Research On Sea Parameters Identification Method Based On Modal Number

With the development of aerospace technology, statistical energy analysis(SEA) has been of great importance to scholars to deepening research since it’s an effective tool for solving complex problems of high frequency acoustic vibration mechanics. The core of Statistical Energy Analysis environment indicatation is to get these three SEA parameters of dampling loss factor, coupling loss factor and the modal density. With the exception of a few simple structure can use obtain theoretical or semi-empirical formula of its SEA parameters, most objects need to be studied by experiment identification. In this paper we analysis the emi-static state space model of SEA parameters by experimental identification,by use of the subspace algorithm or eigensystem realization algorithm to obtain a minimal realization on the basis of the system,put forward to a new method to reconstruct the system SEA parameters directly. Also Analysis the reason why the coupling loss factor with small values can not be effectively identified. At last analysis the SEA parameters in count of incomplete input parameters.In theoretical research,we derived mainly studied the SEA parameter identification of state space model. Researched and deduced the eigensystem realization algorithm and its improved algorithms, including data correlation method to reduce the effect of noise and the state observer to broaden the experimental conditions, pointing out its limitations according to the characteristics of SEA parameter configuration; then, taking into account of the input and output based on noise, modeled on the idea of deduced the generalized observations subspace methods to deduced the state-space equations and get a minimal realization of it.Then with the minimal realization, derived a direct reconstruction method to construct SEA formula parameters, and the selection of mode number is studied and determined of the selection criteria..In the simulation, mainly through the vertical coupling plate, seven local structure model rocket subsystem and ten sub-structure model simulation analysis, to verify the feasibility of direct reconstruction method, and identify that the effect is better than the model correction algorithm. Through the theoretical analysis, the state observe /eigensystem realization algorithm and the subspace algorithms to identify the SEA parameters of satellite model, pointed out that the coupling loss factor with small value can not be effective identified because of its own contribution to the system in response to lower energy. And further stated that unless the measuring device accuracy is very high, and also the SNR, or the SEA parameters identification based on the system state space model is not effectively identified. According to the connection of the structures, combined with the recognition result of divergent elements to determine a value in the range of small values.