linhas ferroviárias, que se materializam no espaço segundo esses preceitos e os irradiam ao entorno. A presença dessas linhas na malha urbana da cidade de Juiz de Fora – MG – influi diretamente na produção do espaço e direciona vetores de crescimento e valoração do solo, representando um importante aspecto a ser considerado nas políticas de produção da cidade. Este trabalho tem por objetivo identificar as características acústicas ao longo das linhas ferroviárias e suas especificidades, propondo alternativas para mitigação dos problemas verificados nas áreas adjacentes. O recorte desta pesquisa, que se insere no campo da acústica ambiental, se dá pela delimitação de estudo das características acústicas das linhas férreas, focando nas características sonoras das áreas habitadas lindeiras à faixa de domínio na malha urbana de Juiz de Fora. Com relação à metodologia de desenvolvimento, utiliza-se a Revisão Sistemática de Literatura – RSL – para definição do arcabouço teórico referencial e, para realização do estudo de caso, utiliza-se a metodologia proposta pelos referenciais normativos, através de medições in loco. O trabalho se desenvolve em seis capítulos que englobam os conhecimentos necessários à sua realização, focando na aplicação dos conceitos ao ambiente ferroviário. As análises realizadas demonstram que o ambiente acústico ao longo da linha ferroviária é drasticamente impactado pela passagem das composições, provocando o aumento do nível de ruído equivalente para além dos parâmetros normativos de conforto. A partir da situação existente, propõe-se a implantação de um misto de dispositivos convencionais e não convencionais para a redução do ruído, focando em soluções que mantenham permeabilidade visual e busquem reduzir a segregação espacial, gerando resultados que podem ser replicados em outras áreas urbanas, tanto na cidade de Juiz de Fora quanto em outras cidades. Conclui-se que a utilização dos dispositivos de proteção auxilia consideravelmente na redução do nível de pressão sonora equivalente nas áreas lindeiras, entretanto, percebe-se que é necessário o atendimento das edificações do entorno às normas construtivas para mitigação mais eficaz do problema. / Acoustic ambience is a predominant factor for adequate performance levels of work, intellect or simply experiential tasks for human beings. The vivency place itself, when possessing factors that enhance the acoustic ambience perception such as traffic noise can cause physiological stress, contributing to an unfavorable qualitative assessment of the urban space. This research’s object of study is the set of acoustic characteristics inherent to the railway lines, which materialize in space according to these precepts and radiate towards the surroundings. The presence of these railways in the Juiz de Fora city – MG – urban network directly influences the spatial production and directs land growth and valuation vectors, representing an important aspect to be considered in the city's production policies. This work aims to identify the acoustic characteristics along the railway lines and their specificities, proposing alternatives to mitigate problems observed its adjacent areas. The research clipping, which is inserted in the environmental acoustics’ field, is given by binding the study to the railway lines acoustic characteristics, focusing on the inhabited areas neighboring to the railway’s domain range sound characteristics in the Juiz de Fora city’s urban mesh. Regarding the development methodology, the systematic revision of literature method– RSL – is used to define the reference theoretical framework and, to carry out the case study, the methodology proposed by the current normative references is used, through on-the-spot measurements. The work develops in six chapters that encompass the knowledge necessary for its realization, focusing on the concept application to the railway environment. The analyses performed shows that the acoustic environment along the railway line is drastically impacted by the rail compositions passages, causing an increase in the equivalent noise level beyond the normative comfort parameters. From the existing situation, it is proposed to implement a mixture of conventional and non-conventional devices for noise reduction, focusing on solutions that maintain visual permeability and seek to reduce spatial segregation, generating results that can be replicated in other urban areas, both in Juiz de Fora and other cities. It is concluded that using protection devices greatly assists in reducing the equivalent sound pressure levels in the railway’s neighboring areas. However, it is understood that it is necessary for the surrounding buildings to meet the constructive norms for a more effective problem mitigation.

L'environnement urbain est le siège de fortes nuisances sonores notamment générées par les moyens de transport. Afin de lutter contre ces nuisances, la réglementation européenne impose la réalisation de cartographies de bruit. Dans ce contexte, des travaux fondamentaux sont menés autour de la propagation d'ondes acoustiques basses fréquences en milieu urbain. Différents travaux de recherche récents portent sur la mise en œuvre de méthodes ondulatoires pour la propagation d'ondes acoustiques dans de tels milieux. Le coût numérique de ces méthodes limite cependant leur utilisation dans un contexte d'ingénierie. L'objectif de ces travaux de thèse porte sur l'approximation unidirectionnelle de la propagation des ondes, appliquée à l'acoustique urbaine. Cette approximation permet d'apporter des simplifications à l'équation d'onde afin de limiter le temps de calcul lors de sa résolution. La particularité de ce travail de thèse réside dans la prise en compte des variations, continues ou discontinues, de la largeur des rues. Deux formalismes sont utilisés : l'équation parabolique et une approche multimodale. L'approche multimodale sert de support à une étude théorique sur les mécanismes de couplages de modes dans des guides d'ondes irréguliers bidimensionnels. Pour cela, le champ de pression est décomposé en fonction du sens de propagation des ondes à la manière d'une série de Bremmer. La contribution particulière de l'approximation unidirectionnelle est étudiée en fonction des paramètres géométriques du guide d'ondes, ce qui permet de mieux cerner les limites de validité de cette approximation. L'utilisation de l'équation parabolique a pour but une application à l'acoustique urbaine. Une transformation de coordonnées est associée à l'équation parabolique grand angle afin de prendre en compte l'effet de la variation de la section du guide d'ondes. Une méthode de résolution est alors spécifiquement développée et permet une évaluation précise du champ de pression. D'autre part, une méthode de résolution de l'équation parabolique grand angle tridimensionnelle est adaptée à la modélisation de la propagation acoustique en milieu urbain. Cette méthode permet de tenir compte des variations brusques ou continues de la largeur de la rue. Une comparaison avec des mesures sur maquette de rue à échelle réduite permet de mettre en avant les possibilités de la méthode. / The urban environment is the seat of loud noise generated by means of transportation. To fight against these nuisances, European legislation requires the achievement of noise maps. In this context, fundamental work is carried around the propagation of acoustic low-frequency waves in urban areas. Several recent research focuses on the implementation of wave methods for acoustic wave propagation in such environments. The computational cost of these methods, however, limits their use in the context of engineering. The objective of this thesis focuses on the one-way approximation of wave propagation, applied to urban acoustics. This approximation allows to make simplifications on the wave equation in order to limit the computation time. The particularity of this thesis lies in the consideration of variations, continuous or discontinuous, of the width of streets. Two formalisms are used: parabolic equation and a multimodal approach. The multimodal approach provides support for a theoretical study on the mode-coupling mechanisms in two-dimensional irregular waveguides. For this, the pressure field is decomposed according to the direction of wave propagation in the manner of a Bremmer series. The specific contribution of the one-way approximation is studied as a function of the geometric parameters of the waveguide, which helps identify the limits of validity of this approximation. Use of the parabolic equation is intended for application to urban acoustic. A coordinate transformation is associated with the wide-angle parabolic-equation in order to take into account the variation effect of the waveguide section. A resolution method is developed specifically and allows an accurate assessment of the pressure field. On the other hand, a solving method of the three-dimensional parabolic-equation is suitable for the modeling of acoustic propagation in urban areas. This method takes into account sudden or continuous variations of the street width. A comparison with measurements on scaled model of street allows to highlight the possibilities of the method.