1、IEEE Std 1531-2003IEEE Standards1531TMIEEE Guide for Application andSpecification of Harmonic FiltersPublished by The Institute of Electrical and Electronics Engineers, Inc.3 Park Avenue, New York, NY 10016-5997, USA24 November 2003IEEE Power Engineering SocietySponsored by theTransmission +1 978 75
2、0 8400. Permission to photocopy portions of any individual standard for educationalclassroom use can also be obtained through the Copyright Clearance Center.Authorized licensed use limited to: IHS Stephanie Dejesus. Downloaded on March 05,2010 at 13:56:43 EST from IEEE Xplore. Restrictions apply. iv
3、 Copyright 2003 IEEE. All rights reserved.Introduction(This introduction is not part of IEEE Std 1531-2003, IEEE Guide for Application and Specification of HarmonicFilters.)This guide addresses the specification of the (1) components, (2) protection, and (3) control of harmonic fil-ters. It does not
4、 address the proper sizing or configuration of harmonic filters to achieve desired perfor-mance. This document provides guidelines for passive shunt harmonic filters for use on 50 Hz and 60 Hzpower systems. No specific standards exist for harmonic filters, although standards do exist for virtually a
5、llof the components that are used in a filter.ParticipantsDuring the time this guide was being developed, the IEEE working group sponsored by the Capacitor Sub-committee of the Transmission and 8% or less for most user busses at less than 1 kV. The total demand current distortion at the point ofcomm
6、on coupling to the utility is limited to the range of 2.5% to 20%, depending upon the size of the cus-tomers harmonic-producing load and other factors. (See IEEE Std 519-1992 for details.) The document alsogives higher limits for conditions lasting less than 1 hour.4.2.2 Equipment withstand capabili
7、tiesSome of the withstand capabilities that are described in existing equipment standards are summarized in thissubclause.When transformers are operating at rated load, the total harmonic current distortion should be limited to 5%as defined in IEEE Std C57.12.00-2000 and IEEE Std C57.12.01-1998.12IE
8、EE Std C57.110-1998 definesthe method for derating transformers when supplying nonsinusoidal loads. UL 1561-1999 and UL 1562-1999 define the transformer K-rating that is intended for use in high harmonic environments.IEEE Std 18-200213states that capacitors are intended to be operated at or below th
9、eir rated voltage.Capacitors shall be capable of continuous operation under contingency system and bank conditions providedthat none of the following limitations are exceeded:a) 110% of rated rms root-mean-square voltageb) 120% of rated peak voltage, i.e., peak voltage not exceeding 1.2 x (square ro
10、ot of two) x rated rmsvoltage, including harmonics, but excluding transientsc) 135% of nominal rms current based on rated kvar and rated voltaged) 135% of rated kvarAdditional application guidelines for capacitors are given in IEEE Std 1036-1992. It should be noted thatcapacitor fuses should be rate
11、d for the voltage and current, including harmonics, in a filter application.The limitation to 135% of rated kvar in IEEE Std 18-2002 is based on dielectric heating at fundamental fre-quency and is based on the thermal stability test in that standard. The 135% limit in IEEE Std 18-2002 isbased on a m
12、aximum operating voltage of 110% of rated voltage and a maximum capacitance tolerance of+15% (the maximum allowable tolerance at the time the 135% limit was set). (1.12 , thus135%.)The total dielectric heating in a capacitor is a function of the force between the electrodes, the capacitance ofthe di
13、electric, and the number of force reversals per second.The force is the result of the attraction of the positive and negative charges on the electrodes. The magnitudeof the charge Q on each of the electrodes is proportional to the voltage difference V between the electrodes.The force is proportional
14、 to the product of the charge magnitudes. Because the positive and negative chargesare equal to each other and are proportional to the applied voltage, the force (and losses) varies as the squareof the applied voltage. 11This information is from the referenced standard(s) and does not transplant the
15、se limits to this guide. See the first paragraph ofClause 2.12See footnote 11.13This quotation is from the referenced standard(s) and does not transplant these limits to this guide.1.15 1.35Authorized licensed use limited to: IHS Stephanie Dejesus. Downloaded on March 05,2010 at 13:56:43 EST from IE
16、EE Xplore. Restrictions apply. IEEEStd 1531-2003 IEEE GUIDE FOR APPLICATION AND6 Copyright 2003 IEEE. All rights reserved.The total dielectric heating varies linearly with the capacitance of the dielectric. Changes in capacitance dueto change in area of the dielectric, thickness of the dielectric, a
17、nd changes in the dielectric constant due tosmall materials variations all affect the total heating linearly.The dc dielectric losses and resulting heating in a high voltage power capacitor are very small. The dielectriclosses are dominated by ac losses. Each time the force is reversed there is an a
