Ensuring the reliability of the traction transformer, which is a critical electrical equipment for trains, is crucial for the entire electrified railway. Hence, increasing the precision of fault diagnosis for train traction transformers is necessary. A model using the dung beetle optimization (DBO) algorithm to optimize the deep belief network (DBN) to realize fault diagnosis of traction transformers is presented. Firstly, the DBO is employed to optimize the learning rate and the quantity of neurons in every hidden layer of DBN. Then, the optimized fundamental parameter values are assigned to the DBN to obtain the optimized DBO-DBN fault diagnosis model. Finally, the traction transformer's DGA online monitoring data and the manual oil sample data were used to verify the method. According to experiment findings, the proposed DBO-DBN model can identify faults with up to 95.8% precision. Compared with the basic DBN, SVM, and BPNN approaches, the proposed method's precision for classification rises by 3.3%, 8.3%, and 12.5%, respectively, which verifies the effectiveness of the proposed method. It furnishes an effective tool for the fault diagnosis of traction transformers.
Fault diagnosis of traction transformer based on improved deep belief network
Fourth International Conference on Mechanical Engineering, Intelligent Manufacturing, and Automation Technology (MEMAT 2023) ; 2023 ; Guilin, China
Proc. SPIE ; 13082
01.04.2024
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Transformer Fault Diagnosis Algorithm for Traction Power Supply System Based on IoT
British Library Conference Proceedings | 2022
|Transformer Fault Diagnosis Algorithm for Traction Power Supply System Based on IoT
Springer Verlag | 2022
|Semisupervised fault diagnosis of aeroengine based on denoising autoencoder and deep belief network
Emerald Group Publishing | 2022
|