Loss Factors Identification from E-SEA Techniques

When designing a Statistical Energy Model (SEA) both the extension of the system with respect to the real structure/fluid and the connections between its parts are key issues in large complex systems. A previous analysis of its practical extent may allow simplifying the model to a smaller one while still taking into account all the significant energy from the paths of higher orders. On the other hand, the analysis of the connections between the parts of the system, will help provide an adequate and accurate SEA matrix where all the connexions (even the non-resonant ones) are considered, and all the non-connected subsystems are identified. In this work, a methodology to identify and classify the coupling loss factors of a given system is proposed. This methodology is based on the application of pattern recognition techniques (such a clustering) to experimental results obtained through an experimental SEA. A description of the methodology is presented, and its advantages and performances are highlighted by an application case.