DC Partial Discharge Research Articles

Computer Vision for DC Partial Discharge Diagnostics in Traction Battery Systems

The tendency towards thin insulation layers in traction battery systems presents new challenges regarding insulation quality and service life. Phase-resolved DC partial discharge diagnostics can help to identify defects. Furthermore, different root causes are characterized by different patterns. However, to industrialize the procedure, there is the need for an automatic pattern recognition system. This paper shows how methods from computer vision can be applied to DC partial discharge diagnostics. The derived system is self-learning, needs no tedious manual calibration, and can identify defects within a matter of seconds. Thus, the combination of computer vision and phase-resolved DC partial discharge diagnostics provides an industrializable system for detecting insulation faults and identifying their root causes.

Configuration of instrumentation wire extraction for ITER magnet feeder system

For the quench detection, the high voltage (HV) instrumentation wires should be led out from the electrical insulation layers of ITER magnet feeders. Based on the latest design, at least four HV instrumentation wires need to penetrate from the busbar joint, which break the integrity of the intact insulation layer. A perforated crack may generate along the instrumentation wire in the insulation layer. At some specific ambient pressure and operational voltage, the crack will cause Paschen discharge inevitably, which is the most serious risk threatening the safe and steady operation of the feeders. There are over 200 busbar joints with the instrumentation wires penetration need to be manufactured on the ITER site, and a reliable and durable insulation structure and technology need to be developed for the extracting HV instrumentation wires of ITER magnet feeder. In this paper, the HV instrumentation wire extraction and the specimen design are presented. The electrical properties of the specimen, including the DC hi-pot test, Paschen test and Partial Discharge (PD) test, are reported and discussed.

Finite Element Analysis of DC Partial Discharges in Failure Types of Thin Insulation Layers in Traction Battery Systems

Finite Element Analysis of DC Partial Discharges in Failure Types of Thin Insulation Layers in Traction Battery Systems

Hybrid numerical simulation of the generation and distribution characteristics of SF6 heavy particles under different DC PD energies

To investigate the decomposition law of SF6 under negative direct current partial discharge (PD) at different energies, a SF6 PD hybrid numerical model based on fluid dynamics and plasma chemical reaction models, in which 14 particle species and 24 chemical reactions are considered, is proposed. The effectiveness of the proposed model is validated with the current pulse waveform and the V–I discharge curve obtained by experiment. The influence of discharge energy on SF6 PD characteristic quantities and SF6 decomposition products is investigated with simulation and experiment. The results show that most of the discharge area of SF6 is neutral, and the cation clouds only exist in the ionosphere (4.79–5 mm). With the increase in applied voltage, the electric field intensity of the needle plate gap does not increase completely and even decreases in some areas. Moreover, different from the traditional opinion, the generation of SO2F2 under PD is mainly generated by the hydrolysis reaction of SOF4, which is formed by [SF5], [SF4], and [OH], [O]. The reaction path of [SF2] with O2 is not important. Thus, c(SO2F2)/c(SOF2 + SO2) can be used as the energy characteristic component ratio because of its ability to represent the low-fluorine sulfide ratio n([SF5])/n([SF4]).

Comparison of AC and DC Partial Discharge by Monitoring of Vacuum Degree in a Distribution Class Vacuum Interrupter

The vacuum degree of Vacuum Interrupter ( <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> ) will deteriorate because of gases emitted by arc heat, and leakage through the joints. It is important to select an appropriate maintenance criteria for a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> by monitoring the vacuum degree. However, monitoring of a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> installed inside a solid insulation apparatus in a transmission system is impossible. Only studies on the insulation deterioration because of the decrease of the vacuum degree have been conducted, and important factors such as the maintenance criteria have not been investigated. Moreover, most studies have only been performed using AC voltage, not DC voltage. Therefore, in this paper, AC and DC partial discharge were compared by monitoring of the vacuum degree in a distribution class <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> to propose maintenance criteria according to voltage type. Through this, for efficient maintenance, maintenance criteria of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> for each voltage type are presented. The vacuum degree at which the dielectric strength rapidly decreases was confirmed according to voltage type. The coupling capacitor was installed directly on the floating shield of <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">VI</i> . Moreover, to measure the pattern and apparent charge of partial discharge according to the vacuum degree, phase resolved partial discharge (PRPD) for AC partial discharge and pulse sequence analysis (PSA) for DC partial discharge were adopted. It is found that maintenance should be performed when the vacuum degree is in the range of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$10^{-3} \boldsymbol {\sim } 10^{-2}$ </tex-math></inline-formula> torr regardless of the voltage type. Moreover, the usefulness of the technology is higher at AC voltage than at DC voltage.

