Incipient Fault Detection in Power Distribution System: A Time–Frequency Embedded Deep-Learning-Based Approach

Abstract

Incipient fault detection in power distribution systems is crucial to improve the reliability of the grid. However, the non-stationary nature and the inadequacy of the training dataset due to the self-recovery of the incipient fault signal, make the incipient fault detection in power distribution systems a great challenge. In this paper, we focus on incipient fault detection in power distribution systems and address the above challenges.In particular, we propose an ADaptive Time-Frequency Memory (AD-TFM) cell by embedding wavelet transform into the Long Short-Term Memory (LSTM), to extract features in time and frequency domain from the non-stationary incipient fault signals.We make scale parameters and translation parameters of wavelet transform learnable to adapt to the dynamic input signals.Based on the stacked AD-TFM cells, we design a recurrent neural network with ATtention mechanism, named AD-TFM-AT model, to detect incipient fault with multi-resolution and multi-dimension analysis. In addition, we propose two data augmentation methods, namely phase switching and temporal sliding, to effectively enlarge the training datasets. Experimental results on two open datasets show that our proposed AD-TFM-AT model and data augmentation methods achieve state-of-theart (SOTA) performance of incipient fault detection in power distribution system. We also disclose one used dataset logged at State Grid Corporation of China to facilitate future research.

Fault Detection and Fault Diagnosis in Power System Using AI: A Review

“…The article proposes an Adaptive Time-Frequency Memory (AD-TFM) cell embedded in Long Short-Term Memory (LSTM) to detect incipient faults in power distribution systems. The model, called the AD-TFM-AT model, uses learnable scale and translation parameters to detect faults in time and frequency domains [28].…”

Enhanced Wavelet Transform Dynamic Attention Transformer Model For Lithium-ion Battery Anomaly Detection

“…As a vital component of these systems, the stability and safety of battery performance directly impact the reliability and economic efficiency of the entire system. Researchers have been dedicatedly analyzing changes in key parameters such as voltage, current, and temperature during the actual use of batteries, striving to predict the health status and potential fault types of batteries [21]- [23]. This analytical approach relies not only on traditional theoretical research based on physical and chemical models but also includes data-driven research utilizing advanced machine learning and signal processing technologies.…”

“…Therefore, accurately monitoring the state of batteries and promptly detecting anomalies have become key technical challenges to ensure the safe operation of EVs. The non-stationarity and complex dynamics of battery charging data further increase the difficulty of monitoring, making traditional monitoring methods inadequate for practical applications [4]- [6].…”

Development of overcurrent relay based on wavelet transform for fault detection in transmission line

“…To solve the restrictions of traditional relays, several methods were studied and proposed 8 – 20 such as the travelling method 8 – 11 , the Fourier transform 12 , 13 , and the Wavelet transform 14 – 16 .…”

“…The reason was the original signal has noise and high frequency effect The accuracy of the traveling wave is reduced when used in more complex electrical systems. It suitable for use in single end terminals Fourier transform 12 , 13 Can be used to analyze faults in high voltage power system Processing time and accuracy are good Usage conditions are limited by having to choose to analyze only frequencies or magnitudes on one-time domain Wavelet transform 14 – 16 Developed further from the Fourier transform. Fourier constraint has been solved which parameter of frequencies and magnitudes can be extracted on one-time domain It is widely used in research related to power systems and the performance is satisfactory Not suitable for analyzing large power systems Most analyzes are based on simulations only.…”