Toshihiro Hoshino

Separation of corona using wavelet packet transform and neural network for detection of partial discharge in gas-insulated substations

It is essential to detect partial discharge (PD) as a symptom of insulation breakdown in gas-insulated substations (GIS). However, the accuracy of such measurement is often degraded due to the existence of noise in the signal. In this paper, a method using wavelet packet transform and neural network is proposed to separate the PD pulses from corona in air, which enables more accurate detection of insulation breakdown of GIS.

Source classification of partial discharge for gas insulated substation using waveshape pattern recognition

Frequency-domain analysis of ultra-high frequency (UHF) signals for source identification of partial discharge (PD) occurring in SF/sub 6/ inside gas-insulated substation (GIS) has been widely covered in literature. In this, Fast Fourier Transform and Discrete Wavelet Transform based techniques have been extensively applied to derive classifying features from transformed patterns. On the other hand, it appears feasible to develop a time-domain classifier, which derives features directly from the original waveshape. The time-domain classifier is conceptually simple, and requires potentially less computing resources and simpler algorithmic interface with other intelligent techniques due to elimination of frequency-domain transformation. A novel classifier to extract features directly from time-domain waveforms is proposed for classifying SF/sub 6/ PD from air corona and among the three types of SF/sub 6/ PD, regardless of changes in PD locations and measurement conditions. Three sets of classifying features are proposed. Encouraging results have been achieved with comprehensive experimental data, which verifies and proves the usefulness and feasibility of the time-domain classifier.

Influence of disconnecting part on propagation properties of PD-induced electromagnetic wave in model gis

Partial discharge (PD) detection using a UHF (ultra high frequency) band signal is a well known advanced insulation diagnosis method in gas insulated switchgear (GIS), and has been actively studied. Detailed investigation of electromagnetic (EM) wave propagation inside the GIS tank is required for significant improvement of detecting PD signal by UHF method. When practically applying the UHF method to GIS insulation diagnostics, it is necessary to examine the effects of GIS components such as circuit breakers, isolators and disconnectors on EM wave propagation properties. In this paper, attention is paid to the effects of a disconnecting part of a high voltage (HV) conductor like a circuit breaker or a disconnector in GIS. To examine the effects of disconnecting part, the gap length of the disconnecting part was set as parameter, and waveforms and frequency spectra of the propagation PD-induced EM wave were measured with UHF sensors. For the purpose of discussing the effects of the disconnecting part theoretically, a finite difference time domain (FD-TD) simulation was also carried out. The experimental results show that the PD-induced wave could propagate through the disconnecting part with higher frequency components over the cutoff frequency components of TE11 mode for disconnecting part, i.e. cylindrical shape formed by GIS tank without HV conductor. The propagation of the lower frequency components below the TE11 mode depended on the gap length of the disconnecting part.

Influence of insulating spacer type on propagation properties of PD-induced electromagnetic wave in GIS

The authors have studied insulation diagnostic technique of gas-insulated switchgear (GIS) by UHF (ultra high frequency) method. In this paper, the authors investigated the effect of insulating spacer type (metal flange type (MT) or bush type (BT)) on the propagation properties of partial discharge (PD) induced electromagnetic (EM) wave in GIS with UHF method. Low pass filtering process for cutoff frequency of TE11 mode was used to eliminate the higher order mode above TE11 of detected signal. Results revealed that reduction of the magnitude of measured EM waveform by passing through spacers. The reduction was found to be not due to attenuation by the spacer, but mainly due to superimposition of reduced TE mode and TEM mode. Furthermore, the authors investigated relationship between the propagating EM wave in GIS and a leaked EM wave from a BT spacer. This result suggests that insulation diagnosis by a sensor placed outside of GIS could be carried out to detect the leaked EM wave through a BT spacer.

Propagation properties of PD-induced electromagnetic wave in 66 kV GIS model tank with L branch structure

The authors investigated the effects of a corner that affects the transmission characteristics of a partial discharge (PD) induced electromagnetic (EM) wave, using a gas insulated switchgear (GIS) model with an L-shaped branch. The authors also experimentally investigated the change of amplitude of PD-induced EM wave around the passage of the Lshaped part using a disc-type ultra-high frequency (UHF) sensor located in a 66 kV model GIS. In addition, the effects of transverse electric (TE) mode were examined using a low pass filtering process, and a finite difference time domain (FD-TD) simulation with a simple model was performed. Subsequently, the effects of the L-shaped part that affected the EM wave transmission characteristic were discussed based on a comparison of the results with the analysis. It was found that the effects of the L-shaped part that affected the transmission characteristics of the PD-induced the EM wave did not affect the transverse electromagnetic (TEM) wave mode component, and reflection was generated with a higher frequency component above TE11 mode. It was also shown that the amplitude of the propagating EM wave changed significantly around the L-shaped part due to the superimposition of the reflected TE mode on the latter part of the incoming EM wave.