Application of multi‐source information fusion based on D‐S evidence theory in insulation defect identification of DC XLPE cable

During the installation and laying process of cross-linked polyethylene (XLPE) cable, it is difficult to avoid the influence of external harsh environment, which leads to insulation deterioration and failure. The partial discharge and leakage current play an important role in formation mechanism, characteristic law and identification technology of DC XLPE cable defect. However, the defect identification accuracy of single source data is limited. The correlation between partial discharge and leakage current of typical defect is analysed here. Based on the shape characteristic parameters of typical DC partial discharge spectrum and the energy characteristic parameters of leakage current wavelet decomposition node, a weighted Dempster–Shafer (D-S) evidence theory is used to fuse and identify different information sources. The results show that the classification rate of the proposed method can reach 88%, which can effectively identify insulation defects.

Time Transition of Activated Conductivity Distribution in Air and Charge Accumulation in Air-solid Composite Insulation Systems under DC Partial Discharge

In this study, charge behavior in gas and on solid insulator due to DC partial discharge (DC-PD) is investigated by using needle-air-PMMA-plane electrode configuration under negative DC voltage. The time evolution of surface potential distributions on solid insulator is measured and compared with the simulation results taking into account the charge behavior in air, such as generation, recombination and motion of positive and negative ions. Especially, a novel transport equation model considering the source of charge carrier due to DC-PD is built. Consequently, in air-solid composite insulation systems under negative DC voltage application, the time transition of activated conductivity distribution in air by DC-PD contributing to the charge accumulation on PMMA surface is evaluated quantitatively. The activated conductivity in air is distributed in the range from 1.0 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-9</sup> to 3.0 × 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">-9</sup> S/m under DC-PD, which is approximately 10 <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">10</sup> times higher than that in non-activated conductivity space by natural ionization. In addition, the activated conductivity distribution in air dominantly contributes to the charge accumulation process from initial state (DC-on) to DC steady-state. The activated conductivity is the crucial parameter for DC electrical insulation in case that charge behavior is activated in gas medium under DC-PD.

Experimental Research on Partial Discharge of High Voltage Direct Current

Under the action of long-term high-voltage DC voltage, the insulation of electrical equipment is inevitably aging, and it is necessary to monitor its insulation status. Since insulation defects of different nature damage electrical equipment to different degrees, the identification of defects is very important. To this end, a DC partial discharge detection system based on AC partial discharge detection equipment was built, focusing on key issues such as the choice of test power supply, the choice of test frequency, and the choice of pressurization methods. The successful completion of the on-site partial discharge test of UHV converter transformers has provided valuable practical experience for the future on-site partial discharge tests of UHV converter transformers, and provided important tests for the establishment of on-site partial discharge test standards for UHV converter transformers. in accordance with.

A simple method to measure DC electric field with space charges

DC partial discharge in gas generates many space charges. Because these charges move with the electric field, change the electric field and accumulate on the surface of sensors, it is difficult to measure the electric field in the space. Based on the relationship between the charge accumulated on the surface of the dielectric in the ion flow field and the applied electric field, this paper proposes a simple method to measure the electric field with space charges using a suspended dielectric as a sensor. By testing the charge accumulated on the suspended dielectric, the applied electric field can be determined. The method is verified and the factors that affect the measurement result are investigated by experiments in a predictable and controllable ion flow field. The results show that when saturation of the accumulation of the space charge on the suspended disc is reached, the total accumulated surface charges are linear with the normal component of the external applied electric field. The ion flow density just affects the speed of the charge accumulation and does not affect the amount of saturated surface charges. The relationship coefficient between the surface charge and the electric field can be obtained through calibration. Unlike other methods that regard the accumulation of charge on the sensor as the key interference, the method in this article just uses the accumulation characteristics of the charge on the sensor, which is simple and feasible.

Optimized extraction of PD fingerprints for HVDC XLPE cable considering voltage influence

Fingerprint parameters of partial discharge (PD) are frequently used for insulation defect identification. In this paper, three kinds of defects, such as inner semiconducting layer breakage, internal cavity and insulating surface scratch defects, common in XLPE cables were fabricated. Then their PD characteristics were tested under different voltage levels and polarities. It was found that DC PD characteristics exhibited distinct for each defect, and varied with voltage amplitude and polarity. Then the PD fingerprint parameters of three defects were calculated by a conventional extraction method. The obtained fingerprints were susceptible to voltage level and polarity. This influence would be not beneficial to the defect identification. To eliminate it, GA-BPNN was applied to re-extract the fingerprint parameters and meanwhile classify defect type. Furthermore, the defect identification rate was compared in the two cases with different extraction methods. For the conventional method, the identification rate was about 97% without considering voltage level and polarity, and it would drop to about 87% if influences of voltage amplitude and polarity were introduced. On the contrary, the identification rate always remained above 97% for the optimized method.

