A hybrid methodology for uncertainty analysis of vibration response in fluid-filled pipes.

Reducing vibration and noise in fluid-filled pipeline systems is critical for enhancing the acoustic stealth of underwater vehicles. However, uncertainties inherent in the complex vibro-acoustic response and transmission of these systems render traditional deterministic methods inadequate. To address this, this paper proposes a hybrid methodology, named ISM-PCE, combining the impedance synthesis method (ISM) and polynomial chaos expansion (PCE), to efficiently estimate the low-order statistical moments of the frequency response function (FRF) of pipelines, validated by experiments and numerical simulations on homogeneous straight pipes. Results show that the normal ISM-PCE accurately estimates the mean FRF under single dimensional parameter (pipe inner diameter) uncertainty, but its variance estimation accuracy is insufficient in the resonance frequency band. Therefore, a stochastic frequency transformation method was introduced, significantly improving variance estimation accuracy and enabling successful multiple dimensional parameters uncertainty analysis. The results demonstrate that the normal ISM-PCE and its improved variant provide an efficient and accurate methodology for uncertainty quantification of vibration responses in fluid-filled pipeline systems. Although only fluid-filled straight pipes have been analyzed in this paper, the proposed methodology is generalizable and applicable to more complex fluid-filled pipeline systems.