Aiming at the accurate extraction of voltage sag characteristic signals in harmonic and noisy environments, an extraction method based on iteration symplectic geometry mode decomposition-difference Hilbert transform (ISGMD-DHT) is proposed. First, the reconstruction trajectory matrix is constructed based on the Hamiltonian matrix and symplectic QR decomposition, and the initial symplectic geometric component is obtained by combining with the symplectic geometric similarity transformation. Secondly, the initial symplectic geometric component is fitted according to the similarity criterion and the residual component is calculated, and then the trajectory matrix is constructed according to the residual component. Next, the above operations are repeated until the iteration termination condition is satisfied to obtain the final independent symplectic geometric components. Finally, the sag characteristic quantities are extracted by the difference Hilbert transform. The analysis of simulation and measured data show that the proposed method can accurately extract the sag feature quantity under severe noise and harmonic disturbance.