[NER]Fault Reduction Strategy by NER

Session Twelve: Fault Reduction Strategy using Neutral Earth Resistor

Electrical Safety & Power System Protection Forum 1

Session Twelve:

Fault Reduction Strategy using Neutral Earth Resistor (NER) Installation

Cheng Lee

Lead Engineer, Peracon

Co-author : Frans Cloete, Peracon

Abstract

Fault level reduction is an ongoing challenge in a growing electrical network. With generation and network capacity continually being added to the system, the fault levels in various locations of a distribution network will soon approach their maximum allowable design level. For electrical power distribution companies there are a number of business drivers that force the implementation of fault level reduction schemes.

While the general operation of Neutral Earthing Resistors (NER) for fault level reduction is well understood, optimisation of the design under various network configurations to yield maximum benefit requires detailed study. This paper presents the studies carried out in assessing the impact of Neutral Earthing Impedances of varying sizes for phase-ground faults on a “generic” network with three different configurations:

Delta-Star Transformer configuration;

Star-Star Transformer configuration; and

Bus Tie CB Open (Delta-star Transformer configuration)

The impact of Transient Recovery Overvoltage (TRV) on selected zone substation configuration resulting from the installation of NERs is also presented.

1. INTRODUCTION

Fault level reduction is an ongoing task on a growing electrical network. With increasing generation and network capacity added to the system, the fault levels in various locations of a distribution network will approach their maximum allowable design level.

For electrical power distribution companies, there are a number of business drivers that result in the implementation of fault level reduction schemes. These business drivers are: Maintain the existing equipment within design ratings to avoid expensive

upgrades. Bushfire risk mitigation. Quality of Supply Improvement. Safety and reliability.

In particular, the quality of supply improvement generates a key driver for the electrically sensitive LV industrial and commercial customers, and the HV customers with step down transformers.



The present 66/22kV systems under studied are solidly earthed (i.e. without NER installed), which results in higher earth fault currents under fault condition. With the expanding network, the situation is quickly worsened because the existing switchgear and other equipment are not rated for the higher fault level.

These are shown graphically and mathematically as follows. Without the NER, it can be seen that the Ifault will be naturally high.

A commonly accepted and cost efficient approach to reduce the single phase to ground fault level is to install a neutral earthing resistor on the transformer neutral (as shown below).

During the fault, the NER will form part of the positive, negative and zero sequence circuits, providing a mean to control the Ifault, consequently a higher NER value will result in a lower fault current. The simplified sequence circuits for the symmetrical components to represent the unbalance condition under single phase to ground fault are demonstrated graphically as follows:

NER

Vs

Earth Potential Rise

22kV 66kV

RL

NER

Re2 (Earth Resistance) Re1 (Earth Resistance)

Ifault

22kV 66kV

NER

Positive Sequence I1 Z+ V1

Negative Sequence I2 Z- V2

Zero Sequence I0

Z0 V0

I1 = I2 = I0



For sequence currents,

For phase currents,

where Z0 = 3 x ZNER

While the general operation of Neutral Earthing Resistors for fault level reduction is well understood, optimisation of the design under various network configurations to yield maximum benefit requires detailed study. The studies involved assessing the impact of Neutral Earthing Impedances of varying sizes for phase-ground faults on a “generic” network with three different configurations:

Delta-Star Transformer configuration; Star-Star Transformer configuration; and Bus Tie CB Open (Delta-star Transformer configuration)

The impact on Transient Recovery Overvoltage (TRV) resulting from the installation of NERs has also been studied to help determine eventual NER selection.

As the use of NERs to reduce the fault level is a widely accepted approach in the electrical industry. This paper aims to provide an insight into the NER effect on the network and to present the outcome of the study and simulation results.

2. METHODOLOGY The modeling and computation associated with fault and Temporary Overvoltage (TOV) calculation was carried out using the load flow software. The work associated with Transient Recovery Overvoltage (TRV) was carried out using ATP software, a widely used EMTP software for the analysis of electrical transient phenomenon.

The study has been performed by modeling the generic network (see Figure 4) in the load flow program, and modifying it to reflect the required transformer configuration and additional bus tie circuit breakers. Load flow studies were then conducted for each configuration. The NER fault studies (for the range of NER values) were conducted according to IEC methodology for phase-ground faults at selected locations. The voltages and fault currents were recorded and tabulated for various transformer configurations with the 22kV bus tie CB both open and closed. From these results, voltage rises on healthy phases for each study were calculated and graphed at both the 22kV zone substation bus level and at the 415V LV customer level for each transformer configuration. The effects of the



NER on the LV customer level within the generic network were investigated because the impact on customer voltages is one of the major benefits of neutral earthed impedance installations. It should be noted that all voltages specified herein are phase to ground voltages. When the neutral earthing resistor value is zero, the phase to neutral voltage is equivalent to the phase to ground voltage, i.e. solidly earthed. The transient recovery overvoltage (TRV) was separately investigated using ATP. The TRV results simulated based on the ATP models were compared with accepted TRV ratings for 24kV circuit breaker specified in the Australian Standard (AS-62271-100), both in terms of the levels of TRV and the rate of rise of the recovery voltage. Figure 1 shows a typical response of the phase voltages following the clearing of a phase to ground fault.

