Entre as etapas do desenvolvimento de automóveis pode-se apontar a definição das características de suas suspensões. A fase de definição da suspensão pode ser dividida dentro do seguinte cenário: a escolha de um determinado tipo de suspensão, os pontos (geometria) e quais os valores de rigidez/amortecimento para todo o sistema irá resultar em um comportamento dinâmico desejado para o veículo, bem como a viabilidade de produção. Além disso, o entendimento da interação entre os parâmetros de suspensão, é crucial para a otimização do desempenho. Este trabalho pretende propor um método para aperfeiçoar a fase de \"tuning\" da suspensão, com foco principal no conforto. O veículo considerado é um caminhão comercial, e entre os seus parâmetros considerados estão rigidezes de molas da cabine e suspensão, amortecimento da suspensão de cabine e curvas do amortecedor da suspensão primária. O modelo virtual do veículo foi desenvolvido em ambiente ADAMS, o qual, previamente à otimização, foi validado contra dados experimentais. Métricas foram especialmente desenvolvidas levando em consideração aspectos subjetivos de conforto veicular, para dessa forma eliminar a variabilidade entre as avaliações subjetivas e análises das simulações. Os resultados mostraram expressivas melhorias no conforto e através de dados experimentais essas melhorias foram confirmadas. / Among the development phases of an automotive vehicle one can point out the definition of the characteristics of its suspensions. Suspension definition phase can be understood as the following scenario: given a suspension type, which hard points (geometric) and what values of stiffness/damping for the whole system will result in a desired dynamic behavior for the vehicle as well as production feasibility. Moreover, understanding the iteration among the suspension parameters, even considering just the tuning ones, is crucial for performance optimization. This work intends to propose a method for vehicle tuning characteristics optimization, having as a target the ride comfort. The vehicle considered here is a commercial truck, and among its parameters one considers cabin and suspension springs, cabin dampers and suspension damper curves. A vehicle model was developed in ADAMS environment and prior to the optimization the vehicle was validated against experimental data. Metrics were specially developed to take into account subjective aspects of ride, and, in this way, eliminating the gap between subjective evaluations and simulations analysis. Results showed improvements in ride comfort. The resulting setup was measured and the improvements were confirmed with experimental data.

Entre as etapas do desenvolvimento de automóveis pode-se apontar a definição das características de suas suspensões. A fase de definição da suspensão pode ser dividida dentro do seguinte cenário: a escolha de um determinado tipo de suspensão, os pontos (geometria) e quais os valores de rigidez/amortecimento para todo o sistema irá resultar em um comportamento dinâmico desejado para o veículo, bem como a viabilidade de produção. Além disso, o entendimento da interação entre os parâmetros de suspensão, é crucial para a otimização do desempenho. Este trabalho pretende propor um método para aperfeiçoar a fase de \"tuning\" da suspensão, com foco principal no conforto. O veículo considerado é um caminhão comercial, e entre os seus parâmetros considerados estão rigidezes de molas da cabine e suspensão, amortecimento da suspensão de cabine e curvas do amortecedor da suspensão primária. O modelo virtual do veículo foi desenvolvido em ambiente ADAMS, o qual, previamente à otimização, foi validado contra dados experimentais. Métricas foram especialmente desenvolvidas levando em consideração aspectos subjetivos de conforto veicular, para dessa forma eliminar a variabilidade entre as avaliações subjetivas e análises das simulações. Os resultados mostraram expressivas melhorias no conforto e através de dados experimentais essas melhorias foram confirmadas. / Among the development phases of an automotive vehicle one can point out the definition of the characteristics of its suspensions. Suspension definition phase can be understood as the following scenario: given a suspension type, which hard points (geometric) and what values of stiffness/damping for the whole system will result in a desired dynamic behavior for the vehicle as well as production feasibility. Moreover, understanding the iteration among the suspension parameters, even considering just the tuning ones, is crucial for performance optimization. This work intends to propose a method for vehicle tuning characteristics optimization, having as a target the ride comfort. The vehicle considered here is a commercial truck, and among its parameters one considers cabin and suspension springs, cabin dampers and suspension damper curves. A vehicle model was developed in ADAMS environment and prior to the optimization the vehicle was validated against experimental data. Metrics were specially developed to take into account subjective aspects of ride, and, in this way, eliminating the gap between subjective evaluations and simulations analysis. Results showed improvements in ride comfort. The resulting setup was measured and the improvements were confirmed with experimental data.

