US 11,756,352 B2
Pilot protection method, device and storage medium
Ruiwen He, Guangzhou (CN); Ruisi Li, Guangzhou (CN); Yuan Jin, Guangzhou (CN); and Yiyu Lin, Guangzhou (CN)
Assigned to GUANGDONG UNIVERSITY OF TECHNOLOGY, Guangzhou (CN)
Filed by Guangdong University of Technology, Guangzhou (CN)
Filed on Nov. 4, 2022, as Appl. No. 18/52,576.
Claims priority of application No. 202111322840.6 (CN), filed on Nov. 9, 2021.
Prior Publication US 2023/0148329 A1, May 11, 2023
Int. Cl. G07C 5/08 (2006.01); G06F 18/25 (2023.01); G06F 123/02 (2023.01)
CPC G07C 5/0816 (2013.01) [G06F 18/251 (2023.01); G07C 5/0808 (2013.01); G06F 2123/02 (2023.01)] 11 Claims
OG exemplary drawing
 
1. A pilot protection method, comprising:
obtaining time-domain signals data of a target element at a preset sampling frequency;
fusing time-domain signals data of multiple first sampling periods to obtain first time-domain signals combination data based on preset fusion rules;
determining whether a fault occurs in the target element and a fault type according to the first time-domain signals combination data based on a trained machine learning model;
when it is determined that a fault occurs in the target element according to the first time-domain signals combination data, after a preset duration, based on the trained machine learning model, determining whether a fault occurs in the target element in the second sampling period and the fault type according to the second time-domain signals combination data, and the second sampling period is the sampling period after the preset duration, the second time-domain signals combination data is obtained by fusing the time-domain signals data of multiple first sampling periods, all sampling periods in the preset duration and at least one of the second sampling periods, or by fusing the time-domain signals data of at least some of the sampling periods within the preset duration and at least one of the second sampling periods, or by fusing the time-domain signals data of multiple second sampling periods; wherein the preset duration is determined according to a change trend of the performance indices of the trained machine learning model; and
when it is determined that the same type of fault occurs in the target element in multiple consecutive second sampling periods, the pilot protection system is controlled to perform the protection action on the target element according to the fault type.