Accounting for Manufacturing Variability in Interior Noise Computations

A formulation that accounts for manufacturing variability in the analysis of structural/acoustic systems is presented. The methodology incorporates the concept of fast probability integration with finite element (FEA) and boundary element analysis (BEA) for producing the probabilistic acoustic response of a structural/acoustic system. The advanced mean value method is used for integrating the system probability density function. FEA and BEA are combined for producing the acoustic response that constitutes the performance function. The probabilistic acoustic response is calculated in terms of a cumulative distribution function. The new methodology is used to illustrate the difference between the results from a probabilistic analysis that accounts for manufacturing uncertainty, and an equivalent deterministic simulation through applications. The probabilistic computations are validated by comparison to Monte Carlo simulations. Significant computational timesavings are achieved by the new probabilistic algorithm in comparison with the Monte Carlo Simulations. Based on its computational efficiency and its accuracy the new methodology is concluded to be a viable method of calculating numerically the probabilistic response of structural/acoustic systems due to manufacturing variability.