Inverse Method to Predict Fan noise

A1 - CEVAS Project (HVAC of cars)

Predicting the noise radiated by a fan can be done using several methods from the simplest (ASHRAE, very fast estimation of the spectrum without linking it to the air-flow phenomenon) to the most demanding in terms of computing resources (CFD and aero-acoustics module for instance).
Model predictions have been developed to characterize the components whose noise sources are due to fluid flow. These methods permit to determine the acoustic behavior of a component such as the fan, taking into account the operating point. In this paper, an inverse method is developed to define the acoustic source terms of a fan in a complex environment which require setting up a protocol for testing and calculating.
Based on a theoretical approach from the Euler equations, the fan is represented by a source term reflecting its ability to produce a sound. The value of the source term is quantified using a test bench to determine the intrinsic capacity of the fan to produce the sound measured, regardless of the measurement environment. Then, one may calculate the noise produced when the fan is placed inside any acoustic medium. In this paper, one will use the cavity of an automotive HVAC to correlate the noise prediction with experimental results comparing the outlet sound power level.
This work has been developed in the FUI project called Cevas.