IDENTIFICATION OF STATISTICAL ENERGY ANALYSIS PARAMETERS USING SUBSPACE METHOD

The reliable estimate of statistical energy analysis (SEA) parameters is one of the most important steps in a SEA prediction work for high frequency vibration response. The theory of the first order power flow model identification as well as the SEA model improvement is studied. A method for SEA parameters identification using subspace method is proposed, which is on the basis of the multi-variable output error state space (MOESP) algorithm and constrained optimization theory. The numerical simulation result demonstrates that the proposed method is feasible and has good anti-disturbance performance. Also experimental SEA on a L-shaped plate structure is processed, the results of a subspace and a power input method are in agreement, which further verify the former. The present study manifests that SEA model update and parameter identification can be carried out by using transient vibration data in time domain, which provides a good complement to experimental SEA.