Draft report of the sub-group constituted by OCC to Review the df/dt or rate of change of frequency relay setting

In Northern Region.
May 2007

Index
1. 2. 3. Introduction/Background Estimation of rate of change of frequency (df/dt): Computation of Inertia Constant (H): 3.1. Based on inertia constant (H) of Individual Generators in the system [1]: 3.2. Based on rate of change of frequency using field data [2,3] 3.3. Based on rate of change of frequency using simulation studies df/dt Measurement principle by relays: Initial findings and conclusion: 5.1. df/dt estimation for combined North-Central System. 5.2. Available df/dt relay settings: 5.3. Recommended df/dt settings and further work: References/Bibliography

4. 5.

6.

Annex-1 Analysis of incident of loss of Generation Analysis of incident of System separation Annex-II Observed Frequency response at Vindhyachal following loss of generation incident. Annex-III Observed frequency response at Vindhyachal following system separation. Annex-IV Frequency response obtained through simulation studies. Annex-V ABB document on Frequency relay behavior. Annex-VI Extract from Toshiba Corporation document on Frequency and Voltage protection. Annex-VII Extract from Technical data sheet of AREVA manufactured MICOM relays.

1. Introduction/Background
The df/dt settings were carried out for the individual regions separately in past as a safety net of the respective regions. However, with the synchronous interconnection of Northern Regional grid and central grid comprising of Eastern region, Western region and North Eastern Region on 26th August 2006 at 1222 hrs the system size has increased from the order of 25000 MW to 60000 MW. Consequently inertia of the combined system has increased, resulting into lower rate of change of frequency for a given loss of generation or load. Therefore, the existing defense mechanism based on rate of change of frequency relay (or df/dt relay) for protection the system from collapse following a loss of generation has become practically ineffective. Present df/dt setting in the Northern region is as tabulated below. Table-1 Existing df/dt relay setting in Northern Region Stage Initiating Hz/Sec Frequency (Hz)

The present sub-group has been constituted by the Operation Co-ordination SubCommittee (OCC) of the Northern Region Power Committee (NRPC) in its 9th meeting held on 15th Dec 2006 to review the setting of df/dt relays. The sub-group was advised to submit its report by 12th Feb 2007. Due to other various pressing issues such as weather related trippings in Northern Region during January-07, February-07 and March-07, the sub-group could not finalise its recommendations. The first sub-group meeting was organized on 18th Jan 2007 and the following was decided. 1. The combined North & Central Grid requires relief through df/dt relays to take care of network loading conditions following a large loss of generation. 2. The combined system has witnessed several incidents of loss of generation up to 1500 MW since the synchronization of Northern Grid with the Central Grid. However, in none of the above occasions, the df/dt relays with present setting have not operated. A list of such incident in the Northern Region is enclosed at annex-I. 3. It was observed that some df/dt relays have a limited storage facility for frequency data. However, this was not archived and analyzed as a matter of practice and therefore practical observation on actual df/dt recorded for different events is not available. Therefore the df/dt relay settings for the combined North-Central grid would be worked out through the following methods.

Northern Region

Load Relief (MW)

