Partial Discharge Detection and Recognition in Insulated Overhead Conductor Based on Bi-LSTM with Attention Mechanism

Insulated overhead conductor (IOC) faults cannot be detected by the ordinary protection devices due to existence of insulation layer. The failure insulated conductors is regularly accompanied partial discharge (PD); thus, IOC are often judged PDs conductors. In this paper, an intelligent PD detection model based on bidirectional long short-term memory with attention mechanism (AM-Bi-LSTM) proposed for judging faults. First, original signals processed using discrete wavelet transform (DWT) de-noising, and then signal statistical-feature entropy-feature vectors fused characterize signals. Finally, AM-Bi-LSTM network detection, in which AM able assign inputs different weights highlight their effective characteristics; identification accuracy computational complexity have been greatly improved. validity were evaluated ENET common dataset. experiment results demonstrate that exhibits a higher performance than existing models, such as LSTM, Bi-LSTM, AM-LSTM.