Statistical energy analysis of noise and vibration

Abstract

Introduction Statistical energy analysis (S.E.A.) is a modelling procedure for the theoretical estimation of the dynamic characteristics of, the vibrational response levels of, and the noise radiation from complex, resonant, built-up structures using energy flow relationships. These energy flow relationships between the various coupled subsystems (e.g. plates, shells, etc.) that comprise the built-up structure have a simple thermal analogy, as will be seen shortly. S.E.A. is also used to predict interactions between resonant structures and reverberant sound fields in acoustic volumes. Many random noise and vibration problems cannot be practically solved by classical methods and S.E.A. therefore provides a basis for the prediction of average noise and vibration levels particularly in high frequency regions where modal densities are high. S.E.A. has evolved over the past two decades and it has its origins in the aero-space industry. It has also been successfully applied to the ship building industry, and it is now being used (i) as a prediction model for a wide range of industrial noise and vibration problems, and (ii) for the subsequent optimisation of industrial noise and vibration control. Lyon's book on the general applicability of S.E.A. to dynamical systems was the first serious attempt to bring the various aspects of S.E.A. into a single volume. It is a useful starting point for anyone with a special interest in the topic.