1st 2nd 3rd 4th

48.8 49.2 49.6 50.0

0.125 0.2 0.3 0.4

750 1500 1500 1250

1

Based on these study. 5. Subsequent to the first meeting of sub-group on 18th January 2007. This might be on account of its limited efficacy in large systems.9 Hz) at 220 kV Gazipur and 220 kV Narela in the DTL system. Overloading of the generators is followed by considerable change in system voltages.1 Hz per second or very close to it as per relay limitation) at higher frequency (i. The findings of the same are given at annex-I & II. North-East and Southern regions for their valuable comments. 4. To analyze the incidents data from number of the substations is collected. 2. In the intervening period literature survey was carried out and work was done as decided in first meeting of the subgroup. causing a 2 . The group also felt that to have greater insight into functioning of the relay specially rate of change of frequency measuring principle and limitations if any. discussions with relay manufacturers shall be held. 49. analysis and discussions this report has been prepared and covers estimation of df/dt for different situations. The RRVPN and BBMB were also requested to revise the df/dt setting as above in their system at selected few locations. simulation data and other findings were discussed and analyzed at second meeting of the subgroup at 10th April 2007. International experience of df/dt relaying in very large power grids might also be studied separately. It was observed that internationally not much literature was available on df/dt relay setting philosophy. The operation of df/dt relay (actual response of the grid under contingencies) shall be monitored by revising the relay settings as follows: Revise setting with lower slope (0.e. It can be appreciated that these values give us good indication of rate of change of the frequency in the system. It is also suggested that the findings of this sub-group might be circulated in Western. Eastern. limitations and the initial recommendations of the sub-group. two instances of separation of Northern Region from the Central Grid have occurred. The estimated values through real time data. It was observed that the Vindhyachal HVDC back-to-back station had a digital recording of frequency at 100ms interval. the measuring principle adopted for df/dt relaying. Estimation of system inertia constant (H) through studies and real time frequency transient response. Thus this goldmine of field data was also used to evaluate ‘H’ for some of the tripping incidents.Estimation of system inertia constant (H) from individual machine’s inertia constant and MVA and assuming that all the units swing coherently (although this would strictly not be true in a real time wide spread interconnected system). Estimation of rate of change of frequency (df/dt): Power system frequency drops when the power supply in the system becomes insufficient due to loss of generation or tie line support.

The general equation of the system behavior following a loss of generation or load is as below . there are oscillations and these could lead to the higher line loading and consequently tripping of the generators/tie lines. changing Pmech. altering frequency ‘f ’ and its rate of change df/dt. While the low frequency ‘f ’ is the final result of the power deficiency. the generator’s governing system utilizes the spinning reserve (if available). In addition. However. df/dt can be computed if we have the value the system inertia constant (H). However. Power change (per unit in system load base) =Load prior to generation Loss in MW =System Generation after Loss in MW =Power/frequency characteristic of the system in pu/Hz =Change in frequency This general equation can be simplified by considering the damping effect to be negligible during early onset of the event. its rate of change df/dt is an instantaneous indicator of power deficiency and can enable recognition of the MW imbalance. Therefore the simplified equation without significant loss of information is 2H . Higher the inertia of the system larger would be the time and vice versa. df f o dt H fo df/dt PL PG D ∆f ∆P + Dx ∆f = ∆ P -----(1) Where = System Inertia Constant on system base (seconds) = Frequency at the time of disturbance (Hz) =Rate of change of frequency (Hz/Sec) =( PL-PG)/PG . The typical values of the H as given below: 3 . protection systems for the low frequency may be activated in the power plants and the consequent shutdown of plant may lead to the separation of the interconnected system or black-out of the power system. If the amount of frequency drop is large. df = ∆ P f o dt -----(2) Therefore. the change in machine speed is oscillatory in nature. The time taken by the system to settle to a low frequency would depend upon the inertia of the system. Thus Pmech and Pelec change in time.fluctuation in load power Pelec. Therefore it is important to grasp the characteristics of the frequency response of the power system to loss of generation in order to stabilize the system faster and avoid catastrophe. The system inertia constant is defined as the kinetic energy stored in the system per MVA. These oscillations depend on the response of the generators and are seen differently at different locations. before frequency settles to lower level. refer [3] 2H .

