UHF PARTIAL DISCHARGE DIAGNOSIS OF PLUG-IN CABLE TERMINATIONS
Denis DENISSOV, University of Stuttgart, (Germany), denis.denissov@ieh.uni-stuttgart.de Ruben GRUND, PFISTERER Kontaktsysteme GmbH, (Germany), und@pfisterer.de Thomas KLEIN, PFISTERER Kontaktsysteme GmbH, (Germany), thomas.klein@pfisterer.de Wolfgang KÖHLER, University of Stuttgart, (Germany), wolfgang.koehler@ieh.uni-stuttgart.de Stefan TENBOHLEN, University of Stuttgart, (Germany), bohlen@ieh.uni-stuttgart.de
ABSTRACT
The experiences in on-line application of the ultrawide band partial discharge (PD) detection in high voltage plug-in cable terminations are prented. Issues of nsitivity of capacitive and inductive field couplers (nsors) are discusd. A ries of comparative tests was done using an artificial defect to establish a link between peak voltage, as well as the energy of UHF puls and the corresponding apparent charge, provided with a conventional IEC 60270 method. Examples of the UHF pha resolved PD measurements and measurements with a power detector are given. Experiences from a field testing are reported as well. The diagnostic system can be applied as a quality check on the just asmbled terminations, as well as for the purpos of the condition asssment of insulation integrity after years in rvice.
KEYWORDS
Partial discharge (PD), UHF, on-line diagnostics, termination, nsors.
INTRODUCTION
A failure of a high voltage power cable caus a rvice interruption, costly location, repairs and loss of revenues. Utility experience shows that poor termination and jointing is a major cau of cable failure. This is due to the fact that, in contrast to the cable itlf, the components have more compl
ex structure, sometimes even with veral dielectrics, and incread field gradients. But moreover they are asmbled and installed under on-site conditions and thus expod to the higher risk of defects and contaminations.
Modern plug-in cable connectors (terminations) for GIS and transformers are made from silicone rubber. The electrical life span of this high polymeric material normally exceeds 40 years, but only in abnce of PD activity that inevitably caus material degradation. Several IEC standards, e.g. IEC 60840 [1], prescribe routine tests on the prefabricated components of HV cable accessories to be carried out by manufacturers. Unfortunately, there are no standards for testing a complete accessory yet. But an improper asmbly done under on-site conditions can strongly affect the long-term performance of the completed accessory. Therefore, to make sure that the asmbly was done immaculately, a quality check is often desired by utilities.
Another application field of the prented test technique is a condition monitoring of cable accessories in order to predict failures before they occur. Tho accessories that are about
to fail can then be replaced, thereby reducing the risk of cable system failures and improving the overall quality of power supply.
Partial discharge measurement is a well established criterion for the condition asssment and quality control of the high voltage electrical insulation. PD, originated from a microdefect, incepts periodically according to ac cycle of the operation voltage and gradually degrades and erodes the polymeric material, eventually leading to breakdown. To detect such a PD activity under conditions of on-site on-line testing, the ultrawide band PD (UHF PD) diagnosis principle can be deployed. This method is bad on nsing the electromagnetic emissions from discharge sites in the insulation. The coupling nsors should be placed possibly clo to the test object and effectively screened against outside interferences.
Although there are veral well known off-line test techniques, which are successfully applied to diagno long power cables including their accessories, they all need load flow re-dispatching and a parate voltage source to energize the cable line apart from the network. The on-line test approach overcomes the difficulties allowing nsitive measurement on the terminations, while the cable is in normal operation. This contribution discuss the experiences in on-line UHF PD diagnostics of high voltage cable terminations.c盘满了怎么扩容
PRINCIPLE OF DIAGNOSTICS
整改措施落实情况
The occurrence of partial discharges in electrical insulation is always associated with the emission of electromagnetic puls. A typical PD pul has a ri time of less than 1 ns and a pul width of veral ns, implying in frequency-domain a bandwidth of veral GHz. The electromagnetic emissions propagate in all directions from the PD source. Different materials impo different attenuation rates to the travelling waves. In general, the attenuation of the PD puls is a function of frequency [2]-[3]. The higher the frequency components will be attenuated rapidly when they travel along the cable. Therefore, detecting PD in the UHF band (300-3000 MHz) has the advantage of the distance lectivity of only veral meters. This can be perfectly ud for the diagnosis of the concentrated equipment such as transformers, GIS, machines and cable accessories. The distributed equipment, e.g. cables, can be effectively diagnod in HF and VHF bands.
