Referans Kısmi Boşalma Kaynağının Yüksek Doğru Gerilim Koşullarında İncelenmesi

Abstract

Bir sivri uç-yarıküre elektrot düzeni olan referans kısmi boşalma (korona) kaynağı, yüksek gerilimde, sabit genlikli ve sıklıklı boşalma darbeleri üreten ve kısmi boşalma ölçme devrelerinin ölçeklenmesinde kullanılan bir düzendir. Alternatif gerilimde kullanım koşulları belirli olan bu düzenin, bu çalışmada, doğru gerilimde kullanım olanakları araştırılmış ve çalışma koşulları belirlenmeye çalışılmıştır. Referans korona kaynağında boşalma büyüklükleri ile elektrot düzeninin biçim ve boyutları arasındaki ilişkiyi incelemek amacıyla kuramsal ve deneysel elektrostatik alan incelemeleri yapılmıştır. Analitik alan incelemelerinde, yaklaşık hesap yöntemi, görüntü yükleri yöntemi ve sferoidal ve paraboloidal koordinatlar; sayısal alan incelemelerinde sonlu elemanlar yöntemi; deneysel alan incele melerinde de elektrolitik banyo ve yarıiletken kağıt yöntemleri kullanılmıştır. Yüksek doğru ve alternatif gerilimde yapılan kısmi boşalma deneylerinde, referans korona kaynağında boşalma büyüklüklerine siv ri uç eğrilik yarıçapının, elektrot açıklığının, sıcaklığın ve basıncın etkileri incelenmiş; kısmi boşalma ölçme frekansının önemi ve kaynağın çevreye yayınladığı radyo parazitlerinin uzaklık ve gerilimle değişimi araştırılmıştır. Deneysel incelemelerde, kuramsal incelemelerde de olduğu gibi, boşalma büyüklüklerine etkiyen en önemli etkenin sivri uç eğrilik yarıçapı olduğu görülmüştür. Yapılan incelemeler, referans korona kaynağının yüksek doğru gerilim koşullarında da kullanılabileceğini göstermiştir.

The sensitivity and the calibration of partial discharge detection systems are important aspects of general subject of partial discharge measurements. Calibration of instruments measuring several discharge quantities in the complete test arrangement should be made by injecting short current pulses of known magnitude into the terminals the measuring circuit. The calibration can be made either in the low voltage mode or in the high voltage mode., Calibration in the low voltage mode is to supply an artifical discharge of known magnitude at the terminals of the detection circuit. This type of calibration devices are known as primary dis charge standard sources. Secondary discharge standards which are used in the high voltage mode are usually made from sources of natu ral discharges, i.e., surfaces discharges, internal discharges and corona discharges. Among these, corona discharges around a sharp point are the most suitable ones. One convenient type of corona discharge reference source (CDRS), which is suitable for direct connection in the high voltage test circuit, is a point-hemisphere discharge gap (Fig.l). The device consists essentially of a needle pointing into a hemisphere. Hemisphere electrode Fig.l. Corona discharge reference source. R=25 mm, r=0.05 - 0.5 mm. Under certain conditions, this gap will give a discharge magnitude which is constant to within +_ 10 %. The discharge level is dependent on the radius of curvature of the point. At the incep tion voltage one or two discharge pulses occur during each negative half-cycle of the 50 Hz sine wave. If the voltage is increased, the number of discharges (repetition rate) increases, but the dis charge magnitude remains constant. vıı Corona discharge reference source have been analyzed for alternating voltages. The discharge phenomena under direct voltage conditions takes approximately 10 nanosecond. This duration is extremely short, compared with the half period of a 50 Hz sine wave. Therefore, the discharge process at direct voltages can assumed to be the same with that at alternating voltages. The calibration of partial discharge measurements under high direct voltage conditions are almost done by primary discharge standard sources as in the high alternating voltage case. In this thesis, several characteristics and the operating aspects of the corona discharge reference source are investigated and demonstrated for high direct voltage conditions. It was aimed to show that the corona discharge reference source can be used as an alternative calibration source under high direct volt age conditions. Moreover, as an extension to alternating voltage studies, the variation of the corona inception and extinction volt age with the temperature and the pressure are determined experimen tally. The studies began with the theoretical analysis of discharge phenomena. Afterwards, potential and field analysis, which yield a close dependence with the discharge quantities (partial discharge inception voltage, apparent charge, repetition rate...) of corona discharge reference source have been conducted by theoretical and experimental methods. Finaly, several partial discharge tests at high alternating and direct voltages have been carried out and the results are compared