Based on rate of change of frequency. 1. Based on inertia constant (H) of Individual Generators in the system [1]: Assuming the machines to be swinging coherently. 1 Unit Type of Machine Water wheel generator Slow speed (< 200 rpm) High speed (> 200 rpm) Turbine Generator Condensing 1800 rpm Condensing 3000 rpm Non-Condensing 3000 rpm Synchronous motor with load varying from 1. P. the first figure applies to the smaller MVA sizes. Based on rate of change of frequency.e 10 samples per second 3. Kothari. calculated using simulation studies 3. 4 . 3rd edition.page-436 3. the equivalent inertia constant of the multi-machine system can be calculated by adding the inertia Constant of individual machine (on System base) Heq = Σ (Hi x Gi mach) / Σ Gi mach (i=1 to n) -----(3) = Σ (Hi x Gi mach) / Gsystem (i=1 to n) -----(4) Heq Where Heq Hi Gsystem Gi mach n =System Inertia Constant on system base =Individual Generator Inertia Constant on its own base =System Size (MVA) or System Base =Machine rating (MVA) or machine Base =total no.0 and higher for heavy fly wheels H* Constant (Sec) 2 to 3 2 to 4 9 to 6 7 to 4 4 to 3 2. New Delhi .0 to 5.Typical Inertia constant of Synchronous machines. Sl. Nagrath and D. J. Computation of Inertia Constant (H): As described above. that was calculate using field data (100 milisec frequency data recorded at Vindhyachal HVDC back to back station) i. “Modern Power System Analysis”.00 2 3 * where range is given . Source: I. Based on inertia constant (H) of Individual Generators in the system 2.1980 . following three methods has been used to determine the Inertia Constant H (Sec) of the Interconnected North-Central Grid. The results obtained are as tabulated below in the system load base. of machines in the system Based on the individual generator inertia constant on its own base used the PTI-PSS/E system studies model of all India network North Central grid Inertia constant was calculated. Tata MacGraw-Hill. No.1. The computation of inertia constant (H) is being described in subsequent section.

1. Based on rate of change of frequency using field data [2.Table -1 Region NORTH EAST WEST NORTH EAST North Central Grid Size (GW) 29 15 31 2 77 Size (MVA) 40542 22483 42555 2900 108479 H (Sec) 4.19 3. 2.93 3. Two major incidents of loss of generation when df/dt was more than 0. 2H . Units tripping at Suratgarh (1250 MW Generation Loss) on 04-102006 at 1822 hrs.06Hz/sec or more is observed only when there is big loss of generation say more than 500 MW or more.90 5. But df/dt of 0.033 Hz/sec. Power change (per unit in system load base) =Load prior to generation Loss in MW =System Generation after Loss in MW Knowing the rate of change of frequency following a known loss of generation.2. system inertia constant can be estimated using equation (3) if the system size in MW and frequency prior to event is known. Units tripping at Anpara (1450 MW Generation Loss) on 15-112006 at 1119 hrs. 5 .05 3.3] Based on the study of under frequencies following generation outages system inertia can be estimated using the simplified equation of inertia (derived from swing equation) considering damping effects to be small early onset of the events. It was found that system is regularly observing the rate of change of frequency of 0.06 Hz/sec were taken for inertia constant calculation.85 3. df = ∆ P f o dt H fo df/dt PL PG ∆P Where = System Inertia Constant on system base (seconds) = Frequency at the time of disturbance (Hz) =Rate of change of frequency (Hz/Sec) =( PL-PG)/PG .05 3.93 Table-2 Region NORTH CENTRAL North Central Grid Size (GW) 29 48 77 Size (MVA) 40542 67938 108479 H (Sec) 4. A number of incidents since NR-Central grid synchronization were analyzed.60 3.

It is observed that it contain oscillatory component. 4. 1311 hrs rate of change of more than 0. df/dt of more than 0. Higher value of df/dt is observed near the node where generation is lost. Some relays measure the change in frequency (∆ f) over a time interval (∆t=100ms). The rate of change of frequency measurement is based on two successive frequency measurements and the time difference between the frequency measurements. Based on rate of change of frequency using simulation studies Power system dynamics simulation studies were carried out for measuring the frequency transients. This frequency measurement is carried out through time measurement of a cycle (time between two zero crossing).2.5. Different values of df/dt are observed at different locations for the same loss of generation. 4.5 seconds in the system load base. Refer Annex-IV. it is inferred that relay adopts following measurement principles. Refer Annex-II Frequency recorded at 10-samples/second (i. The measurements static error depends on the quartz crystal’s absolute accuracy.1. 100 millisecond intervals) at Vindhyachal back to back station for both NR side bus as well as WR side bus was taken for analysis.1 Hz/sec was observed. The measured frequency value for df/dt calculation is averaged over three cycles. They operate when the frequency 6 . However. 3.On none of the occasion loss of generation up to 1500MW in the interconnected system. The biggest error in frequency measurement is in the range of +/. Result of the estimation based on loss of generation simulation shows that the inertia constant of the North-Central combined system is around 6.6 mHz considering the time resolution of 1 micro second.1 Hz/sec was observed. on one of the occasion of system split on 28th Feb 2007. 4. Refer Annex-I Result of the estimation based on loss of generation field data shows that the inertia constant of the North-Central combined system is around 5.3.6 to 9. Refer Annex-V: ABB document on Frequency relay behaviour. To restrain the influence of this component average of rate of change of frequency over 200 milliseconds was taken and from that inertia constant of the system was determined.8 to 10. df/dt Measurement principle by relays: Going through the technical information provided by various relay manufacturers.4 seconds in the system load base.e. Rate of change of frequency observed at NR and WR side bus of Vindhyachal are different for same incident of loss of generation. The accuracy of df/dt measurement depends on the accuracy of frequency measurements. and calculate the df/dt every 5ms.