Fig. 1 demonstrates the principle of UHF diagnosis of the plug-in cable connectors. A portable metallic sleeve is clamped around the cable immediately behind the
connector and fulfils two functions: firstly, as a housing for field couplers (antennas) and condly, as a grounded screen against the disturbances from outside.
Figure 1: Object and principle of the UHF diagnostics The nsors are mounted inside the sleeve and terminated with BNC jacks. The capacitive nsor reprents a copper disc with the diameter of 2.5 cm, soldered to the copper pin in the middle. The inductive nsor is a two-winding coil made from an insulated wire. One end of the coil is grounded; the other end is connected via BNC jack to a measuring coaxial cable.
EXPERIMENTAL ARRANGEMENT
Fig. 2 prents an experimental arrangement built in a laboratory. It includes commercial 550 kV GIS (1), HV test transformer (2), power cable (4) connected to the GIS via connector (5) and terminated at SF 6-filled test joint (7) using the connector (6) with an artificial PD defect. The GIS is equipped wi
th a coupling capacitor and a conventional IEC 60270 PD measurement system. An UHF PD detection circuit, shown also in Fig. 6, contains the portable screening sleeve (8) with capacitive and inductive nsors inside, coaxial cables, high pass filter with the cut-off frequency of 200 MHz, a 40 dB preamplifier (9) and an 1 GHz high speed digital oscilloscope (10).
1
2
3
5
4
6
7
8
9
10
C orona L ocation 1
C orona L ocation 2
Figure 2: Laboratory t-up
Power supply of the measurement circuit is realized via insulating transformer. An artificial insulation defect was built-in to initiate a discharge at interfaces between materials. This kind of fault is very typical for defects in cable accessories. For that purpo a plastic lath covered with some silicon grea was introduced along the boundary between the cable insulation and a stress cone.
The intensity of PD was influenced by changing the penetration depth of the lath.
SIMULATION OF EIGENMODE FREQUENCIES
Since the prented UHF PD diagnostic concept implies the measurement of radiated puls outsid
e the termination housing, it is of high interest to figure out what own resonance frequencies (eigenmodes) this body has. Becau once excited, the frequencies will be resonated according to the well known physical phenomenon. Fig. 3 demonstrates the results of eigenmode simulation on a model prented in Fig. 1. Calculation was done considering loss in materials. As a result, Q-factors, a ratio between the total signal power and power loss in materials of the model, were evaluated.
I n t e n s i t y (Q -F a c t o r )
Frequency (GHz)
Figure 3: Eigenmode frequencies (simulated) The eigenmode frequencies are almost uniformly distributed over the range of interest. The range of 550-850 MHz is denly occupied with eigenmode frequencies, which agrees perfectly with the working range of the nsors and hardware ud.
LABORATORY MEASUREMENTS Sensor types and functionality
In general, the field emissions caud by PD puls can be coupled in a capacitive or inductive manner.
Capacitive couplers (nsors) u the electric component of the transient field. The capacitive coupling generally depends on the dimensions of the nsor, distance to the test object and the electric field strength of the emitted waves. In the vicinity of the PD source, the discharge puls are carried by a small portion of the neighbouring materials and they need some space until they will distribute uniformly along the circumference of the accessory or cable [2]. Therefore to increa the spatial detection nsitivity of capacitive nsing one needs to put veral nsors along the circumference or simply move the single nsor along the circle.
