Abstract: Based on the section model of two-dimensional bending-torsional coupling free vibration of wind tunnel, a new hybrid algorithm based on Particle Swarm Optimization (PSO) and Grey Wolf Optimizer (GWO) is introduced on the basis of the existing least square principle to search relevant parameters, in order to accurately identify the flutter derivatives of bridge section. By comparing the flutter derivative identification results of the traditional iterative method and HGWOP under three kinds of weighted matrices for the ideal plane plate and a bridge, it shows that HGWOP can not only avoid the problem of initial value sensitivity and carry out accurate parameter search when a certain mode signal decays rapidly, but also have more stable and accurate identification under relatively complex weighted matrices. A new flutter derivative identification method is proposed, which makes it possible to apply a more scientific and complex weighted matrix.