Technical data sheet of AREVA manufactured MICOM relays indicates that for frequency supervised rate of change of frequency ‘f + df/dt’ protection accuracy is +/.08 Hz/sec and.075 Hz/sec is observed. for a 2000 MW loss it would be 0. whichever is greater. Initial findings and conclusion: 5.3.5 % or +/-55m Hz/s.11 Hz/sec 7 df dt = ∆P . The system inertia constant for the combined North-Central grid based on simulation studies and field data works out to about 5.On one occasion it has operated on 9th Feb 2007 when there were multiple trippings going on in the system due to fog and on another occasion it has operated on 28th Feb 2007 when Northern region system (Size about 18000 MW) got separated from central grid losing about 600 MW generation through tie line. However.0 can be taken for the finding out the frequency trend following a loss of generation.9 Hz /0. Without a significant loss of information in the result inertia constant value of 8. DTL has reported that df/dt relays with modified settings have operated on two occasions during the observation period . Refer Annex-VII 5. Based on this a 1500 MW loss at a frequency of 50 Hz in a 60.fo .1 Hz/sec can be achieved even with a loss of only 960 MW. From the response of df/dt relay set at 49.1 Hz per sec as decided in the first meeting of sub group it has been observed that these relay had not operated on any of the occasion of loss of generation listed in Annex-I.change exceeds the setting value 50 consecutive times. Analysis of 100 millisecond interval frequency received from Vindhyachal also indicates that on none of the occasion of loss of generation listed in Annex-I frequency trend of more than 0. The wide variation in system inertia may be due to daily variation in generation dispatch and system load.4 sec.93 secs) seems to be inaccurate since with this value of inertia constant frequency trend of 0.1. df/dt estimation for combined North-Central System.6 to 10. Refer Annex-VI: Toshiba Corporation document on Frequency and Voltage protection.000 MW system would result in 2H =50 x 1500/(60000-1500) 2x8 =0. 4. The system inertia constant value based on individual generator’s inertia constant as calculated above (3.

Further work is required to be done in the area of employing Phasor Measurement Units (PMUs) to detect loss of generation & activating load shedding in neighbouring nodes. 1 2 3 4 5 Contingency Loss of Talchar STPS (StageI & Stage II) Loss of Vindhyachal STPs Loss of Korba STPS Loss of Singrauli STPS Loss of Rihand STPS MW Loss 3000 3260 2100 2000 2000 Approximate load relief through such relay may be taken as 4000 MW spread over combined North Central system. • Recommended df/dt settings and further work: The combined North Central system could have a large number of df/dt relays at minimum setting of 0. The quantum would depend on net import by any zone/flow gate. Available df/dt relay settings: The minimum df/dt relay setting available is 0. 5.mHz (whichever is greater) is expected. 8 . • • Each region could have a df/dt setting at a higher value say 0. Based on the proximity of loss of generation some relays will pick up and some will not because df/dt will not be same everywhere. Some of the single largest contingency in the combined system are: Sl No.2.1 Hz/sec at 50 Hz /49.2 Hz/sec only.3.9 Hz based on single largest contingency.5 % or +/.1 Hz/sec Many df/dt relays have a minimum setting of 0. Further as mentioned in section – 4 error of +/.5.2 Hz /Sec and above as that would signify separation of system.