18、mount of loss. The ac lossesare linear function of the applied frequency.Therefore, for a single frequency, the dielectric losses are proportional to the square of the applied voltage,the capacitance, and the frequency, as shown in Equation (1):(1)Note that, for a capacitor, the reactive power Q is(
19、2)(3)(4)Note the similarity in the expressions for dielectric loss and Q. For a single frequency, the dielectric heatingin the capacitor is proportional to the reactive power (measured in kilovars).For a capacitor in a filter, there are multiple frequencies generating the dielectric heating. For fil
20、ter applica-tions where (1) there is no significant dc voltage present, (2) the harmonic voltages across the capacitor aresmaller than the fundamental frequency voltage, and (3) the highest significant frequency is less than about1 kHz, the dielectric heating will be within the 135% limit if(5)or(6)
21、wheref is the rated frequency of the capacitor and system (Hz),C is the actual capacitance of the capacitor (F),h is the harmonic order, for all significant harmonics including the fundamental (h = 1),V(h) is the capacitor voltage (rms) at the h harmonic (kV),I(h) is the capacitor current (rms) at t
22、he h harmonic (A), Qratedis the capacitor rated reactive power (kvar).The inequality in Equation (6) is the one normally used to determine whether the dielectric heating is accept-able. While dc losses in a capacitor are small, the presence of dc voltage can increase the maximum chargeand significan
23、tly increase the ac losses. That increase is not reflected in the above inequalities, limiting theuse of these inequalities to applications where no significant dc voltage is present.Higher frequency currents may result in eddy current or induced losses in addition to the dielectric losses.Where the
24、 harmonic currents are smaller than the fundamental current, the error caused by ignoring theseDielectric Loss f CV2QV()I()=V()2 f CV()=2 f C V()2=2000 f C hV h()2()h1.35 QratedVh()Ih()h1.35 QratedAuthorized licensed use limited to: IHS Stephanie Dejesus. Downloaded on March 05,2010 at 13:56:43 EST
25、from IEEE Xplore. Restrictions apply. IEEESPECIFICATION OF HARMONIC FILTERS Std 1531-2003Copyright 2003 IEEE. All rights reserved. 7losses is negligible. Where the harmonic currents are larger or there are significant harmonic currents above1 kHz, higher frequency capacitor equipment designs, which
26、are beyond the scope of this guide, may berequired.Proposed derating curves for harmonic voltages for constant speed motors are given in various references(see IEEE Task Force B3 and Rice B6). Typically, these curves indicate that the derating of the motoroccurs for voltage distortions greater than
27、5%. For a typical distribution of harmonics, significant derating ofthe motor begins at about 8% total harmonic distortion (see IEEE P519.1/D8b14).4.3 Normal system conditionsThe normal system operating conditions are generally evaluated to assure that the harmonic filter design willmeet specific re
28、active power (kilovars) and harmonic performance requirements for these conditions. Thesenormal system operating conditions include the following:a) All harmonic voltages and currents, including1) Characteristic harmonics of all expected loads.2) Uncharacteristic harmonics. Frequencies that are not
29、theoretically characteristic of a perfectlyoperating device may sometimes occur. (These uncharacteristic harmonic frequencies mayinclude even harmonics, triplen harmonics, and harmonics that are not integral multiples of thepower system frequency.) Analysis, experience, and field measurements will o
30、ften help toquantify these values. 3) Background and future harmonic loads. Harmonics generated by other loads near the proposedload location will affect the harmonic current in the harmonic filter. In addition, some futureharmonic generating loads should be anticipated to reduce the probability of
31、overloading theharmonic filter.b) System voltage variation. Overvoltages to +5% are typically considered for normal load conditionsand +10% for unloaded system conditions. Undervoltage conditions are generally not critical forharmonic filter design, unless voltage is lost completely. In that case, t
32、he harmonic filters should bedisconnected from the system immediately until the system is restored to normal conditions.c) System frequency variation. On the interconnected power system, frequency variations beyond 0.1 Hz are rare. Larger frequency variations may occur when the system is fed from a
33、local gener-ator. They can affect the duty of the harmonic filter and can also have a profound impact on theoverall harmonic performance of the system.d) Power system configurations. Possible variations in the power system configuration that may affectthe filter should be evaluated. For example, in
