Identification of Modal Parameters Using an Improved Sparse Blind Source Separation Method

During the past decade, blind source separation (BSS) method has become an effective tool to characterize and identify modal parameters of linear systems. However, in practical engineering, the assumptions of guaranteeing conventional BSS method successful application cannot be frequently satisfied, which often lead to some challenging issues. One of these challenges is how to tackle the challenge of modal identification in situations of under-determination, which means that the total sensors should be not more than that of the active modals. In this paper, we explore an efficient under-constrained BSS approach called sparse BSS (SBSS) for addressing this problem. The drawbacks of conventional BSS are first listed and an improved SBSS method is proposed to deal with the above mentioned problem, which is shown to be more suitable for engineering applications. A 5-degrees-of-freedom numerical system and two analyses are employed to confirm the effectiveness of the suggested approach. The identified outcomes of modal parameters show highly satisfied accuracy via comparative analysis, which illustrates that the proposed technique has a potential application in structural engineering.