Figure 1 Typical phase voltage response during fault clearance The ratings of relevance in this study are the TRV rating and the short-time power frequency-withstand voltage rating of the 22kV circuit breaker. The first refers to the peak TRV value the circuit breaker can tolerate during mechanical breaking and the second is relevant to the TOV, i.e. the breaker’s tolerance to withstand phase voltage displacement during the fault.

(f ile STA_020805_R2_w TERT.pl4; x-var t) v:SECA v:SECB v:SECC

0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40[s]

-40

-20

0

20

40

60

[kV]

Typical ATP Output

TRV generated when clearing fault TOV during fault

conditions



3. VOLTAGE DISPLACEMENT CAUSED BY NER In the event of a single phase to earth fault, the fault current flows through the star-point of a delta-star connected transformer and then through the NER as shown in Figure 2, causing a rise in the voltage of the neutral point.

Figure 2 Flow of Fault Current through NER As a result, the neutral phase voltage is displaced and causes other phases (relative to it) to shift as well. This is also the fundamental cause of temporary overvoltage (TOV). See Figure 3. The unbalance voltage also provides a means of detecting earth faults and is a technique many relays employ.

Figure 3 Vector Diagram - Demonstration of Voltage Displacement

4. KEY FINDINGS FROM THE GENERIC NETWORK STUDY A “generic” network as shown in

Figure 4 was modified to include a 22kV/415V delta-star to ground transformer connected at an arbitrary location on Feeder 2. Simulation studies were then carried out to observe the effect of phase to ground faults and neutral earthing impedances at the low voltage 415V customer level as it was identified that quality of supply is one of the key business drivers, particularly for industrial and commercial customers. The key findings related to the impact of the NER with different transformer configurations, bus tie CB states, and fault types were as follows.

Regardless of transformer configuration, opening the bus tie CB reduces the impact of all types of faults but reliability and quality of supply issues remain for feeders on the affected bus section.

NERs have no impact on phase to phase type fault mitigation but it would help to reduce phase to phase faults developing from phase to ground fault.

NERs maintain voltage supply at the 415V level but do cause 22kV phase voltages to rise. However, NERs do not have any reliability or voltage depression issues associated with opening the bus tie CB.

At the 415V low voltage customer level, neutral earthing resistors would continue to assist distribution companies to meet their fault level mitigation

66kV (Pri)

a

c

b

Voltage displacement resulting from NER

TOV

22kV (Sec)

NER

If

Displaced Voltage



and quality of supply objectives regardless of 66/22kV transformer configurations and bus tie CB status.

Figure 4 A Generic Distribution Network

4.1 Study Results With the 8Ω NER installed in the transformer, it is observed that at the customer (415V) level, there is almost no neutral displacement of the two healthy phase voltages and an improved voltage on the faulted phase. Thus, a 22kV fault anywhere on Feeder 1 would not cause any noticeable disruption to LV customers on other feeders. The study results are shown in Figure 5 and Figure 6.

NER

66kV Bus

66kV Bus

22kV Bus

To LV (415V) customer

To LV (415) customer

Normally open CB

220kV Bus



Figure 5 LV Voltage Results with 8Ω NER

Figure 6 22kV Voltage Results with 8Ω NER On a network with no neutral earthing resistors, and for faults along Feeder 1, only the faulted phase is displaced at the 22kV bus level while the two healthy phases remain fairly constant, as shown in Figure 8.

Figure 7 LV Voltage Results with 0Ω NER

Figure 8 22kV Voltage Results with 0Ω NER At the LV (415V) bus level, however, two phases (Va & Vc) experience some displacement while the other healthy Vb is relatively constant at close to 1 pu (Figure 7). In fact, the voltage displacement compared with the healthy phase is as high as 36% for a fault directly on the 22kV bus. As such, a fault on Feeder 1 very close to the 22kV bus would cause some noticeable disruption for customers on other feeders, quantifying as a voltage phase failure and can