Design of Transmission Line and Electromagnetic Field Sensors for DC Partial Discharge Analysis

In this paper, a testbed is designed and constructed for the investigation of DC PD pulses. The testbed is equipped with a 50 Ω transmission line (TL) that terminate to an oscilloscope for measuring the charge displacement current generated by PD pulses. Besides the oscilloscope measurements, two types of electromagnetic field sensors (D-dot and B-dot) were developed to capture the EM fields of the PD pulses propagating through the TL. The main goal of this paper is to investigate the DC PD pulses through the EM fields and the corresponding discharge current pulses that are considered as calibrating signals for the developed D-dot and B-dot sensors. The results of DC cavity discharge measured by the constructed testbed and the EM field sensors demonstrate close agreement with the reference PD pulses measured via oscilloscope.

Study of DC partial discharge on dielectric surfaces: Mechanism, patterns and similarities to AC

This paper presents the investigation on surface discharge behavior of various dielectric samples under DC. It sequentially develops the knowledge base for the study and analysis of the partial discharge (PD) defect with the goal of PD defect identification under DC. In order to facilitate this, the material properties of the dielectric are measured. Finite Element (FEM) simulation is used to obtain the preliminary estimates of the electric field and dielectric properties that concern partial discharge behavior. The DC-PD tests performed on the surface dielectric samples demonstrate a highly plausible behavior based on simulation results and other literature. It also displays a great degree of similarity towards the AC surface discharge behavior. The paper concludes by presenting novel partial discharge fingerprints for the surface PD defect that will aid in defect identification under HVDC.

Classification and comparison of AC and DC partial discharges by pulse waveform analysis

This study aims to classify and compare AC and DC partial discharge (PD) based on PD pulse waveform analysis. To achieve this goal, we designed a testbed that enables precise measurements of individual PD pulses. The testbed is used for collecting data from four different types of PDs including cavity, surface, corona, and floating potential discharges generated by individual PD source samples. All samples were examined under AC, positive DC, and negative DC electrical stresses, through which we captured thousands of PD pulses. We classify the waveforms of each PD type into representative groups associated to their discharge mechanisms. The statistical data of the measured pulses are utilized to identify the differences between AC and DC PDs while the clustered patterns of PD amplitude versus their temporal characteristics serve as a means to classify the types of PDs under AC and DC electrical stresses.

Pattern Recognition of DC Partial Discharge on XLPE Cable Based on ADAM-DBN

Pattern recognition of DC partial discharge (PD) receives plenty of attention and recent researches mainly focus on the static characteristics of PD signals. In order to improve the recognition accuracy of DC cable and extract information from PD waveforms, a modified deep belief network (DBN) supervised fine-tuned by the adaptive moment estimation (ADAM) algorithm is proposed to recognize the four typical insulation defects of DC cable according to the PD pulse waveforms. Moreover, the effect of the training sample set size on recognition accuracy is analyzed. Compared with naive Bayes (NB), K-nearest neighbor (KNN), support vector machine (SVM), and back propagation neural networks (BPNN), the ADAM-DBN method has higher accuracy on four different defect types due to the excellent ability in terms of the feature extraction of PD pulse waveforms. Moreover, the increase of training sample set size would lead to the increase of recognition accuracy within a certain range.

Partial discharge identification of DC cross-linked polyethylene cables based on GA-BP algorithm

In this paper, XLPE DC cables were used to construct three kinds of defects of inner semiconducting layer breakage, internal air gap defect, and insulating surface scratching defect. And their DC PD characteristics were tested under different voltage amplitudes and polarities. The study found that these three defects exhibit different DC partial discharge characteristics and change with voltage amplitude and polarity. Subsequently, two methods were used to identify the type of defect. The recognition rate of the traditional defect type identification method reached 97.1%, and the recognition rate of the optimized defect type identification method based on GA-BP was increased to 98.9%, which solved the problem that the traditional method was affected by the voltage level and polarity, and improved the recognition rate.

Decomposing Mechanism of SF6 under Positive DC Partial Discharge in the Presence of Trace H2O.