“Power System Stability and Control..] [6.pdf [4. 435-439 I. “Power System Control and Stability”. Ikeguchi. GRIDSAFE – 1995.1999. Ames. Nagpur. India http://arxiv. Tata MacGraw-Hill .New Delhi .] [7. Nagrath and D. C. Ramirez and C. K. P. Taniguchi.Z. Donnelly. pp. and A.1980. Estimation of WECC System Inertia Using Observed Frequency Transients Anderson. The Iowa State University Press. Nagrath and D. I. Hassler . H. J. 1997 . Chassin . Fouad. 1 .] [2. pp. 3rd edition. M. New Delhi . “Under Frequency Trend Relay as Power System Savior”. Yoshida. Tata MacGraw-Hill .6. P. Feb. (I). References/Bibliography [1.org/ftp/nlin/papers/0409/0409036. Ray . Inoue. “Estimation of power system Inertia Constant and Capacity of spinning reserve support generators using measured frequency transients”. P.] T.] [3. “Power System Engineering”. International Seminar on “Grid Stability and Load Management”. 136-143 D. and K. 12-14 January. Paul M. 1994 [5. Y. J. vol. 1996. 12 no. New York. “Modern Power System Analysis” .] I. 487 Dwarkanath. IEEE Transactions on Power Systems. McGraw-Hill.] 9 . Iowa.pp. Huang.A. E.E. 1977 Kundur. Kothari. P. Kothari.

10 .

1 2 System separated NR system separated from rest of the Central grid NR system separated from rest of the Central grid Date & Time 25th Feb 2007 (1634 hrs) 28th Feb 2007 (1311 hrs) Loss of generation (MW) 165 600 NR Grid Size (MW) 17500 18000 Df/dt (Hz/Sec) 0.5 4 Farraka.0 --9.03 0.Annex-1 Analysis of incident of loss of Generation S No . Flow on 400 kV Purnea-Muzaffarpur one circuit touched 1000 MW (other ckt out) Fault on 400 kV Biharsharif-Kahalgaon ckt-II fed from Biharsharif end for 2.4 HZ/sec (on NR Side) 11 .075 5. 49.014 HZ/sec (on NR Side) 0.067 --- 5 7th Feb 2007 (0610 hrs) 1200 49.03 0.71 0.6. 6.97 0. Auxiliary supply failure.19 48.37 0.86 48.06.5 minutes. Kahalgaon & Tenughat Anpara 6th Jan 2007 (1610 hrs) 1750 48. 0. 1 2 3 Station Date & Time Generation Loss in MW Remarks Freq prior to incident (Hz) Df/dt (Hz/Sec) Observed at Vindhyachal North Bus H (Sec) Suratgarh Singrauli Anpara 4th Oct 2006 (1822 hrs) 13th Nov 2006 (1432 hrs) 15th Nov 2006 (1121 hrs) 1250 (750 + 500) 800 1450 600 MW load throw off in Rajasthan CW pump failure Fault on 132 kV Pipri bus.03 --- Analysis of incident of System separation S No.

This page has been intentionally left blank. 12 .

65 48. Units Tripping at Anpara (1450 MW Loss) on 15-11-2006 at 1119 hrs 48.06 Hz /Sec (NR Side) 0.6to 6.13 49.027 Hz /Sec (WR Side) Hz 49.15 49.85 48.70 48.75 0. Loss) on 04-10-2006 at 1822 hrs 49.60 H= 9.08 Hz/Sec 48.55 11:17:45 11:17:51 11:17:57 11:18:03 11:18:09 11:18:1 Time Frequency Profile during Units tripping at Suratgarh (1200 MW Gen.19 49.05 0.10 49.80 WR Frequency 48.0 Sec Based on NR Side Frequency of Vindhyachal 49.25 49.71 NR Frequency Hz 48.03 Hz /Sec (NR Side) 49.71 48.16 0.04 Hz /Sec (WR Side) 49.5 Sec Based on NR Side Frequency of Vindhyachal 0.09 H=5.20 0.00 18:23:00 18:23:06 18:23:12 18:23:18 18:23:24 18:23:30 18:23:36 18:23:42 18:23:48 18:23:54 18:24:00 Time 13 .Annex-II Observed Frequency response at Vindhyachal following loss of generation incident.68 48.04 Hz/Sec 48.67 48.