Inductive nsors, also called high frequency current transformers (HFCT), are usually made in form of coil or toroid. They couple the magnetic component of the field. Thus, the voltage induced in the coil is proportional to the
Capacitive UHF nsor
Portable screening sleeve PD defect
Inductive UHF nsor
Connector
Bushing
BNC jack
A m p l i t u d e (m V s )
Frequency (MHz)
Figure 4: Frequency spectrum of a PD pul (15 pC)
captured with the capacitive nsor in time-domain
A m p l i t u d e (m V s )
Frequency (MHz)
Figure 5: Frequency spectrum of a PD pul (15 pC) captured with the inductive nsor in time-domain rate of change of the current passing through the coil. The magnetic coupling depends on the number of turns, distance to the test object, the magnetic field strength outside the cable and the frequency. Sensor coils can have a ferrite core. In that ca they are more nsitive, but no longer linear, which is undesirable from the calibration point of view. Their advantage over the capacitive nsors is that they control the whole circumference of the cable.
The Fourier transforms of the impuls (15 pC measured by the IEC 60270 method) picked up in time-domain by capacitive and inductive nsors are shown in Fig. 4, 5 respectively. It can be en that the capacitive nsor captures broadband frequency components in a range of 500-1000 MHz. The inductive nsor provides a different FFT spectrum with the highest signal-noi ratio in a range from 200 (filter’s cut-off frequency) up to 350 MHz. Two main narrowband interferences of approx. 730 and 940 MHz can be found in all laboratory measurements, the are the local DVB-T and GSM signals accordingly.
订金和定金的区别
Sensitivity of UHF nsors
It is known that the calibration of the UHF method in terms of apparent charge is impossible. Hence in order to be able to judge in the future, if a termination is faulty or not, a so called nsitivity check must be performed under laboratory conditions. For this purpo a conventional PD
Figure 6: Structural diagram of the comparative measurement between an UHF and IEC60270 PD
detecting methods
P e a k v o l t a g e o f U H F p u l s e s (m V )
Apparent charge measured acc. to IEC 60270 (pC)
Figure 7: Sensitivity lines of the UHF nsors (without
pre-amplification) measurement according to the IEC 60270 standard is carried out simultaneously with the UHF diagnostics [4]. The structural diagram of the nsitivity check is shown in Fig. 6. As a result, the nsitivity below 5 pC turned out to be detectable by both types of UHF nsors inside the sleeve.
To establish functionality between the key parameters of both methods, namely the peak voltage of t
he UHF puls and the corresponding values of the apparent charge, a ries of simultaneous measurements was performed. The PD intensity was incread by pushing the plastic lath farther beneath the stress cone of the termination. AC voltage of 50 kV (one pha, rms) was constantly applied during the test. A 40 dB pre-amplification was omitted due to the saturation effect at the higher levels of discharge activity. The results are shown in Fig. 7. Comparing the nsors it can be en that capacitive coupling has generally a higher nsitivity, than the inductive one. This is due to the narrower band of the inductive nsor, limited by the highpass filter downwards and lf resonance frequency upwards. The latter is determined by the relation of the inductivity to the parasitic capacitance of the coil.
>200 MHz 1-1000 MHz
Correlation between the UHF pul energy and apparent charge
Several publications note that energy is more objective criterion to compare the UHF and IEC60270 methods [5]. The latter reacts on a charge, which eventually reprents the stored energy. Peak voltage, on the contrary, is more a parameter of the discharge current growth rate and can differ greatly for different types of PD defects and strongly depends on damping characteristics of materials on the way from defect’s origin to the nsors.
A ries of laboratory measurements was made on the termination with an artificial defect to establish the correlation between the UHF pul energy and the apparent charge. The IEC 60270 signals were taken directly at the output of the quadropole after a preamplifier, to enable time respons of the same order. Energy of the puls was calculated according to the formula:
dt u E ∫
Ω=
2
501
凝血时间(1)
Fig. 8 reflects around 200 PD events in the range of 5-900 pC acquired at constantly applied voltage of 50 kV.