34、industrial power systems, changes in the supply trans-former and reconfigurations of the medium voltage feeders have the largest effect on the systemimpedance. Changes in the source power system (e.g., generation, transmission/distribution, trans-former changes) may also have a significant impact. T
35、hese evaluations should include a realisticrepresentation of the system and its componentsgenerators, transformers (with proper windingconnections), lines (with conductor skin effect and capacitive charging), capacitors, reactors, and allsources of harmonics electrically close to the proposed filter
36、 site.e) Loading conditions. Variations in the system load, as it affects the harmonic filter design, should beconsidered. Conditions to be considered include variations in the harmonic-producing loads, in thestatus of ac motors, and in the status of system capacitor banks and other harmonic filters
37、. Linear(resistive) loads also should be included as they help to dampen harmonic resonant peaks. f) System voltage unbalance. System unbalance can result in increased harmonic injections fromdistortion-producing equipment, particularly triplen harmonics, and facilitate their propagation onthe syste
38、m.14Numbers preceded by P are IEEE authorized standards projects that were not approved by the IEEE-SA Standards Board at the timethis publication went to press. For informaiton about obtaining drafts, contact the IEEE.Authorized licensed use limited to: IHS Stephanie Dejesus. Downloaded on March 05
39、,2010 at 13:56:43 EST from IEEE Xplore. Restrictions apply. IEEEStd 1531-2003 IEEE GUIDE FOR APPLICATION AND8 Copyright 2003 IEEE. All rights reserved.4.4 Normal harmonic filter conditionsFilters are seldom tuned to their exact calculated values. It is necessary to allow for the following parameterv
40、ariations when evaluating the performance of the harmonic filters:a) Component tolerances. Manufacturing tolerances must be considered for the inductance, capaci-tance, and resistance.b) Ambient temperature variations. Capacitance and resistance both vary with temperature. The appro-priate temperatu
41、re range depends upon the location. Capacitance variation with temperature is typi-cally in the range of 0.4% to 0.8% decrease per 10 C increase in temperature.c) Capacitor element or unit failures. Capacitor failures will result in a change in the harmonic filtertuning and may result in overvoltage
42、 on some parts of the capacitor bank. Larger harmonic filtersmay continue to give acceptable performance with a limited number of failed capacitor elements orunits. However, for smaller harmonic filters, the failure of one capacitor element or unit can cause arelatively large change and require the
43、harmonic filter to be immediately disconnected.4.5 Contingency system conditionsThe contingency system operating conditions are generally evaluated to assure that the harmonic filterdesign will be rated adequately to handle these conditions although the normal system distortion limits maybe exceeded
44、. These include the following:a) Switching. The switching of harmonic filters or other system components may result in significantovervoltage duties for the harmonic filter components.The energization of large transformers can result in severe dynamic overvoltages on harmonic filtercomponents. When
45、several single-tuned harmonic filters are energized simultaneously, the transient overvoltagescan be very severe on small harmonic filters tuned to relatively low frequencies.When multiple-tuned steps are switched individually, it may be necessary to assure that they areswitched on and off in the pr
46、oper order so that an undesirable parallel resonance does not occur.When switching harmonic filters, it is important that an adequate delay time be maintained betweenan open and a subsequent close. This delay allows time for the trapped charge on the capacitors todecay so that the resultant energizi
47、ng transient voltages are not excessive. A delay time of 5 min isused with high-voltage capacitors (rated over 1000 V, where IEEE Std 18-2002 requires a dischargeto 50 V in 5 min), and 1 min for low-voltage capacitors (rated at 1000 V or less, where IEEE Std 18-2002 requires a discharge to 50 V in 1
48、 min). Shorter time delays may be acceptable if a modestincrease in the energizing transient is acceptable. See 5.2.1 and 6.4. Where very short delays are required, discharge devices to rapidly discharge the capacitors may beused. Switching devices with insertion resistors or reactors, or switching
49、devices designed to closewith near-zero voltage across the open contacts, may also be used to control system transientvoltages.b) Application of filters tuned to the same frequency. When harmonic filters are applied at the samelocation and are tuned to the same frequency, care must be taken to assure that there is acceptablesharing of the harmonic currents among the harmonic filters. This current sharing is a function of thedifferences in the impedances of the harmonic filters.c) System frequency variation. Frequency variations greater than the frequency v
