Pulse signal recovery method and related equipment
By fitting pulse signals in the power system and training deep learning model, the problem of poor recovery effect of traditional methods in high noise environments is solved, and higher recovery accuracy and robustness are achieved, and the performance of the power time synchronization system is improved.
CN119719734APending Publication Date: 2025-03-28FIBRLINK NETWORKS
0 Cites 0 Cited by
Patent Information
- Application Number
- CN202411704650.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-11-26
- Publication Date
- 2025-03-28
AI Technical Summary
Technical Problem
The traditional pulse signal recovery method is poor when processing signals in high noise environments in power systems, and cannot effectively process nonlinear signals, and is poorly robust to noise and interference.
Method used
By fitting the acquired pulse signals, a recovery model to be trained is constructed, and the model is trained using a deep neural network to obtain the recovery model, and then the recovery pulse signals to be restored.
Benefits of technology
It improves the accuracy and robustness of pulse signal recovery, can handle noise interference and complex signal characteristics more effectively, and improves the performance and safety of power time synchronization systems.
✦ Generated by Eureka AI based on patent content.
Abstract
The invention provides a pulse signal recovery method and related equipment. The method comprises the following steps: performing fitting processing on an acquired pulse signal to obtain a first signal; constructing a to-be-trained recovery model based on the features of the pulse signals; training the to-be-trained recovery model based on the pulse signal and the first signal to obtain a recovery model; and performing recovery processing on a to-be-recovered pulse signal based on the recovery model. According to the embodiment of the invention, the pulse signal is recovered by using a deep learning method, and noise features which are difficult to process by a classical algorithm can be deeply captured, so that more accurate optimization compensation is realized, and the accuracy and robustness of pulse recovery are remarkably improved.
Need to check novelty before this filing date? Find Prior Art