E n e r g y o f U H
F p u l s e s (p J )
Apparent charge measured acc. to IEC60270 (pC)
Figure 8: Correlation between the energy of the UHF puls and the corresponding apparent charge
ENHANCED OPPORTUNITIES OF ON-SITE METHOD’S APPLICATION
Pha resolved UHF PD measurements
Often in a noisy on-site environment a repetitive pul activity picked up by the diagnostic system can be fally interpreted as possible PD. It is known that partial discharges in electrical insulation only occur at certain moments depending on the pha of applied voltage. So one can easily differentiate between power cycle related puls and the rest. Fig. 9 shows an example of the so called pha resolved PD measurement using the UHF field nsing technique. Triggering the oscilloscope on the AC sine curve of 50 Hz, the UHF pul activity can be monitored using “peak hold” mode of the oscilloscope. Clear evidence that defect incepts at the rising part of the applied voltage proves that we deal with an internal PD in this ca.
Figure 9: Pha resolved UHF PD measurement with
“peak hold” function Regarding an on-site testing in GIS substations there are very few possibilities to extract each pha signal. Line triggering with the own supply voltage is sufficient to proof an AC cycle relation of repetitive puls.
UHF PD measurements with Power Detector
The UHF method implies per definition the usage of a high speed digital oscilloscope. It is necessary to resolute events lasting only a few nanoconds. Thus, for the ur this assumes substantial invest
ments into the test equipment and training of workmanship. In attempt to make the diagnosis more affordable the so called power detector (PwD) was developed in the laboratory in place of the filter and pre-amplifier shown in Fig. 6. Basically, a PwD is a linear or logarithmic amplifier that monitors signal’s power, not just the voltage output [6]. This allows extremely nsitive triggering. Fig. 10 shows a PD measurement with a magnitude of 3 pC clearly detected by the PwD and concealed for our regular UHF circuitry.
Figure 10: Sensitive measurement with the PwD, regular UHF and according to IEC 60270 (3 pC) The cost advantage is in the limited output bandwidth of the PwD that allows us to deploy simpler oscilloscopes. Fig. 11 shows a measurement picked up by an oscilloscope with an analogue bandwidth of only 350 MHz. On the other hand having a lower frequency output eliminates an opportunity of spectral analysis and limits the possibilities of results interpretation.
UHF PD measurements with an integrated voltage nsor
A significant number of plug-in connecting systems are equipped with a capacitive voltage nsor, integrated in the bushing. The output signal of this nsor is normally ud to signalize if the operating voltage is applied or not. It can be ud for PD diagnosis as well.
AC cycle of the applied voltage 2 ms/div
UHF nsor output at a moment
…Peak hold“ mode for UHF nsor output after a few conds
Regular UHF diag. circuit (filter, ampl.) 5 mV/div
PwD output
康巴藏族500 mV/div 100 ns/div
IEC 60270 output 5 pC/div明天的英语单词
Figure 11: Pha resolved PD measurement carried out with the PwD comparing to the conventional
testing
Figure 12: Connection of a coaxial cable to the
voltage nsor inside the bushing Fig. 12 shows a connection of a coaxial cable to the voltage nsor’s output. The measuring cable was directly (without filtering and pre-amplification) connected to the 50 Ohm input of an oscilloscope.
A plug-in cable connector with an artificial PD defect was plugged into the bushing equipped with the voltage nsor. A few comparative tests to establish nsor’s nsitivity in accordance with the scheme shown in Fig. 6 were conducted. Fig. 13 shows nsor’s output at a partial discharge intensity of 100 pC picked up with a oscilloscope in time-domain. An FFT transform of this signal is given in Fig. 14. Frequency components below 100 MHz dominate in the signal that allows a usage of oscilloscopes with narrow bandwidth. Maximal nsitivity without any amplification lies below 10 pC.
FIELD EXPERIENCE
On-site PD measurements were made on cable terminations in the manhole of GIS, while the cables
were in rvice. The portable screening sleeve equipped with a t of UHF nsors was clamped around each termination in turn as shown in Fig. 15. A 40 dB pre-amplification was generally ud during the on-site testing.