Concevoir un composant automobile et s’assurer que celui-ci atteindra un niveau de fiabilité cible requière une connaissance précise de la variabilité des chargements que ce composant est susceptible de rencontrer dans son environnement d’utilisation. La grande diversité des chargements appliqués à différents véhicules par différents clients, ou à un même véhicule tout au long de son historique d’utilisation, représente un défi statistique majeur. Généralement, l’acquisition d’information relative à la variabilité des chargements imposés aux composants des véhicules, repose sur la réalisation de campagnes de mesures. La complexité, la durée et le coût de telles campagnes limite naturellement la taille des échantillons statistiques constitués et les chargements enregistrés sont inévitablement dépendants du véhicule utilisé pour la mesure.Le travail présenté dans ce manuscrit explore la possibilité de changer fondamentalement d’approche, en se basant sur la simulation plutôt que sur la mesure et en concentrant l’effort d’analyse statistique non pas directement sur la variabilité des chargements mais sur la variabilité des facteurs qui les déterminent. Dans ce but, des modèles stochastiques sont proposés pour décrire l’évolution de la géométrie des surfaces de routes rencontrées par les véhicules ainsi que l’évolution de la vitesse à laquelle les conducteurs les parcourent. La caractérisation de la variabilité de ces facteurs est couplée à la notion de situations de vie. Ces dernières permettent de segmenter l’historique d’utilisation des véhicules, afin de faciliter l’analyse statistique de leur évolution au sein d’une population de clients. Pour finir, la réponse dynamique du véhicule à l’excitation générée par la route est déduite par la simulation.Des données statistiques relatives à la variabilité des facteurs de route et de vitesse sont évidemment nécessaires. L’information sur les routes parcourues peut par exemple être acquise à moindre coût au moyen d’une méthode d’estimation des profils de route proposée dans ce manuscrit. Cette information peut ensuite être exploitée afin de constituer, par la simulation, à un coût très faible et pour n’importe quel véhicule dont les caractéristiques sont connues, un échantillon d’historiques de chargements aussi important que souhaité. Cette méthodologie basée sur la simulation offre la possibilité d’analyser plus largement la variabilité des chargements de fatigue provenant de la route, l’influence des différents facteurs qui les déterminent ainsi que l’effet sur la fiabilité des composants du véhicule étudié. / In order to design vehicle components that will achieve a prescribed reliability target, it is imperative to possess a precise description of the variability of the loads to which such components may be subjected within the environment in which they are used. The strong diversity of the loads imposed on different vehicles by different customers, or on a particular vehicle throughout its life, constitutes a formidable statistical challenge. Generally, the acquisition of information about the load variability experienced by vehicle components is based on the use of load measurement campaigns. The complexity, duration and cost of such campaigns naturally limit the size of the statistical samples that may be collected. Moreover, the recorded load histories are inevitably dependent on the vehicle used for the measurements.The work presented within this manuscript explores the possibility of a fundamental change in the approach to load characterisation. The objective is to make use of simulation rather than measurements and focus statistical analysis efforts not directly on load variability itself but on the variability of the factors that determine such loads. Stochastic models are proposed to describe the evolution of the geometry of road surfaces covered by vehicles, as well as the evolution of vehicles’ speed on those road surfaces. The characterisation of the variability of such factors is performed in combination with the use of life situations. The latter may be employed to divide the load histories associated to different vehicles, within a population of customers, and analyse their variation more easily. Eventually, the dynamic response of the vehicle to the excitation imposed by the road can bederived through simulation.Statistical data on the variation of the road and speed factors obviously have to be acquired in order to apply the methodology. For example, road-related information may be obtained through the use of a road profile estimation algorithm proposed within the framework of this manuscript. Such information may then be exploited to constitute, through simulation, an arbitrarily large set of load histories at a very low cost and for any vehicle whose mechanical characteristics are known.The proposed methodology based on simulation enables us to study more extensively the variability of road-induced fatigue loads, the influence of the different factors that determine such loads, as well as the effect they have on the reliability of any considered vehicle component.