The Automotive Suspension Design and Analysis Considering Uncertainties
Published August 22, 2005 by Society of Automotive Engineers of Korea in South Korea
In this research, the design considering uncertainties and uncertainty analysis are sequentially executed for an automotive suspension system. In the suspension design, random design variables, such as the hardpoint and bushing stiffness, have a great influence on the suspension's performance. In order to analyze the individual effects of the hardpoint and bushing stiffness on the K&C characteristics, the design problem can be considered in three cases. The first case is where the hardpoint is the only random design variable, the second is where only the bushing stiffness is the random design variable and the third is where both are considered random design variables. FRAMAX, a general-purpose, computer-assisted design framework, is used to integrate the design components and execute the optimization and variability analysis. In order to find the most robust design, an efficient reliability-based design optimization algorithm, known as the single-loop-single-vector approach, was utilized. The target reliability level was set to 3-sigma. To estimate the variability or fraction defective, descriptive sampling was applied, often considered to be one of the most efficient sampling methods available. Using this sampling method, the probability of design satisfaction and descriptive statistics can be calculated. In addition to the uncertainty analysis, the sampled data can also be utilized in a regression analysis to determine the individual effect of each input variable on the responses. It can then be utilized as a guideline for tolerance allocation.