Discharge Detection Research Articles

Expert level of detection of interictal discharges with a deep neural network.

Deep learning methods have shown potential in automating the detection of interictal epileptiform discharges (IEDs) in electroencephalography (EEG). We compared IED detection using our previously trained deep neural network with a group of experts to assess its potential applicability. First, we performed clinical validation on an internal data set. Seven experts reviewed all EEG studies. Performance agreement between experts and the network was compared at both the EEG and IED levels. All EEG recordings were also processed with Persyst. Subsequently, we performed external validation, with data from four centers, using a hybrid approach, where detections by the deep neural network were reviewed by an expert. In case of disagreement with the original report, the EEG recording was annotated independently by five experts. For internal validation we included 22 EEG studies with IEDs and 28 EEG studies from controls. At the EEG level, our network showed performance similar to that of the experts. For individual IED detection, the sensitivities between experts ranged from 20.7%-86.4%, whereas the sensitivity of our network was 82.5% (confidence interval [CI]: 77.7%-87.4%) at 99% specificity and a false detection rate (FDR) of <.2/min, outperforming Persyst, with 64.6% sensitivity (CI: 61.4%-67.9%) at 98% specificity. External validation in 174 EEG studies demonstrated that all 85 EEG recordings classified as normal in the original report were classified correctly, with an FDR of .10/min. Of the 89 EEG studies with IEDs according to the report, 56 were correctly classified (Cohen's κ = .62). Visual analysis of the remaining 33 EEG recordings showed high interobserver variability among the five experts (Fleiss' κ = .13). Our deep neural network detects IEDs on par with clinical experts. The external validation in a hybrid approach showed substantial agreement with the original report. Disagreement was due mainly to high interobserver variability. Our deep neural network may support visual EEG analysis and assist in diagnostics, particularly when human resources are limited.

Acoustic emission analysis for corona discharge detection in medium-voltage cubicles: a review

Acoustic emission analysis for corona discharge detection in medium-voltage cubicles: a review

Review on peak detect and hold circuits and their applicability in partial discharge detection

Abstract This article reviews various designs of electrical peak detect and hold circuits and tests their applicability for the detection of partial discharge pulses. Since partial discharges are very short pulses in the nanosecond range, the peak detector circuit must be very fast and thus work precisely over a wide bandwidth up to the GHz range. The article begins with a comprehensive literature review and gives an overview of the state of the art of peak detector circuits. As a result of the literature research, four different peak detector designs are identified that appear suitable for measuring partial discharges. These four circuits are then simulatively tested for their performance in measuring the pulse height of various partial discharge pulses. It turns out that two of the identified circuits are able to measure nanosecond pulses of small amplitude. These two circuits must be further optimized for partial discharge detection in the future. In addition, the circuit should be tested prototypically with real partial discharge pulses.

Combining multi-level feature extraction algorithm with residual graph convolutional neural network for partial discharge detection

Partial discharge (PD) identification is critical for the insulation diagnosis of cable conductors; however, there is still scope for enhancement in the existing adaptive extraction capabilities and feature utilization for PD signals. In this context, this paper introduces a method that integrates a one-dimensional convolutional neural network (1D-CNN) with a residual graph convolutional neural network (ResGCN) to recognize PD signals. Noise analysis is performed using various combinations of mother wavelets, and Bayesian optimization is employed to mitigate background noise. Key features of PD signals are progressively abstracted through a 1D-CNN-based multilevel automatic feature learning method, while signal timing attributes are maintained to minimize manual intervention. The graph data is constructed using the signal feature matrix and the signal timing feature similarity matrix. This is followed by the development of a ResGCN utilizing a graph attention mechanism to integrate node feature information and the topology of the PD graph data. This approach aims to fully exploit the correlation between local regions of the feature space and the temporal numerical properties of the signals. Additionally, it jointly optimizes feature extraction and model classification to facilitate adaptive diagnosis. The method is validated with extensive real experimental data obtained from medium voltage overhead power lines. It demonstrates exceptional performance and practicality, achieving an accuracy rate of 97.3% and a recognition rate of 96.1% for PD samples, thus offering reliable theoretical support for effective PD detection.

Deep learning based automatic detection and dipole estimation of epileptic discharges in MEG: a multi-center study.

