Uncertainty analysis for improved correlation of airborne SEA model

摘要

Statistical Energy Analysis (SEA) is a well-known framework to analyze the flow of acoustic and vibration energy in structures. However, constructing a reliable SEA model is still a challenging task since the model has to synthesize various energy-transfer mechanisms between subsystems. A good SEA model requires accurate properties for various subsystems within the model and robust test-correlation techniques that support SEA modeling assumptions. In general, reciprocity measurement methods are employed to provide the power transfer functions. Uncertainties of the results represent the main drawback and this is related to different parameters such as build variation, the number of microphones and their locations used during validation, source orientation and so on. Car interior environment constitutes another factor that contributes to these uncertainties where significant absorptive or dissipative materials are present. In such an environment the direct field component from the sound source may dominate the total sound field rather than the diffuse field which is the basis of the SEA model. This paper represents an attempt to address this challenge and enable best correlation. The diffuseness of the field of the car interior will be evaluated and discussed by measuring the standard deviation based on the decay measurements. Next, transfer functions measurements to various points of the vehicle interior will be performed for a typical FCA Vehicle.