The influence of H2O on SF6 decomposition characteristics under positive DC partial discharge (PD) is significant. To evaluate PD fault severity in DC SF6-insulated equipment using the production characteristics of SF6 decomposition components, the corresponding relationship and mathematical expression between the production of SF6 decomposition components and the H2O content should be identified and achieved. Thus, SF6 decomposition experiments under positive DC PD are performed to reflect the influence of H2O on SF6 decomposition components. Results show that the total discharge quantity and the discharge repetition rate averaged for 1 s decrease slightly when the H2O content increases from 0 to 970 ppmv and then increase when the H2O content increases from 970 to 5120 ppmv. The effective production rates of SO2F2, SOF2, and SO2 increase with the H2O content, whereas that of SOF4 decreases. Finally, the corresponding relationship and mathematical expression between the characteristic ratio (c(SO2F2) + c(SOF4))/(c(SOF2) + c(SO2)) of components and the H2O content have been achieved, which can afford references for PD fault diagnosis in DC SF6 gas-insulated equipment.

An Investigation of Negative DC Partial Discharge Decomposition of SF6 Under Different Metal Materials

For determine the influence of metal materials on the negative DC partial discharge (PD) decomposition of SF6 and put forward for the method of SF6 DC gas insulated equipment condition monitoring and fault diagnosis based on decomposed component analysis, this study investigates the decomposition characteristics of SF6 under three common metal materials (Al, 304 stainless steel, and Cu) in gas insulated equipment by using SO2, SOF2, and SO2F2 as characteristic component gases under negative DC PD. Results show that the types of metal materials have substantial effects on the negative DC PD decomposition characteristics of SF6. The simulation results based on the first principles confirm the phenomena observed in the experiment: the amounts and generation rates of SOF2 were the most abundant, followed by SO2F2 and SO2. The calculation based on free electron cloud and finite depth particle in a box model shows that the electron current density of field emission has the relationship with metal work function $\gamma $ and electrical field $E$ : $j(0{,E})\approx (BE^{2}{/}\gamma)\exp (-D\gamma ^{3 / 2}{/}E)$ , where B and D is constants. The electron current density by field emission is inversely proportional to the work function of metal with constant electric field. The metal material with smaller work function means the greater electron current density, further promoting the decomposition of SF6 to generate more characteristic component gases. The results can well explain the phenomena observed in the experiment, that is, the decomposition of SF6 is the most serious under Al electrode with lowest work function, followed by 304 stainless steel, and Cu.

Partial discharge in a cylindrical cavity in XLPE under DC voltage: Simulation and experiment

A simulation model is established to investigate the influence of applied voltage and temperature on DC partial discharge characteristics of a XLPE sandwich internal cavity sample. The field and temperature dependent conductivity of XLPE is considered, and its influence on the electrical field distribution is computed. The calculation results show that the voltage across the cavity changes with time under constant DC, and the steady cavity voltage is 32.7% higher than that just after voltage application. When calculating the cavity voltage at 70 °C, the error is over 500%, if the characteristics of the conductivity of XLPE is not considered. Residual cavity voltage with the opposite polarity after decreasing voltage is discovered by computation, and discharges with the polarity opposite to the applied voltage are detected by experiment. By simulation, the partial discharge repetition rate increases by two orders of magnitude when the temperature increases from 23 to 70 °C. The repetition rate measured by experiment increased by over one order of magnitude. The change of repetition rate is attributed to the reduction of cavity voltage recovery time influenced by the temperature dependence of the XLPE conductivity.

Evaluation of insulation condition detection methods applied to DC XLPE cable

At present, with the construction and operation of new energy grid-connected and offshore wind power projects in China, DC XLPE cables have been widely used. Under the operating conditions, there are mainly two feasible technical means of DC cable insulation state detection, partial discharge (PD) and leakage current. However, it is still a blank in the world to apply it to the validity and condition of DC cable operation detection. Based on the semi-conductive layer residual defect model of XLPE cable, the characteristics of PD and leakage current under different voltage values and polarities are studied by setting up an experimental platform for detecting PD and leakage current of DC XLPE cable. The results show that both PD and leakage current can detect cable defects, among which PD is easier to detect cable defects under negative polar voltage and higher voltage. Leakage current characteristics change obviously under positive polar voltage. When applied voltage is lower than rated voltage, obvious changes can also be measured. In this paper, the feasibility of live detection of leakage current of DC cables under operating conditions is proposed, which is of great value to improve the detection capability of DC XLPE cables.

Editorial: Polymeric materials for HVDC insulation

High voltage direct current (HVDC) power transmission plays a key role in the global power grid today and in the future, particularly for high-capacity, long-distance, and regional power grid interconnections. With the high voltage level of the HVDC power system, the most important concern is the operating safety of polymeric materials as well as solutions for HVDC insulation. During polarity reversal and overvoltages on the HVDC system, polymeric insulation can breakdown in cables, cable accessories, apparatus bushings and gas insulated switchgears. The issues regarding polymeric materials which includes the DC conductivity, charge transportation and partial discharge, have become key problems in restraining the development of the HVDC power system.