Estudo de metodologias de avaliação da força de resistência ao deslocamento de um veículo

• Ano de defesa: 2018
• Autor(a) principal: Arthur Braga Thiriet
• Tipo de documento: Dissertação
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Veículos; Desaceleração livre; Resistência ao deslocamento; Engenharia mecânica
• Link de acesso: http://hdl.handle.net/1843/RAOA-BCGGW6

The coastdown test, described in Brazil by the standard ABNT 10312, consists in letting the vehicle coast freely in a speed interval. With its results is possible to estimate the road loads acting on the vehicle, that is, the sum of all forces that oppose its displacement. However, this method only allows to evaluate the vehicle as a hole, becoming difficult to analyze certain subsystems individually. Moreover, it is highly susceptible by uncontrollable external factors, such as environmental temperature and pressure, air relative humidity and wind speed and direction, which increases the test uncertainty. Another negative point from those external factors is the dependence on environmental specific conditions which could turn in delays and high costs during the development of a project. Therefore, this work seeks to evaluate two methodologies that have the potential to remedy the difficulties presented, one experimental and another virtual. The first consists in measuring the resistive force at each subsystem separately, at the main four sources of resistance: driveline, tires, brakes and aerodynamic. The latter performs the same traditional coastdown test but in a virtual environment. In order to test them, three vehicles from different segments were selected with two kinds of tire each. Every set of vehicle-tire was teste accordingly to the standard for reference values. Also, each tire had its rolling resistance measured before and after the references tests in order to evaluate the tires behavior ahead and afterwards of breaking-in. Then, all vehicles were tested with the coastdown breakdown methodology. Next, the virtual models were simulated via the commercial software Adams/Car R . Finally, the results were analyzed during the driving schedules from the standard ABNT 6601 to calculate the demanded energy from the vehicle. It could be seen that the coastdown breakdown methodology is valid as the deceleration curves from two out of three vehicles were extremely close. The mean deviation was approximately 1% with peak of 2,03% difference, which is very satisfactory considering that the uncertainty of the standard method is higher than that. The same cannot be said for the virtual test, in which the results differed from 0,02% to 5,71% in demanded energy, showing no pattern and always values much lower than the real. At the end it was concluded that the breakdown method is valid and useful due to the fact that has better repeatability than the ABNT 10312 method as it is performed in a controlled room. Furthermore it has the advantages of evincing each subsystem, helping the development of components separately. Finally, the virtual method with this specific software showed incapable to represent the reality and did not presented a conclusive pattern, which reinforces the idea that the experimental test cannot be replaced.