Wyniki wyszukiwania - BazTech

Ograniczanie wyników

1 Diagnostyka

1 2025

Znaleziono wyników: 1

Liczba wyników na stronie

Wyniki wyszukiwania

Sortuj według:

Ogranicz wyniki do:

Fault pattern recognition and prediction of transformer voicing based on neural network

Chen Jieyuan, Zhao Tiancheng, Wang Yiyu, Jin Peng, Lin Chuqiao, Zhang Hengyuan

Diagnostyka

2025

Vol. 26, No. 3

art. no. 2025315

With the complexity of power system, transformer fault detection and early warning face challenges. Traditional methods fail to identify potential failures, leading to increased risk of equipment damage and power outages. In this study, an efficient fault recognition and prediction model based on voice print signal was developed by combining convolution neural network and long and short term memory network. After training and verification, the accuracy rate of the model's recognition of common fault modes reached 95.1%. In the complex fault mode, although the recognition rate has decreased, the whole early warning system still has high reliability and practicability. The research results provide technical support for the intelligent maintenance of transformers, and help to improve the safety and stability of power system.