Magnetoencephalography (MEG) provides crucial information in diagnosing focal epilepsy. However, dipole estimation, a commonly used analysis method for MEG, can be time-consuming since it necessitates neurophysiologists to manually identify epileptic spikes. To reduce this burden, we developed the automatic detection of spikes using deep learning in single center. In this study, we performed a multi-center study using six MEG centers to improve the performance of the automated detection of neuromagnetically recorded epileptic spikes, which we previously developed using deep learning. Data from four centers were used for training and evaluation (internal data), and the remaining two centers were used for evaluation only (external data). We used a five-fold subject-wise cross-validation technique to train and evaluate the models. A comparison showed that the multi-center model outperformed the single-center model in terms of performance. The multi-center model achieved an average ROC-AUC of 0.9929 and 0.9426 for the internal and external data, respectively. The median distance between the neurophysiologist-analyzed and automatically analyzed dipoles was 4.36 and 7.23mm for the multi-center model for internal and external data, respectively, indicating accurate detection of epileptic spikes. By training data from multiple centers, automated analysis can improve spike detection and reduce the analysis workload for neurophysiologists. This study suggests that the multi-center model has the potential to detect spikes within 1cm of a neurophysiologist's analysis. This multi-center model can significantly reduce the number of hours required by neurophysiologists to detect spikes.

Partial Discharge Detection Technique Based on Full Convolutional Network with Channel-wise Convolution and Channel Mixing

Abstract The occurrence of partial discharge (PD) serves as an indication of power equipment failure, which, if not given due attention, may result in severe power accidents. Traditionally, the detection of PD involves applying various pre-processing techniques to the original signal and extracting specialized signal features that are subsequently classified using a classifier. The process is intricate and demands extensive expertise. One drawback of the conventional detection approach lies in its failure to consider the information conveyed between the phases of a three-phase electrical signal during the process of identifying PD in medium voltage three-phase cables. The present study proposes a novel method for detecting PDs, which is based on channel-wise convolution and channel mixing(C2CM) using full convolutional networks. Firstly, a set of pulses for each phase of a three-phase signal is extracted and combined to form a three-phase pulse set, serving as the input to the network. Subsequently, we introduce a FCN-C2CM network designed specifically for PD detection, capable of learning both the time-domain features of the input signal and the phase features of the three-phase signal, thereby enhancing recognition accuracy. We evaluated our approach using online testing tools on publicly available datasets, with experimental results demonstrating its superiority over existing.

On image transformation for partial discharge source identification in vehicle cable terminals of high‐speed trains

AbstractPartial discharge (PD) detection of cable terminals is crucial for the safe operation of the traction power system in trains. However, similar PD signals in complex train‐operating environments cause difficulty to recognise the insulation defects. Therefore, a PD signal image transformation recognition method is proposed for PD detection of cable terminal defects to identify defects in cable terminals with similar PD characteristics accurately. In the proposed method, the raw PD signals are firstly transformed to images via the Gramian angular field (GAF) representation. This can reveal the discriminative characteristics embedded in the original PD signals and subsequently facilitate differentiating the PD sources, which exhibit similar characteristic in the time domain. The obtained GAF representation of PD signals (named as PD GAF images) is extracted from local and global features to train an efficient MobileVIT model, which is then utilised to identify similar types of PD sources in cable terminals. The results show that the proposed method achieves 97.5% recognition accuracy in the field experiment, which is superior to other methods.

The implementation of a cost‐efficient damped AC testing methodology for transmission cables based on DC–AC conversion and distributed partial discharge detection

AbstractThis research introduces a novel damped alternating current (DAC) testing methodology, which integrates an enhanced DAC generator with a pulse‐based distributed partial discharge (PD) detection technique. The advanced DAC generator is designed without the need for high voltage (HV) solid‐state switches, thereby offering a cost‐effective solution for field‐testing voltage generation through a direct current–alternating current conversion approach. To meet the demand for high instantaneous power, a capacitor bank is employed as the power supply. Furthermore, the implementation of a distributed PD detection technique enhances sensitivity and eliminates limitations associated with cable length. To minimize the construction costs of the distributed PD‐detection system, a pulse synchronization technique has been employed. Efforts to reduce the system's weight were informed by simulations, resulting in the design and development of a prototype weighing 850 kg for a 64/110 kV power cable. The reduction in the number of HV solid‐state switches contributes to significant cost savings, amounting to tens of thousands of dollars, when compared to conventional DAC generators. Laboratory and field tests validated the effectiveness of the cost‐efficient DAC testing methodology.