Comparison of sensitivity between UHF method and IEC 60270 for onsite calibration in various GIS

It is more important to obtain the sensitivity line, which is defined as a correlation between the ultra-high-frequency (UHF) signal (mV) and apparent charge (pC), for onsite calibration regarding partial discharge (PD). Synchronous measurement of the UHF method and IEC 60270 enables us to obtain the sensitivity line and to estimate the apparent charge in on-site substations. The sensitivity line of 3-phase gas-insulated switchgear (GIS) is compared to that of single-phase GIS, and differences in defects are also discussed. The fundamental data in this paper contribute to on-site insulation diagnosis of GIS

Simulation of Propagating Electromagnetic Wave Due to Partial Discharge in GIS Using FDTD

It takes a lot of trouble to measure and calibrate the sensitivity of ultra-high-frequency couplers which are set in various gas-insulated switchgears (GIS). This paper is related to the approach of calculating the attenuation of the electromagnetic (EM) wave due to the partial discharge propagating through insulators constituting a GIS. The use of a simulator based on the finite-difference time-domain method for analyzing the three-dimensional transient EM fields shows the degree of precision of the modeling descriptions employed and their ability to confirm experimental results.

Study of partial discharge radiated electromagnetic wave propagation characteristics in an actual 154 kV model GIS

A method to detect partial discharge (PD) is considered effective for gas insulated switchgear (GIS) insulation diagnostics. In this paper, for a 154 kV model GIS, the influence of the enclosure diameter on PD propagation characteristics was initially investigated using model GIS by varying the enclosure size. Secondly, an experiment was conducted for metallic particles placed in different locations as a source of PD (particle adhering to the center conductor or free metallic particle). Thirdly, an actual PD experiment was conducted in an Lshaped form to study the influence of GIS shape on PD-radiated electromagnetic (EM) waves. In addition, in order to study the experimental results in detail, a simulation using the FD-TD method was conducted for comparison with these experimental results. Following the experiment, the larger the enclosure diameter, the greater the measured amplitude of PD having occurred around the center conductor. In addition, it emerged that, where the particle was close to the center conductor, EM waves propagating due to PD were less likely to be damped compared with the case where particles were placed on the bottom of the enclosure. Furthermore, following the investigation of the propagation characteristics in the L-shaped form, it emerged that, while a high frequency component of frequency exceeding that of the TEM wave component was reflected at the L-shaped part, the TEM wave component passed through almost unaffected by the latter. The propagation of the TEM mode component could be also found via FD-TD analysis.

A unique sensitivity verification combined with real-time partial-discharge identification method

The sensitivity verification with the ultrahigh-frequency (UHF) method will be discussed in this paper. The sensitivity line with the UHF method based on the conventional (IEC 60270) one is different in the radial position of the defects inside gas-insulated substation (GIS), because transverse-electromagentic mode can be more weakened while transverse electric (TE)/transverse magnetic (TM) modes can be more enhanced as the defect comes close to the enclosure than the conductor. The result makes it difficult to detect the free particle with IEC 60270. A unique sensitivity verification, which is combined with real-time PD identification method, first enables us to identify the defects and thereafter to calibrate the apparent charge of each specified defect according to the sensitivity line.

Real-time PD identification in diagnosis of GIS using symmetric and asymmetric UHF sensors

A new practical method for real-time insulation diagnosis of GIS is introduced: a conical coupler is used as an asymmetric UHF sensor while a loop electrode embedded in barrier insulator as a symmetric one. This method can immediately identify the kinds of defects in GIS and discriminate inside partial discharge (PD) from noises by taking single PD pulse. This is because the output ratio of those UHF sensors shows the instinct difference in the kinds of defects due to dissimilar electromagnetic field distributions in GIS. The method will lead to sequential risk assessment based on each identified PD pulse.