Abstract: Fiber reinforced materials made by adding shape memory alloy (SMA) to composite materials have the advantages of high strength, high damping and good self-repair, which can significantly improve the rigidity and peel resistance of repair structures. In order to explore the vibration characteristics of SMA repair plates with different laying quantities and methods embedded in glass fiber patches under thermal environment, a user-defined material subprogram (UMAT) was developed based on Liang-Rogers' SMA constitutive model to achieve accurate finite element solving. The changes of mode shape, natural frequency and damping ratio of traditional repair plate and SMA repair plate prepared in this paper were compared, and the influence of SMA laying configuration on the vibration characteristics of double-sided patch repair fiberglass laminate was explored at different temperatures. Experimental and simulation results show that compared with single-line (D) and vertical cross (C), the first three-order mode shape of the diagonal cross (X) SMA repair plate is closer to the traditional repair plate (Q₀), and the mode shape is independent of the number of embedded SMAs. When the SMA phase changes, the embedded SMA increases the natural frequency of the first second order of the repair plate, and the rise rate is positively correlated with the number of SMA, which is related to the SMA laying mode as follows: C-type> D-type> X-type. The embedded SMA makes the energy dissipation capacity and damping of the repair board better than that of the Q₀ board, which is positively correlated with the number of SMA, and the relationship with the SMA laying method is as follows: X-type> C-type> D-type. In the laying configuration studied in this paper, the X₂₄ type has the best repair effect, and its natural frequency of the first and second orders can be restored to 108.47% and 112.26% of the Q₀ plate, and the damping ratio is 5.19 times higher than that of the Q₀ plate. This work can provide a reference for the technical transformation of existing traditional composite repair to intelligent composite material repair.