This page has been intentionally left blank. 14 .

5 51 50.200 0.600 13:10:40 13:10:46 13:10:52 13:10:58 13:11:04 13:11:10 13:11:16 15 .600 0.5 47 13:10:40 13:10:46 13:10:52 13:10:58 13:11:04 13:11:10 13:11:16 0.5 WR Frequency Hz 50 49.5 48 47.200 -0.000 -0. Frequency profile during System separation on 28-02-2007 at 1311 hrs 52.4 Hz/Sec NR Frequency Time df/dt profile during System Separation On 28-02-2007 at 1311 hrs during system separation 0.5 52 51.400 -0.Annex-III Observed frequency response at Vindhyachal following system separation.5 49 48.400 0.

16 .This page has been intentionally left blank.

00 4.5 -1.00 3.50 Average df/dt (over 200 ms) Rihand 0.00 2.8) NJPC 0.10 0.0 0.00 1.1 Frequency (Hz) 50.50 5.00 Time (sec) Average df/dt (over 200 ms) Rihand 0.00 0.8 49.00 Time (Sec) df/dt (Rihand-1800 MW loss of Generation) df/dt (NJPC-1250 MW Loss of Generation) 17 .00 1.5 df/dt (Hz/sec ) 1.5 0.10 0.045 Hz/s (H=10.Annex-IV Frequency response obtained through simulation studies.50 1.7 49.0 49. Frequency Profile observed through simulation of loss of generation 50.00 0.00 2.4) 3.4) Freq (Rihand-1800 MW loss of Generation) Freq (NJPC-1250 MW Loss of Generation) df/dt observed through simulation of loss of generation 1.8) NJPC 0.00 4.50 4.096Hz /s (H=6.50 1.50 3.045 Hz/s (H=10.50 2.0 0.9 49.6 49.0 -0.50 5.05 0.05 0.50 4.5 0.00 3.50 2.096Hz /s (H=6.

Jaipur through simulation for loss of 1250 MW generation at Jhakri.Jaipur through simulation for loss of 1800 MW generation at Singrauli. Korba.Tala. 18 . Comparison of fequency response observed at Singrauli. Korba.Comparison of fequency response observed at Singrauli.Tala.

19 .Annex-V ABB document on Frequency relay behavior.

.

.

.

.

.

.

.

.

21 .Annex-VI Extract from Toshiba Corporation document on Frequency and Voltage protection.

r'".. GRF100 ..: . VOLTAGE RELAY. ( Var..0) .***8 f .-."... 2..TOSHIBA 6F2S081 I . "o ~~.".'" ". ~ J L TOSHIBA CORPORATION @ TOSHIBA Corporation 2005 All Rights Reserved.. INSTRUCTION MANUAL FREQUENCY. .. ...:iI!!lJitIJIIi!l. !.

~ DANGER I Indicates an inuninently bazardous situation which will result in death or serious injury if you do not follow the instructions.CAUTION CAUTION -1- . A WARNINGI . ...i:'~~:. may result in minor injury I)f'moderate itiJuiY: Indicates a potentially hazardous situation which if not avoided." ~.:_- . tbis chapter must be thoroughly read and understood.::':.'. This chapter describes tbe safety precautions recommended when using the GRFlOO. . Indicates a potentially hu.. pIrase read this chapter carefully.'~. Explanation of symbols used Signal words such as DANGER. and two kinds of CAUTION.' . ' .. """'~_"IR""'M"".. '\ TOSHIBA 6F2S0816 S~fetyPrec~utions Before using this product.. WARNING..'.ardous situation which could rtSult in death or serious injury if you do nol follow the instructions. .A. may result in property damage.ardous situation which if not avoided.".:~rY.. Indicates a potentially ha7. will be followed by important safety information that must be carefully reviewed. "L:!::.'.. Before installing and using the equiPmeoI.

