Abstract: Partial discharge detection and localization of indoor electrical equipment is an effective means to ensure the long-term stable operation of the equipment, and the time-delay estimation accuracy is an important factor affecting the accuracy of partial discharge detection and localization. In order to solve the problem of delay estimation accuracy of partial discharge signal under the influence of noise and multipath effect, this paper proposes a partial discharge signal preprocessing method based on double ensemble empirical mode decomposition (EEMD) decomposition and reconstruction, which effectively improves the signal-to-noise ratio of the partial discharge signal and the accuracy of the delay estimation algorithm. Simulation and experimental test results show that the proposed method is more stable and robust and can effectively improve the delay estimation accuracy compared with the generalized mutual correlation algorithm. The proposed method is used for the accurate estimation of partial discharge signal delay in indoor environments with low signal-to-noise ratio and obvious multipath effect.