US 11,875,500 B2
Failure diagnosis method for power transformer winding based on GSMallat-NIN-CNN network
Yigang He, Hubei (CN); Liufei Shen, Hubei (CN); Liulu He, Hubei (CN); Hui Zhang, Hubei (CN); and Jiajun Duan, Hubei (CN)
Assigned to WUHAN UNIVERSITY, Hubei (CN)
Filed by WUHAN UNIVERSITY, Hubei (CN)
Filed on Jan. 29, 2021, as Appl. No. 17/161,687.
Claims priority of application No. 202010513387.6 (CN), filed on Jun. 8, 2020.
Prior Publication US 2021/0382120 A1, Dec. 9, 2021
Int. Cl. G06T 7/00 (2017.01); G01R 31/62 (2020.01); G06N 3/04 (2023.01); G06N 3/08 (2023.01); G06F 18/24 (2023.01); G06F 18/213 (2023.01); G06F 18/214 (2023.01); G06F 18/21 (2023.01); G06F 18/25 (2023.01); G06V 10/42 (2022.01); G06V 10/80 (2022.01); G06V 10/82 (2022.01)
CPC G06T 7/0004 (2013.01) [G01R 31/62 (2020.01); G06F 18/213 (2023.01); G06F 18/214 (2023.01); G06F 18/217 (2023.01); G06F 18/24 (2023.01); G06F 18/251 (2023.01); G06N 3/04 (2013.01); G06N 3/08 (2013.01); G06T 7/0002 (2013.01); G06V 10/431 (2022.01); G06V 10/803 (2022.01); G06V 10/82 (2022.01); G06T 2207/20064 (2013.01); G06T 2207/20081 (2013.01); G06T 2207/20084 (2013.01); G06T 2207/20216 (2013.01); G06T 2207/20221 (2013.01)] 10 Claims
OG exemplary drawing
 
1. A failure diagnosis method for a power transformer winding based on a GSMallat-NIN-CNN network, comprising:
Step 1: measuring a vibration condition of the transformer winding by using a multi-channel sensor to obtain multi-source vibration data of the transformer;
Step 2: converting the multi-source vibration data obtained through measurement into gray-scale images through GST gray-scale conversion;
Step 3: decomposing, by using a Mallat algorithm, each of the gray-scale images layer by layer into a high-frequency component sub-image and a low-frequency component sub-image, wherein the high-frequency component sub-images are fused through region-based property measurement, and the low-frequency component sub-images are fused through weighted averaging;
Step 4: reconstructing fused gray-scale images, and coding vibration gray-scale images according to respective failure states of the transformer winding;
Step 5: establishing a failure diagnosis model for the transformer based on the GSMallat-NIN-CNN network;
Step 6: randomly initializing network parameters to divide the fused gray-scale images and the corresponding failure state codes into a training set and a test set based on a predetermined ratio, and training and tuning the network by using the training set; and
Step 7: preserving the network which has been trained, testing the network by using the test set, and performing a failure diagnosis on a transformer to be diagnosed according to the GSMallat-NIN-CNN network which has been trained.