both with eachother and with of those reported in the literature. Both the analytical and numerical electrostatic field calcu lations require the solution of Laplace (or Poisson) equation. Analytical methods can easily be applied in case of simple electrode configurations. However, most electrode systems in nature do not posses such a simple configurations. In case, one of the following methods must be applied: (1) Approximated solutions, (2) Coordinate transformations or conformal mapping, (3) Numerical methods, (4) Experimental methods. Since the corona discharge reference source yields a non uniform field, it is impossible to perform an exact analytical cal culation without coordinate transformation. Theoretical field analysis of corona discharge reference source have been done by approximated solutions, coordinate trans formations and numerical methods. In approximated solutions, point- hemisphere configuration of corona discharge reference source is transferred to concentric-hemisphere and sphere-plane electrode systems. The calculated maximum electric field strength and capa citance values of the electrode system are found to be nearly the same for both approximations. Coordinate transformation can be done by choosing the most adequate coordinate system for the electrode configuration. Prolate spheroidal and parabolic coordinate systems are preferred for corona discharge reference source and the solutions of Laplace equation are viii carried out in that coordinate systems. The point-hemisphere con figuration is considered a hyperboloid point-plane in prolate spheroidal coordinate system and two confocal paraboloids in para bolic coordinate system. In these coordinate transformations, utilization factor r) is obtained as: n = sin2(a/2).tanh 1 [cos(a/2)] cos(a/2) (1) for a hyperboloid point-plane configuration and ln(2p-l) n " 2(p-l) (2) for two confocal paraboloids. Where a is the top angle of point electrode, p is R/r as a geometrical characteristic. R and r are hemisphere radius and point radius, respectively. For the applied voltage U and electrode spacing a, the maximum field strength E^ may be calculated as: max E = U/a.n max (3) Among these coordinate transformations, parabolic coordinate system is more adequate for corona discharge reference source, because of the close similarity between a parabolic and a point electrode. Fig. 2 shows the equipotential lines (surfaces) calculated in para bolic coordinate system. Fig. 2. The equipotential lines calculated in parabolic coordinate system. Numerical methods have become more and more attractive with the increasing availability of modern high-speed digital computers, Among them, finite element method (FEM) is the most preferred one. The basic principle of the finite element method is to divide the entire solution domain into arbitrarily chosen finite elements and then to find a piece-wise solution of potential function in each element in which approximate potential function is expressed by a xx polynomal. Here, two dimensional and rotationally symmetric field produced by corona discharge reference source is examined by using triangular elements. The region of solution is divided into a nonuniform mesh by automatic grid generation. Cylindrical coordi nates are used for the formulation. The resulting sparse, simulta neous, lineer equation system is solved by successive over-relaxation method. Fig. 3 shows the equipotential lines and grid obtained from finite element method. Fig. 3. The equipotential lines and grid obtained from finite element method. Experimental field analysis of corona discharge reference source have been conducted both by electrolytic tank method and semiconductive paper method for the aim of checking and extending the theory. Experiments are done with several kinds of 1/5 and 1/10 magnified models. The results obtained by these experimental methods are nearly the same of the theoretical ones and therefore will not be reproduced here. Final studies are about partial discharge measurements. The equipment used in the experimental works and the measurement funda mentals are described. At 50 Hz AC and DC voltages, partial dis charge tests are performed to investigate. (1) The importance of measuring frequency, (2) The versus of the partial discharge inception voltage and discharge intensity with the radius of curvature of point electrode, (3) The effect of electrode spacing on the partial discharge (corona) inception voltage, (4) The effect of temperature on the partial discharge inception voltage, (5) The effect of pressure on the partial discharge inception voltage, (6) The RIV (Radio Interference Voltage) of corona discharge reference source, (7) The partial discharge repetition rate versus applied voltage and (8) The determination of breakdown voltage. The overall results may be summarized as follow: (1) The