: Seismic signals contain a variety of attributes, but different attributes are distributed in different frequency bands, so it is necessary to accurately decompose seismic signals into different frequency components, and carry out subsequent denoising processing more effectively. Variational modal decomposition is an effective method for analyzing non-stationary signals. Aiming at the problem of difficult parameter selection of unknown signals, a seismic signal denoising method based on the combination of variational modal decomposition based on quantum particle swarm optimization and Teager energy operator is proposed. The quantum particle swarm optimization algorithm makes the variational modal decomposition algorithm adaptively decompose the signal into eigenmodal functions of different frequency bands, which is convenient for subsequent filtering of signals with different frequencies. In the simulation experiment, the traditional time-frequency analysis method and this method were used to analyze and compare the synthetic signal. According to the comparison results of time-frequency diagram and spectrum diagram, the results show that this method is a more effective seismic signal denoising method.