fragment supply to customers. This is eliminated with the inclusion of a neutral earthed resistor. The study showed that the correction and improvement a NER of 8Ω provides at the 415V customer level is quite significant, regardless of the 66/22kV transformer configuration. NERs maintain quality of supply by allowing voltages to be sustained on all three phases at LV during fault condition. The phase voltages are steady at approximately 1pu. This is one of the key benefits of using neutral earthing impedances. NERs provide protection for all non-faulted feeders from phase to ground faults and their effects are comparable to opening the bus tie CB for phase to ground type fault mitigation. However, NERs do not have the reliability and voltage loss issues associated with opening the bus tie CB. Rather, they help maintain quality of supply at the LV 415V level at the expense of an increase on 22kV phase voltages. NERs are effective in reducing the impact of all phase to ground faults, which represent approximately 90% of all faults on the distribution network. NERs have no impact on phase to phase faults (representing approximately 10% of all faults) in the system under study and are also relatively expensive compared to opening the bus tie CB. Different NER values, including NEX which has similar effect in reducing fault levels are investigated in later sections. 4.2 Discussions As illustrated in the diagrams below, the zero-sequence currents will flow in both primary and secondary circuits in the case of star-star connection. For delta-star, the zero-sequence currents will flow in the secondary circuit. It was found that the 66/22kV transformer configurations have more minor and a secondary impact on fault current mitigation, this outcome is not usually expected because the configuration of source transformers (66/22kV) does impact the fault currents. The reason for this minor impact is likely due to the range of NER values being investigated.

Z0

Z0

Transformer Connection Zero Sequence Representation



But NERs do cause 22kV voltage rise for networks with bus tie CB closed, for this reason, many utilities are still maintaining the solidly earthed approach. Opening bus tie CBs offer a cost-effective means for mitigation of all fault types. The drawback, though, is voltage depression on feeders connected to affected bus section due to a weaker system. On a network with a 8Ω NER, the phase voltage at the 415V customer level is very steady at approximately 1.0 pu. This differs from the voltage displacement observed at the 22kV bus voltage level where there is quite some disparity between the phase voltages. 5. TRANSIENT RECOVERY VOLTAGE (TRV) 5.1 General The TRV is the voltage occurring at the circuit breaker contacts on the faulted phase at the moment of opening the circuit breaker to clear the fault. The installation of neutral earth impedance could potentially increase the Transient Recovery Voltage, causing the circuit breaker to re-strike and fail to interrupt fault currents. The maximum TRV is generated when clearing ‘close-in’ phase to ground fault at 22kV network. Technical studies using the specialist Electro-magnetic Transient Program (ATP) were carried out to examine the TRV behaviours under selected values of neutral earth impedances installed on the 66/22kV transformers at three specific zone substations in the distribution network. For each zone substation, we examined the installation of the following four (4) different neutral earthing impedance values:

Solidly earthed (i.e. zero neutral impedance in the neutral of the 22kV transformer);

1Ω reactor in each 22kV transformer neutral;

2 resistor in each 22kV transformer neutral; and

8 common neutral earthing resistor in 22kV transformer neutral. The ATP models are constructed based on detailed network information and are properly validated. It should be noted that unless otherwise specified, all voltage waveforms shown in the figures represent line-to-ground voltages. For example, the nominal peak value for the 66 kV voltage level is 54 kV ( kV 543/266 ) and the nominal peak

value for the 22 kV voltage level is 18 kV ( kV 183/222 ).

The EMTP simulation is usually performed on a small network size (reduced system). Working with the reduced system is technically justified since the electromagnetic transients generated in the power system due to either component switching (e.g. cable, capacitors, and lines) or lightning and faults are much localized to the source of disturbance.



The reduced system model is mainly associated with finding suitable equivalent impedance (or fault level) used to represent the system outside the study area. The system outside the study area is normally represented by impedance equivalent to the available short circuit level at a point distant from the disturbance. The single line diagrams of the zone substations and fault levels at the relevant points of the network under study have been provided. To accurately capture the electromagnetic transient response, all the relevant power system components were modeled. As all circuit breakers under investigation operate based on 3 pole operation, it is appropriate to check both the level of TRV and TOV because while the faulted phase is experiencing the TRV, the other two healthy phases are experiencing TOV that could potentially present a risk of re-strike, even though such risk is minimal. Additionally, during all simulations, due to the time dependent nature of transient studies, faults were applied at 0.001 second incremental time intervals along the sine wave to find the point of worst case (i.e. producing maximum TRV level) scenario.

Figure 9 Determination of Circuit Breaker Opening Time The TRV sensitivity with respect to the circuit breaker opening time has been analysed and it was found that the point of worst case scenario is identified as close to the 200ms (or 210 ms) in the simulation cycle of 20ms (see Figure 9). This corresponds to the peak on the voltage waveform (90° or 270°). As such, the simulation of the circuit breaker opening time is targeted at about 200ms (90°) to obtain a more conservative result.

