2022
DOI: 10.1088/1361-6501/ac7a09
|View full text |Cite
|
Sign up to set email alerts
|

A novel federated deep learning framework for diagnosis of partial discharge in gas-insulated switchgear

Abstract: In recent years, many different deep learning methods have been developed to ensure the safe and stable operation of gas-insulated switchgear (GIS). However, the use of these methods to achieve excellent results depends on obtaining as much training data as possible, which is difficult to accomplish because of conflicts of interest among different clients and privacy concerns. To address this issue, this paper proposes a novel federated deep learning (FDL) for the diagnosis of partial discharge (PD) in GIS. A … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
4
1

Citation Types

0
6
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
6

Relationship

2
4

Authors

Journals

citations
Cited by 7 publications
(6 citation statements)
references
References 32 publications
0
6
0
Order By: Relevance
“…The proposed method Simultaneous diagnosis and location of small samples 98.73 9.06 11.09 [1] PD location 10.85 13.89 [7] Multi-source PD diagnosis 95.42 [11] PD diagnosis 96.14 [14] PD location 10.36 13.29 [19] Simultaneous diagnosis and location 92.56 12.09 15.63 proposing a novel approach to GIS PD diagnosis and location based on the digital twin. Furthermore, in terms of diagnostic and location performance, the proposed method achieves simultaneous PD diagnosis and location, significantly improving the performance in the target domain.…”
Section: Accuracy (%) Mae (Cm) Rmse (Cm)mentioning
confidence: 99%
“…The proposed method Simultaneous diagnosis and location of small samples 98.73 9.06 11.09 [1] PD location 10.85 13.89 [7] Multi-source PD diagnosis 95.42 [11] PD diagnosis 96.14 [14] PD location 10.36 13.29 [19] Simultaneous diagnosis and location 92.56 12.09 15.63 proposing a novel approach to GIS PD diagnosis and location based on the digital twin. Furthermore, in terms of diagnostic and location performance, the proposed method achieves simultaneous PD diagnosis and location, significantly improving the performance in the target domain.…”
Section: Accuracy (%) Mae (Cm) Rmse (Cm)mentioning
confidence: 99%
“…Recently, data security policies promote the requirement that private data cannot be shared with other users, which poses a challenge for training fault diagnosis models without data aggregation. Federated learning (FL) is developed rapidly in intelligent fault diagnosis [3,4] for providing a privacy protection mechanism to train machine learning models using decentralized data and computing resources. To protect * Authors to whom any correspondence should be addressed.…”
Section: Introductionmentioning
confidence: 99%
“…In federated learning, data owned by each client is not shared, making it difficult to ensure that it meets the independent and identically distributed (IID) assumption. Addressing the challenge of statistical heterogeneity among multiple parties has been a focus of federated learning research [3,4]. As for fault diagnosis, different equipment or equipment fleets' working conditions often vary, resulting in a shift in data distribution that can slow down or even prevent the convergence of federated models.…”
Section: Introductionmentioning
confidence: 99%