Output-Only Time-Varying Modal Parameters Identification Using Recursive Subspace Algorithm Based Modified FAPI

Online output-only modal identification of linear time varying (LTV) systems is an interesting problem in monitoring the health condition and controlling the dynamic properties of engineering structures. The output-only time-varying modal parameters identification method based on rapid convergence subspace tracking is proposed in this paper. First, the time-varying form of the stochastic time-varying state-space (TSS) model is described, then the fast approximate power iteration (FAPI) algorithm is modified and introduced to track the time-varying pseudo-observable matrix, and then identify the time-varying pseudo modal parameters. The effectiveness and robustness of the proposed method are verified by numerical examples with 10% noise. Compared with a recursive method based on projection approximation subspace tracking (PAST) and stable generalized yet another subspace tracker (SGYAST), the proposed method possesses rapid convergence and low computing cost.