Indirect input identification in multi-source environments by principal component analysis

The paper deals with problems in the indirect input identification when the number of simultaneously usable measurement and processing channels is smaller than the total number of response points. In such a case, the output spectral matrix essential to the indirect input identification can be obtained by processing the responses with respect to a reference point in a sequential manner, which is the so-called transmissibility function approach. This conventional transmissibility function approach is applicable only to a case where the number of independent excitation sources is just one regardless of the number of input points at which the input forces are to be estimated. In this paper, the technique is extended to other cases where the number of independent sources is greater than one based upon the principal component analysis. A method to identify the number of independent excitation sources is also presented. The validity of the proposed method is demonstrated by a numerical example. (C) 2002 Elsevier Science Ltd. All rights reserved.