2.4I1('.'}' op~tc times.1 Frequency nrte-of-change element Underfrequency and O~ertrequency Element - The frequency I"8te-of-cl\ange clement calculates thegradicnt of frequency change (df/dt). exceeds the setting value 50 consecutive times.1 shows cl1ari1~te. Th~y openlt!"..' IInti(.5. a fre-queocy d«ay rate ele-mt'nt and II fT~uency rise rateelement. .ncy chllnet'. ~- Uncfetfrequency Ind ovetfrequency element Undcr&cqucne)' ~Icmcnt UP operates when II power S)'Stcm fi"cqucney stll)'5 under the settin value. .ry~ m..5." c.5.vt'. l' Hz . Overfrtqumcy ~Iemeot OF Opmttes wh~n a power systnn hqucocy stays over tM s~tting valu~. freqllt'.n .change Element Figure 2. T'h.5.\1 .... 8S . i.wht'.17."" ...ht'.. .hown Fi~. 1 ..1I11I't'.\1 -. GRFJ 00 provi~ f'. I Figufe. . These clementsmeasurethe &equencyand check the undet&cqucncyor ovcrUequency every5 MS.) Hzt OF setting UF setting I OF 1--1------UF 0 UVBLK setting t v ~ Figure 2. These elements measure tbe cbaoge in ftequcncy ('" f) over a time interval ( ~t= lOOms).ht'. 'I ~C ~ Frequency Rate-of.2 .1 Frequency Elements 6F2S0816 .5 Characteristics of Measuring Elements ~ 2. fl.r t~ 'TOSHIBA 2.fisti~s of{JF i11\dOF elements. wben the und~ucnc)' or overm:qucucycondition is detected 16 consecutive Borh lIF end OF dmJe1lt~ output is invalidate'dby undervoltag~ block elemmt (UVBlK) operation dwing uodcrvoltagecondition..vQf'l!te-of-<"hangeelemmts.

UVIS and UVIG: t OVIS and OVIG: t where.3 Undervoltage (b) Overvollage elements and Overvoltage Elements ~ 18-- I I ..CVNs) x TMS .5.. I I = (VNs)1 . voltage setting TMS '" time multipliC'-f~tting Figure 2.TOSHIBA 2. lIVIa.5.3 shows characteristics ofundervoltage clements and overvoltage elements...'{pr~~C'd by tbe following equations.4 shows operating time characteristics ofUVIS.. I J I .1 x TMS. PhasC'-to-ph8s~ OVe1'VOltflg~ C'1m1C11tS OVS and OVIS find phas~-to-ground o. UVHSS.ltAg~ ~I~m~nt Th~y ~VM"Y 5 m~ operate when tbe undcrvoltage or ovcrvoltage condition is detected 2 consecutive times..wer system voltage stays over the setting values. Figure 2. OVIS and °YIG / v - 'T0I '\ ) V 1_. OVIS find OVIO hav~ invC'rsC'timC'characteristics.5. 1'hC'!:C' C'1~t1l mC'ft~1I1"l"th~ V""Ag~ And chC'Ck thC' lmdl"f'VoltagC' Qf OV~. I V = power $ystem voltAge Vs . UVHSG :md UVIG operoie when 0 power s)'Stemvolwge s~j's under the seNing values. The UVIS. UVIS and phase-to-ground undervoltage elements UVG./ (a) Undervollage elements Flgur':! 2. find thC'ir charactC'ri~tk~ !!rC'C'.5.~olt8g~ avo and OVIO operate when 8 JX>. t 0:opcmting time = 1 1 . UVIG.2 Undervoltageand Overvoltage Elem~nts 6F2S0816 Pbase-to-pbase uodervoltage clements UVS.

23 .Annex-VII Extract from Technical data sheet of AREVA manufactured MICOM relays.

1":r~' ': 'c.fA AREVA MiCOM P941. .::.}:c.""':::~. .. ~ "...~ . .(".):{i.. . .tf. .~"'h!Jt1 '1.P943 Frequency Based Load Shedding Rela'ys Technical Data Sheet P94x1EN T051 A11 Date: Hardware Suffix: Software Version: Connection Diagrams: 20th January A or C 10 or 11 10P94yxx (y 1 or 3) (x =01 to 06) 2006 = '.."".'I." .~I:DHfj.