-0,05
0,00
0,050,100,15 100 pC, no amplification
图式理论V o l t a g e o u t p u t (V )Time (ns)
Figure 13: Output of the voltage nsor at 100 pC
050
100150200250300350400
12345
Frequency (MHz)
A m p l i t u d e (m V s )
Figure 14: FFT spectrum of the pul
Figure 15: On-site test t-up: portable screening sleeve with the UHF nsors mounted on each
termination in turn Fig. 16 shows a typical pul signal taken on-site on a faulty termination with a 3 GHz oscilloscope. The frequency spectrum of this pul and the spectrum of background noi are plotted in Fig. 17. Besides some discrete broadcast and GSM frequency spikes there are veral other broadband frequency components that clearly indicate the prence of PD activity.
Output of PD monitor acc. IEC 60270 60 pC/div
500 mV/div, 2 ms/div
0300
60090012001500
12345Frequency (MHz)
A m p l i t u d e (m V s )
-100
100
200
300
400
500
-
0,4
-0,20,0
0,20,4
0,6
A m p l i t u d e (V )
Tim e (ns)
Frequency (MHz)
A m p l i t u d e (V s )
Figure 17: Frequency spectra of the pul (light curve) vs. background noi (black filled area)
CONCLUSIONS
The experiences in on-line application of the ultrawide band partial discharge (PD) detection in high voltage plug-in cable terminations were discusd in the paper. The prented UHF PD detection method can be applied to proof the quality of asmbly work during commissioning, as well as on a regular basis after years in rvice to detect aged and risky terminations as a part of the condition-bad maintenance. The UHF nsors developed can be ud without screening sleeve too, if there are any difficulties mounting, with lower nsitivity though.
Comparative study of capacitive and inductive UHF PD nsors, as well as other laboratory tests has been shown: o the nsors ud were effective in the frequency range of 500-1000 MHz for the capacitive and 200-350 MHz for the inductive ones respectively,
o The maximal nsitivity of the screened nsors is below 5 pC for both types,
o A correlation between the key parameters of the UHF method, namely peak voltage output and en
ergy of the puls and the apparent charge of IEC 60270 method has been established for both types of nsors,
o Pha resolved measurements are effective to cope with repetitive pul noi,
o Usage of a power detector can reduce costs of the test equipment. Detecting with the PwD showed higher nsitivity but limited options of results interpretation in frequency-domain.
REFERENCES
[1] IEC 60840/Rev: Power cables with extruded insulation
and their accessories for rated voltages above 30 kV up to 150 kV – Test methods and requirements
[2] N.H. Ahmed, N.N. Srinivas, 1998, "On-line partial
discharge detection in cables"- IEEE Trans. Diel. Elect. Insul., Vol. 5, No. 2, pp. 181-188
[3] N. de Kock, B. Coric, R. Pietsch, 1996, "UHF PD
detection in GIS – suitability and nsitivity of the UHF method in comparison with the IEC270 metho红烛颂
d"- IEEE Elect. Insul. magazine, Vol. 12, No. 6, pp. 20-26
[4] M. Hanai, F. Endo, S. Okabe, T. Kato, H. Hama, M.
Nagao, 2006, …New development for detecting partial discharge using an UHF method and its application to power apparatus in Japan“, prented at CIGRE Session, Paris, France, paper Nr. D1-106
[5] L. Yang, B. Stewart, A.J. Reid, M.D. Judd, R.A.
Fouracre, 2005,"Study on combining UHF techniques with the IEC60270 standard for monitoring partial discharges in HV plant"- Int. Symp. on High Volt.. Eng. (ISH), Beijing, China, paper Nr. G-011
[6] S. M. Hoek, U. Riechert, T. Strehl, S. Tenbohlen,
2007, "New Procedures for Partial Discharge Localization in Gas Insulated Switchgears in Frequency and Time Domain"- 15th International Symposium on High Voltage Engineering, Ljubljana, Slovenia (accepted)