Ultra-High Frequency Sensors for Detecting Partial Discharge in Power Cables

Partial discharge (PD) detection is vital for evaluating the health of power cable insulation, where failure can lead to significant disruptions. This study investigates the use of Ultra-High Frequency (UHF) sensors to detect PD signals, which typically occur in nanoseconds. The sensors operate in the GHz range and minimize interference by selecting frequencies above 400 MHz, where background noise is reduced. Through comparison with other detection methods, UHF sensors demonstrate superior sensitivity in measuring minor discharges, providing a reliable means for on-site condition assessment of high-voltage equipment. The study concludes that UHF sensors offer a practical solution for improving PD detection accuracy, particularly in challenging field conditions.

Differences in subcortical functional connectivity in patients with epilepsy.

Epilepsy is a disease characterized by abnormal paroxysmal bioelectrical activity in the brain cortex and subcortical structures. Seizures per se change brain metabolism in epileptic focus and in distal parts of the brain. However, interictal phenomena can also affect functional connectivity (FC) and brain metabolism in other parts of the brain. We hypothesised that epilepsy affects functional connectivity not only among cortical, but also between subcortical, structures of the brain in a resting state condition. Investigating functional connectivity in patients with epilepsy could provide insights into the underlying pathophysiological mechanisms. Better understanding may lead to more effective treatment strategies. Functional connectivity was analysed in 35 patients with epilepsy and in 28 healthy volunteers. The group of patients was divided into generalised and focal epilepsy (temporal and extratemporal subgroups). Each patient and healthy volunteer underwent an fMRI resting-state session. During the study, EEG signals were simultaneously recorded with fMRI to facilitate the subsequent detection of potential interictal epileptiform discharges (IEDs). Their potential impact on BOLD signals was mitigated through linear regression. The data was processed and correlation coefficients (FC values) between the BOLD signal from selected structures of the central nervous system were determined and compared between study groups. The results were presented as significant differences in correlation coefficients between brain/subcortical structures in the epilepsy and control groups. Lower FC values for the epilepsy group compared to the control group were shown for connections related to the MPFC, hippocampus, thalamus, amygdala, and the parahippocampal gyrus. Epilepsy alters the functional connectivity of resting state subcortical networks. Patterns of pathological changes of FC differ between epilepsy subtypes, with predominant lower FC between the hippocampus, parahippocampal gyrus, amygdala and thalamus in patients with epilepsy. This study suggests that epilepsy affects subcortical structures. Identifying distinct patterns of altered FC in epilepsy subtypes may help to tailor treatment strategies. Changes in FC detected by fMRI may precede clinical symptoms, aiding in the early diagnosis of cognitive and emotional disorders in focal epilepsy.

AI-Based Electroencephalogram Analysis in Rodent Models of Epilepsy: A Systematic Review

About 70 million people globally have been diagnosed with epilepsy. Electroencephalogram (EEG) devices are the primary method for identifying and monitoring seizures. The use of EEG expands the preclinical research involving the long-term recording of neuro-activities in rodent models of epilepsy targeted towards the efficient testing of prospective antiseizure medications. Typically, trained epileptologists visually analyse long-term EEG recordings, which is time-consuming and subject to expert variability. Automated epileptiform discharge detection using machine learning or deep learning methods is an effective approach to tackling these challenges. This systematic review examined and summarised the last 30 years of research on detecting epileptiform discharge in rodent models of epilepsy using machine learning and deep learning methods. A comprehensive literature search was conducted on two databases, PubMed and Google Scholar. Following the PRISMA protocol, the 3021 retrieved articles were filtered to 21 based on inclusion and exclusion criteria. An additional article was obtained through the reference list. Hence, 22 articles were selected for critical analysis in this review. These articles revealed the seizure type, features and feature engineering, machine learning and deep learning methods, training methodologies, evaluation metrics so far explored, and models deployed for real-world validation. Although these studies have advanced the field of epilepsy research, the majority of the models are experimental. Further studies are required to fill in the identified gaps and expedite preclinical research in epilepsy, ultimately leading to translational research.