8Srelevant) and no intentional time delay.1 Hz/s (DfiDt threshold)" Drop-off: Setting +20mHz. :t5% or :t55mHz/s. A v:Cyc/es senlng J timedelay.'.. 1Hz) ~I.47 (for Df setting.O5Hz/s..Cycles= 0 p~ I f'V4JSuI.1Hz/s) SetUng +O. <400ms for df/dt. Dt.Av.3x setting or greater Reset time"': <250ms e Fs/Ff ratios as stated. whichever is greater (for settings greater than 0. :i:10mHz(underfrequency) Setting -20mHz.Av.. whichever Is greater (for settings greater than O.Av... :t:10mHzls (for settings between 0. :t:5%or :t55mHz/s. "':":.156xDf + 0.23 (for Of setting <: 1Hz) Dt > 0. . settlng~greater than 0.1Hz/s) Drop-off (rising frequency): Setting -O.005Hz/s.01 Hz/s and 0. :t:10mHzJs Operating & Reset Time Operating timee (underfrequency): <100ms for FslFf ratios less than 2 <160ms for FslFt ralios less than 6 <230ms otherwise -. ""~'".1 HzJs) Average Rate of Change of Frequency 'f+Df/Dt' Protection: [81RAV] I Pick-up: Setting:t:10mHz (frequency threshold) Setting to.1 Hz/s) Setting -O.. :t10mHz (falling frequency) Setting -20mHz.005Hz/s. :10mHz (tmderfrequency) Setting -20mHz.. :t10mHz/s (for settings between 0. rising frequency): Setting -0.1a $1\01.Cycles setting = 0 Independent Rate of Change of Frequency 'df/dt+t' Protection [81R] Accuracy Pick-up: Setting:3% or :10mHzls.01 Hz/s and 0.1HzJs) Operating t!mer: %2%or QOms whichever is greater Operating & Reset Time Operating timee: <200ms for ramps 2x setting or greater <300ms for ramps 1. :t10mHz/s (for settings between 0.Cyc/es setting = O.frequency = setting Ff =frequency setting .1 Hz/s) Drop-off (df/dt threshold.005Hz/s. . 0 and no Intentional (for settings between 0.O5Hzls.whichever is greater Drop-off (tailing frequency): Setting +0. :t10mHz (overfrequency) Operating timer: %2%or 50ms whichever is greater Operating & Raiet Time Instantaneous operating time: <125ms for Freq.::..IlI:r .1 HZ/s) Setting +0.Cyclessetting = 0 Re~et time: ._9J>~rating timee (overfrequency): <125ms tor FslFt ratios less than 2 <150ms tor FslFt ratios less than 30 <200ms othe(Wise Reset time-: <200ms . where: Fs start frequency .>. falling frequency): Setting +0.v5Hz/s.end frequency Frequency Supervised Rate of Change of Frequency 'f+df/dt' Protection [81RF) Accuracy Pick-up: Reference conditions: Tested with df/dt. :t10mHz (rising frequency) Operating timer: :t2% or 30ms whichever is greater e Reference conditions:Tomaintain accuracy the minimum time delay set/mg...01Hz/s and 0. .:5% or :t:55mHzJs.1Hzls (positiveor negative.~v. -.O5Hz/s.whichever is greater (df/dt threshold) Drop-off (frequency threshold): Setting +20mHz.'" '.375xDf + 0..:' . :t10mHz (overfrequency) Drop-off (df/dt threshold. should be: Dt > 0. Ve'""n II.Reference conditions: Tested using step changes in frequency with Freq.. :t5% or :t55mHz/s. P'O4) "-0.' Operating Time Typically<125ms withFreq.1 Hz/s) Setting '-0. whichever is greater (for settings greater than 0.01 HzJs and 0. .Protection Functions Over/Under Frequency 'f+t' Protection [81U/810] Accuracy Pick-up: Setting:t10mHz Drop-off: Setting +20mHz. Setting:t10mHz (frequency threshold) Setting i3% or :!:10mHz/s. whichever is greater (for settings greater than 0.O05Hz/s. for df/dt .