Validation of Dynamic Simulations of Fluid and Structure Noise and Vibration

Abstract Finite element analysis (FEA) and Computational Fluid Dynamics (CFD) are powerful tools for the design and evaluation of dynamic systems excited by fluid and acoustic sources and the vibration or acoustic response propagating through systems. However, the use of unvalidated approaches can lead to incorrect results that are then used to make decisions to approve or mitigate a system. One of those systems is piping, where the vibration of the pipe and attached equipment can result in unstable flow conditions and fatigue damage that causes pressure boundaries to be broken. Pipe vibration is also a source of potentially high noise levels that need to be reduced for worker safety and community noise impacts. One typical vibration source is flow generated noise from piping system components such as valves and fittings such as tees and expanders. Another vibration source is flow induced turbulence impacting the structure. The simulation approaches range from including analytical predictions coupled with a CFD or FEA analysis and others include a completely computational approach. However, in either approach, there are multiple parameters that need to be adjusted for accurate predictions. Further, the potential formulations and parameter values used for a simulation of the noise generated in a valve or pipe fitting is critical for accurate simulation results. With modern software packages, it is not challenging to generate simulation results, however, to ensure that a proper simulation is performed it is critical to have data to validate the simulations. Several examples will be shown of the value of test data to validate simulations. In some cases, the process facilitates the choice of parameters used in the simulation and in other cases, the validation process demonstrates how the simulation cannot be used for sufficiently accurate predictions.