Detecting and Locating Corona Discharges in Low-Pressure Aircraft Environments Using Optical Sensors

Today, there is a clear trend toward electrification of transportation systems, including aircraft. Due to the enormous power requirements, they must operate at high voltages. However, the combined effect of higher voltage levels and low pressure environments is conducive to the occurrence of electrical discharges in electrical systems. Therefore, there is an urgent need to develop cost-effective systems to detect the discharges in the early stages before major failures can occur. This paper compares two optical sensors for early discharge detection in a simulated aircraft environment. The experimental study is performed in a low pressure chamber using a needle plane electrode. The pressure is changed from that corresponding to ground level to that corresponding to flight altitude, i.e., from 100 kPa to 20 kPa. The effect of the supply frequency is also studied, since modern aircraft operate in a wide range of frequencies up to about 800 Hz. Both variables, especially pressure, have shown significant effects on the CIV value. The results have also shown a similar sensitivity of both sensors for all the experimental conditions analysed, allowing a fast and sensitive detection and localization of incipient electrical discharge activity.

Design and research of high-frequency electromagnetic wave detection antenna for discharge of transmission lines

Abstract Overhead lines have a wide range of applications in power systems. When early insulation faults occur in the long-term operation of overhead lines, corona discharge may occur. Timely detection of corona discharge can guide maintenance personnel to discover and eliminate faults in a timely manner. The height and distribution of overhead lines are relatively high, and close-range detection cannot quickly locate the discharge area. Using radio frequency antennas to measure the electromagnetic wave signals generated by corona discharge can achieve long-distance positioning detection of corona discharge. This article uses a log-periodic antenna as the detection antenna and simulates the antenna structure to prove that its directionality, gain, and measurement frequency band meet the design requirements. Finally, the discharge detection performance of the antenna is verified through experiments.

Novel Antenna for Partial Discharge Detection and Classification: A Convolutional Neural Network-Based Deep Learning Approach

Novel Antenna for Partial Discharge Detection and Classification: A Convolutional Neural Network-Based Deep Learning Approach

Partial discharge detection, diagnosis, and location technology based on a novel transient earth voltage-ultrasonic integrated sensor used for switchgear equipment.

Medium-voltage (MV) switchgear and gas insulated switchgear (GIS) are essential components of power-grid systems. Their safe operation is crucial for the electricity consumption of users. However, insulation faults often occur during the operation of MV switchgear and GIS because of pollution, humidity, heat, mechanical shock, and other factors. Partial discharge (PD) measurements are the most effective indicator to prevent insulation failure. Transient earth voltage (TEV) and ultrasonic methods are the most popular PD measurement methods for switchgear. Currently, these two methods are used widely and independently. In this study, a novel TEV-ultrasonic integrated sensor is proposed based on the independent structure of the TEV and the ultrasonic sensor. The performance parameters of the proposed sensor were tested on test platforms. PD measurement experiments were conducted in a 10kV switchgear and 220kV GIS to analyze the performance of the integrated sensor. The results show that the sensor can simultaneously measure the TEV and ultrasonic signals in the same location. The integrated sensor can realize effective and sensitive detection, precise location, and accurate diagnosis of PD in the MV switchgear and GIS.

Research on Miniaturized and Wideband UHF Sensing Technology for Partial Discharge Detection of Power Equipment Based on Serpentine Bending

Research on Miniaturized and Wideband UHF Sensing Technology for Partial Discharge Detection of Power Equipment Based on Serpentine Bending

IoT-enabled advanced monitoring system for tubular batteries: Enhancing efficiency and reliability

Monitoring battery health is crucial for predicting the lifespan of batteries in electric vehicles. Real time voltage, current, and operating temperature are the key factors influencing longevity. An efficient data acquisition (DAQ) system becomes essential for accurately monitoring real-time battery operating parameters. For commercially viable three-wheeled electric vehicles, a cost-effective yet highly accurate DAQ is imperative. Current commercially available lead-acid three-wheeled electric vehicles lack a dedicated DAQ for monitoring battery health. In this investigation, we address this gap by developing a low-cost, multi-channel DAQ tailored for such electric vehicles, integrated with a cloud server for subsequent data analysis and Remaining Useful Life (RUL) prediction. The implemented prototype excels in measuring individual battery terminal voltage and battery pack voltage under operating conditions, employing the node voltage subtraction method. It also features real-time detection of high current discharges up to 30A using a Hall effect-based current sensor. A custom temperature sensor probe was designed to detect battery electrolyte temperatures up to 60 °C, with multiple probes operating in multiplexed mode to reduce circuit complexity. The system is proficient in storing recorded data locally and in the cloud. A mobile device platform has been developed and interfaced with the system for real-time monitoring of battery parameters. After testing multiple prototypes in running vehicles for over 1000h without failure, the system demonstrated a high accuracy level of 98.6 %. Multiple prototypes were designed to validate the measured accuracy.

