In order to improve the safety and stability of the power system and solve the defect that traditional time-frequency transform methods cannot accurately separate and extract disturbance signals in power quality analysis, a new approach was proposed for power quality analysis based on synchronous compression conversion based on Multiple Matching Synchronous Transform (MMSST). Firstly, the method uses MMSST to decompose power signals into a series of Intrinsic Mode Functions (IMF), and then the instantaneous frequencies and amplitudes can be achieved by applying the Hilbert transform (HT) to the intrinsic mode functions. The results of simulation and measured experiments show that, compared with the Empirical Mode Decomposition (EMD) algorithm, MMSST can accurately separate and extract each disturbance component in the voltage disturbance signal, realize the accurate extraction of the instantaneous frequency and amplitude of each disturbance component, and has strong robustness.