Developing an accurate partial discharge (PD) monitoring system for switchgear has attracted significant attention in recent times. Detecting and distinguishing PDs with a portable PD detector is challenging due to the inherent noise interference and the similarity among different PD signals in field conditions. In this study, we propose an innovative approach that combines knowledge distillation (KD) and multitask learning (MTL) to efficiently transfer valuable insights acquired by an advanced network to a more lightweight student network. Notably, a novel energy-adaptive sampling strategy is applied to frequency feature to represent acquired pulse waveforms efficiently. Furthermore, a novel spatial information distillation module is also proposed to enhance the knowledge transfer, thereby enhancing PD recognition efficiency and accuracy. Signals captured in substations undergo clustering through a $k$ -means model, effectively separating PDs from noise. Using the phase resolved PD (PRPD) patterns generated from clustered signals, we train our PD classification network. With the integration of KD and MTL modules, our PD classification model outperforms the baseline model in terms of both speed and accuracy. A comprehensive assessment of each proposed module was conducted through ablation studies in our experiments. Furthermore, our constructed PD recognition dataset was used to conduct a comparative analysis against other methodologies, showcasing the superior performance of our approach.

Detection of partial discharge (PD) in switchgears requires extensive data collection and time-consuming analyses. Data from real live operational environments pose great challenges in the development of robust and efficient detection algorithms due to overlapping PDs and the strong presence of random white noise. This paper presents a novel approach using clustering for data cleaning and feature extraction of phase-resolved partial discharge (PRPD) plots derived from live operational data. A total of 452 PRPD 2D plots collected from distribution substations over a six-month period were used to test the proposed technique. The output of the clustering technique is evaluated on different types of machine learning classification techniques and the accuracy is compared using balanced accuracy score. The proposed technique extends the measurement abilities of a portable PD measurement tool for diagnostics of switchgear condition, helping utilities to quickly detect potential PD activities with minimal human manual analysis and higher accuracy.

Developing intelligent methods for partial discharge (PD) diagnosis, capable of handling various types of insulation defects in switchgear, has garnered significant attention in recent years. Certain PD signals exhibit similar characteristics, often leading to their confusion with noisy signals during data acquisition. To mitigate noise interference and enhance the precision of PD recognition, this article introduces a novel framework for separating PD signals from noise and acquiring discriminative features for identifying different types of PDs. Specifically, the proposed approach incorporates an adaptive frequency sampling strategy to extract effective and efficient features for the separation of PD signals and noise, followed by the clustering of the captured signals. Phase Resolved PD (PRPD) patterns are then generated for each clustered signal group, forming the PRPD pattern database. In order to identify the informative region within the PRPD patterns, we introduce spatial correlation attention and discriminative feature learning modules. These modules aim to reduce intraclass variance and increase interclass differences in the PRPD patterns. To evaluate the effectiveness of the proposed method in separating PD signals from noise and recognizing different PD patterns, we constructed a PD recognition dataset that encompasses noise as well as three types of PDs: 1) corona, 2) internal, and 3) surface. By conducting experiments and comparing the results with state-of-the-art methods, we demonstrate the performance of our method in achieving accurate PD recognition with a notable improvement of 1.9% on the constructed PD dataset.

Partial discharge is a prevalent sign of insulation aging and defects in high-voltage power equipment. Extended partial discharge can lead to ongoing physical and chemical reactions, resulting in accelerated insulation damage and eventual equipment failure. To early detect partial discharge, this paper proposes a novel wideband high-frequency coil antenna sensor-based measurement method. It maintains a lift-off distance from the conducting ground wire without the need for clamping, unlike the typical high-frequency current transformer (HFCT) sensor. The coil antenna sensor is designed with a multimode resonant ladder structure and measured with a broad impedance bandwidth. The in-lab experiment is conducted in an anechoic chamber. Subsequently, a systematic experimental setup is established with the antenna sensor for the field trial. The front end is situated in the transform and switchgear substations, while the collected data is transmitted via the Internet network to the back end for convenient monitoring. The phase resolved partial discharge (PRPD) pattern is clearly discernible. The test results demonstrate the feasibility of the proposed measurement method, affirming the efficiency of the designed coil antenna sensor along with its supporting system in partial discharge detection.

The task of detecting and classifying Partial Discharge (PD) in switchgear remains challenging due to the presence of inherent noise and the subtle differences among various PD signals under different operational conditions. Traditional PD detection methods have primarily relied either on analyzing time-series waveforms or on employing Phase-Resolved Partial Discharge (PRPD) patterns, which utilize extensive statistical information for classification. However, these approaches are notably susceptible to variations in hardware configurations and signal-to-noise ratios (SNR), limiting their effectiveness in real-world applications. This paper proposes a novel approach that enhances PD detection through multimodal fusion of time-series signals and PRPD patterns. Our method innovatively integrates these modalities by leveraging the statistical features from PRPD to enrich the time-series data, thereby extracting features that are both hardware-robust and domain-invariant. This dual-modal fusion provides a deeper insight into the characteristics of PD, enabling better discrimination among different types of discharges. Furthermore, our strategy emphasizes the learning of domain-invariant features, ensuring the model’s flexibility across different hardware setups and environmental conditions. We conducted comprehensive experiments with data collected from multiple sensors under varied conditions, demonstrating the superiority of our method compared to existing methods. The results reveal a significant improvement in PD detection accuracy and robustness, highlighting the practical utility of our method in complex and varied field environments.

Transient earth voltage (TEV) measurements are widely used for partial discharge (PD) monitoring in medium-voltage equipment due to their non-intrusive nature and suitability for field deployment. However, TEV-based PD analytics remain challenging in practical environments because PD signatures often overlap with noise characteristics. Recent advances in deep learning enable the image-based analysis of phase-resolved partial discharge (PRPD) representations, offering improved robustness compared to conventional signal-based methods. This paper presents a structured experimental investigation of classification-based and localisation-based deep learning models for TEV PRPD analytics. A multi-column convolutional neural network (MCNN) is evaluated as a classification model, while YOLOv5 and YOLOv12, based on the You Only Look Once (YOLO) framework, are investigated as object detection frameworks capable of PD localisation. Experiments are conducted using both original and expanded datasets, with additional analysis on the impact of hyperparameter optimisation. The results show that dataset expansion and hyperparameter tuning improve classification and detection performance for several models. The MCNN achieves strong classification accuracy for PD/noise screening, while YOLOv5 demonstrates substantial improvement in localisation performance after optimisation. In contrast, YOLOv12 maintains stable detection performance under the evaluated training configurations. The results highlight the trade-offs between classification accuracy, localisation capability, and annotation complexity when selecting deep learning models for PD analytics. These findings provide practical insights into the deployment of deep learning techniques for TEV-based PD monitoring systems and highlight the complementary roles of classification and localisation frameworks in PRPD analysis.