An automatic robot for ultrasonic partial discharge detection of gas-insulated switchgear

Purpose Gas-insulated switchgear (GIS) stands as a pivotal component in power systems, susceptible to partial discharge occurrences. Nevertheless, manual inspection proves labor-intensive, exhibits a low defect detection rate. Conventional inspection robots face limitations, unable to perform live line measurements or adapt effectively to diverse environmental conditions. This paper aims to introduce a novel solution: the GIS ultrasonic partial discharge detection robot (GBOT), designed to assume the role of substation personnel in inspection tasks. Design/methodology/approach GBOT is a mobile manipulator system divided into three subsystems: autonomous location and navigation, vision-guided and force-controlled manipulator and data detection and analysis. These subsystems collaborate, incorporating simultaneous localization and mapping, path planning, target recognition and signal processing, admittance control. This paper also introduces a path planning method designed to adapt to the substation environment. In addition, a flexible end effector is designed for full contact between the probe and the device. Findings The robot fulfills the requirements for substation GIS inspections. It can conduct efficient and low-cost path planning with narrow passages in the constructed substation map, realizes a sufficiently stable detection contact and perform high defect detection rate. Practical implications The robot mitigates the labor intensity of grid maintenance personnel, enhances inspection efficiency and safety and advances the intelligence and digitization of power equipment maintenance and monitoring. This research also provides valuable insights for the broader application of mobile manipulators in diverse fields. Originality/value The robot is a mobile manipulator system used in GIS detection, offering a viable alternative to grid personnel for equipment inspections. Comparing with the previous robotic systems, this system can work in live electrical detection, demonstrating robust environmental adaptability and superior efficiency.

Centimeter-level 2D ultra-small interdigital sensor for ESD detection of spacecraft

The spacecraft in geosynchronous orbit interacts with the space plasma environment, which is prone to electrostatic discharge and affects the safe operation of the spacecraft. In order to realize simple, accurate and reliable electrostatic discharge detection in space environment, based on the energy spectrum characteristics of spacecraft electrostatic discharge electromagnetic radiation electromagnetic pulse and the electromagnetic pulse sensing principle of interdigital electrode, a centimeter-level two-dimensional planar ultra-small interdigital sensor with a size of only 4 cm × 3 cm × 0.01 cm is developed. The impedance matching circuit is established to greatly improve the transmission efficiency of electromagnetic pulse, which is 59.34 % higher than that before the matching circuit is added. Finally, a spacecraft vacuum electrostatic discharge simulation experiment platform is built to test the frequency response characteristics of the sensor. The results show that the frequency response characteristics of the electromagnetic pulse signal perceived by the sensor developed in this paper are consistent with the main frequency characteristics of the spacecraft electrostatic discharge radiation electromagnetic pulse energy spectrum, which can be used for spacecraft electrostatic discharge detection in space environment.

Research and Simulation of Partial Discharge Phenomenon Sensing Technology for Substation Inspection Robots

Abstract This paper thoroughly investigates the phenomenon of partial discharge in substations and proposes a set of simulation experimental apparatus to replicate various types of partial discharge defects in switchgear. By modifying the switchgear, incorporating models for different types of partial discharge, and equipping various standard partial discharge sensors, accurate simulations of phenomena such as point discharge, suspended discharge, gap discharge, and particle discharge are achieved. Through this system, the effectiveness of different detection methods, such as ultrasonic, high-frequency, and transient voltage methods, is validated for various defect models. Furthermore, utilizing an indoor inspection robot guide rail, the paper simulates the environment of partial discharge in substations, enabling the validation of the robot’s partial discharge detection performance and typical patterns under different conditions. This work is dedicated to providing situational awareness technology for partial discharge phenomenon detection for substation inspection robots, so that the inspection robot can detect various abnormal partial discharge phenomena in transformers or other substation equipment through micro sensors to promote the stable operation of the substation.