The influence of road surface properties on vehicle suspension parameters optimized for ride-design trends for global markets

Suspension design is influenced by many factors, especially by vehicle dynamics performance in ride, handling and durability. In the global automotive industry it is common to "customize" or tune suspension parameters so that a vehicle is more acceptable to a different customer base and in a different driving environment. This paper seeks to objectively quantify certain aspects of tuning via ride optimization, taking account of market differences in road surface spectral properties and loading conditions. A computationally efficient methodology for suspension optimization is developed using stochastic techniques. A small (B-class) vehicle is chosen for the study and the following main suspension parameters are selected for optimization-spring stiffness, damping rate and vertical tire stiffness. The road is characterized as a stationary random process, using scaling and shaping filters representative of comparable roads in India and the USA. A standard quadratic function is used, including terms for vertical body acceleration, tire load variations and workspace usage. Using the stochastic nature of the ride metrics obtained, it is found possible to further penalize large excursions in the suspension working space. Weighting parameters in the ride metric are selected to match a given design to the US market condition, so that re-optimization assuming conditions relevant to the Indian market immediately provides trend directions for retuning the suspension. It is also shown how assumed increases in payload (passenger numbers etc.) or different driving speeds may be included in the tuning process. Future integration of durability and handling criteria is also discussed. The result is an efficient tuning tool that may be used to support more detailed studies in simulation and track testing, helping to reduce the time to market when suspension tuning is required. Copyright © 2012 SAE International