LAUSR

Abstract Modeling, analysis and optimizing the stealthiness of the submarine platforms are in the focus of many research studies. Almost all of these methods have been applied to deterministic models solved by analytical or numerical techniques. However, since the instrumentation used for measurement and fabrication usually contains some levels of uncertainty and/or error, the methods should be used in the stochastic form. In this paper, a non-intrusive stochastic approach is used for stochastic modeling and analysis of the sound absorption coating used for submarines. It consists of the stochastic modeling using Polynomial chaos expansion and sensitivity analysis with Sobol’s sensitivity indices. To this end, a simple coating is employed when all its geometrical and physical parameters are uncertain. Stochastic models are developed for its three main acoustic responses, i.e. reflection, echo reduction, and transmission loss over frequencies. Then, these stochastic models are used for fast sensitivity analysis and rank the importance of the parameters on these outputs. The results indicate that the reflection is a better response for stochastic modeling than echo reduction. Moreover, they show that considering reflection and transmission loss, the Poisson ratio and thickness of the coating are the most influential parameters.