magnitude of partial discharges is mainly determined by the radius of curvature of the point. The partial discharge incep tion voltage also varies with the radius. The partial discharge quantities vary with the top angle of the point. The radius of the hemisphere does not substantially affect the discharge magnitude. (2) The experiments performed on the change, in magnitude with change in point-to-hemisphere distance show that over a 5 mm range, the discharge magnitude can be reduced by approximately 28 % at negative voltage and 20 % at 50 Hz AC without seriously affecting the stability of discharge pattern. Increase of the separation increases the discharge inception voltage but only by approxi mately 6 % at all cases over the range indicated, as this change is almost a linear function of separation. Here, the normal position of the point, i.e. where it coincides with the diametric plane of the hemisphere, has been taken as zero separation. (3) Under high direct voltage conditions, corona discharge reference source provides discharge pulses of constant magnitudes as in alternating voltage conditions, but the magnitudes vary with the polarity. The partial discharges appear sooner at negative than at positive voltage; with AC voltage they occur often during the negative half-cycle of sine wave only. (4) The discharge (corona) inception voltage with temperature reduce by approximately 10 % from the initial value for a 25-75 C temperature range. (5) The partial discharge inception voltage decrease considerably with pressure at low pressure range. Discharge inception volt age with pressure increase by approximately 75 % from the initial value for a 0-5 bar pressure range. (6) The partial discharge repetition rate is strongly dependent on the voltage. The repetition rate increases with applied voltage as in alternating voltage case. (7) The RIV (Radio Interference Voltage) level of corona discharge reference source is not of importance. (8) Corona discharge reference source can be used as an alternative calibration source under high direct voltage conditions. It is sufficiently reliable to be accepted as a secondary discharge standard. It is simple and of reasonable size, and a range of specific discharge magnitude can be obtained using only a small number of points. xx SEMBOLLER a b c C C a Cb D A Af E E o E max r f(n) f.. o Sivri uç elektrodun tepe açısı; iyonlaşma katsayısı Elektrot açıklığı; elips ve hiperbolde büyük eksen uzunluğu Elips ve hiperbolde küçük eksen uzunluğu Elips ve hiperbolde yarı odak uzaklığı Kapasite Deney cisminin kapasitesi Bağlama (kuplaj) kondansatörü Akı yoğunluğu Korona tabakasının kalınlığı Band genişliği Elektrik alan şiddeti Ortalama elektrik alan şiddeti Maksimum alan şiddeti Dielektrik sabiti -12 F/: m Boşluğun dielektrik sabiti = 8,86.10 Bağıl dielektrik sabiti Kısmi boşalma ölçü aletinin boşalma tekrarlama sıklığına bağlı bir fonksiyonu : Ölçme frekansı h, h, h : Genel metrik katsayılar u v w J h, h_, h, : Sf eroidal koordinat sisteminde metrik katsayılar n 8 (J> h, h, h, : Paraboloidal koordinat sisteminde metrik katsayılar -*??*?-*? i, iQ, i : Silindirik koordinat sisteminde birim vektörler r o z xıı } n n, RK q Q t T U Ud u u dA U. i U. ıa U. in U. IP Sferoidal koordinat sisteminde birim vektörler Paraboloidal koordinat sisteminde birim vektörler Akım yoğunluğu Düzeltme faktörü Öziletkenlik Anten uzaklığı Kısmi boşalma tekrarlama sıklığı Faydalanma faktörü Sferoidal koordinatlar Paraboloidal koordinatlar Basınç; geometrik karakteristik Sivri uçlu elektrodun eğrilik yarıçapı Yarıküre elektrodun yarıçapı; direnç Yarıçap değişkeni Silindirik koordinatlar Koruma direnci, ön direnç Boşalma genliği (pC); görünen yük Elektrik yükü Sıcaklık Zaman sabiti Gerilim Delinme gerilimi Korona tabakasının delinme gerilimi Kısmi boşalma sönme gerilimi Kısmi boşalma (korona) başlangıç gerilimi Alternatif kısmi boşalma (korona) başlangıç gerilimi Negatif kısmi boşalma (korona) başlangıç gerilimi Pozitif kısmi boşalma (korona) başlangıç gerilimi xııı U : Boşalma genliği (yV) ; radyo parazit gerilimi u, v, w : Genel koordinatlar x : Uzaklık değişkeni x, y, z : Kartezyen koordinatlar V : Potansiyel W : Enerji Z : Kısmi boşalma ölçü empedansı KISALTMALAR SYK : Sivri uç-yarıküre elektrot düzeni RKK. : Referans korona kaynağı AG : Alternatif gerilim DG : Doğru gerilim TS : Türk Standardları IEC : International Electrotechnical Commission BS : British Standards CISPR : Comite International Special des Perturbations Radioelectriques NEMA : National Electrical Manufacturers Association RIV : Radyo parazit gerilimi (Radio Interference Voltage) MWB : Messwandler - Bau GMBH HFM : MWB RIV-metresi ÖA : Ölçü aleti, RIV-metre, pC-metre KIO : Katot ışınlı osiloskop PVC : Polivinilklorür IEE. Institution of Electrical Engineers IEEE : Institute of Electrical and Electronics Engineers