TRV Sensitivity Versus Circuit Breaker Opening Time

0.00

0.20

0.40

0.60

0.80

1.00

1.20

200 205 210 215 220

Breaker Opening Time (ms)

TR

V S

en

sit

vit

y F

acto

r (p

.u)

Circuit Breaker Opening Time



5.2 Model Validation In Figure 10, the green curve represents the equivalent voltage displacement (or TOV) conducted in steady-state load flow program. The blue TOV curve was generated using an electromagnetic transient program (ATP) used for the TRV studies.

Figure 10 Comparison of TOV results

It can be shown that the two TOV curves are very comparable in both magnitude and trend. The slight discrepancy is likely due to the use of generic model in the load flow program. 5.3 Simulated Results Figure 11 to Figure 14 show the time domain response of the individual phase voltage following the clearing of 100ms single phase fault, simulated for 0Ω, 1Ω, 2Ω and 8Ω Neutral Earthed Impedance.

Figure 11 Solidly Earthed 5.4 Discussions The temporary overvoltage (TOV) takes place on the two healthy phases during the occurrence of phase-to-ground fault. There are two components associated with the TOV – the transient surge component and the fundamental frequency component. The transient surge component attenuates with a time

NER vs Voltage Levels

0.800

1.000

1.200

1.400

1.600

1.800

2.000

2.200

2.400

2.600

2.800

3.000

3.200

0 1 2 3 4 5 6 7 8

NER Values (Ohms)

Pe

ak

Vo

lta

ge

Le

ve

ls (

p.u

.)

TOV-NER-PU

TRV-NER-PU

TOV-NER-S2-PU



constant that depends on the circuit damping and the fundamental frequency component will remain until the fault is cleared.

Figure 12 NEX = 1Ω

Figure 13 NER = 2Ω

Figure 14 NER = 8Ω (Common)

The TRV curve (orange line) in Figure 10 shows that the peak TRV levels occur

with NERs between 2 to 3 for one of the zone substations investigated. However, this can vary slightly from one zone substation to another because it is found that the behaviours of TRV are site specific and are dependent on the transformer impedances, line impedances, and feeder impedances.



We have assessed all TOV values corresponding to each TRV case simulated and the results are presented for each zone substation under study. The study found no TOV levels that are of any significant concern and the cases investigated are generally in compliance with the Australia Standard. The TRV results of the EMTP studies conducted for the three separately located zone substations, identified as Substation A, Substation B and Substation C are summarised in Table 1.

Table 1 Summary of maximum TRV generated when clearing ‘close-in’ phase

to ground fault at 22kV network The one ohm neutral earthing reactors will neither provide effective limitation of earth fault current nor provide adequate benefits to quality of supply at the 415V level. Furthermore, in a zone substation with capacitor bank installations, neutral earthing reactors could increase the risk of resonance situations that may be difficult to predict. Based on the simulation results, there is evidence in the case of reactor neutral grounding that multiple low to medium frequency oscillation superimposed on the 50Hz system frequency. The resonance condition, particularly due to interactions with local loads was not modelled in this report due to the lack of information of the other harmonic current sources. Unlike the neutral earthed resistor, the neutral earthed reactor appeared to have fluctuating levels of TRV depending on the values selected and it does not follow a constant trend. This fluctuation can be attributed to the frequency dependant nature of TRV, that vary based on each reactor value and their interactions with one or all of the zone substation’s capacitor banks, the line inductance, transformer impedances, and feeder configurations. It is evident from the transient simulation that an 8Ω common NER provides an effective natural damping to the high oscillation frequency components of TRV and therefore presents the lowest TRV level across all the three zone substations studied. In addition to lower TRV levels, the 8Ω NER provides additional technical benefits in terms of improved voltage quality and supply at 415V LV level, as discussed in the previous sections. Although the temporary overvoltage (TOV) appears to be higher1, it is not considered a threat because the TOV level is within the circuit breaker TOV rating.

1 TOV observed is 1.87pu for Substation A, 1.89pu for Substation B, and 1.79pu for Substation C

NEI Solidly Earthed NEX=1Ω NER=2Ω NER=8Ω (common)

TRV

peak

value

Within

41kv

Limit

TRV

peak

value

Within

41kv

Limit

TRV

peak

value

Within

41kv

Limit

TRV

peak

value

Within

41kv

Limit

SA 28.94kV 27.36kV 51.68kV 21.49kV

MLN 31.53kV 31.73kV 42.85kV 24.94kV

LVN 32.55kV 32.89kV 29.50kV 26.34kV

Zone

Sub



The Australian Standards (AS 62271.100) for 24kV circuit breakers have the following TRV rating - TRV peak value of 41kV and time of 87µs. This is the standard by which the results within these studies were compared against. The current set of 22kV bus circuit breakers are rated within the Australian Standards specifications.

It is observed that all solidly earthed, 1 NEX and 8 NER (common) have shown that the TRV values measured have response time within the Australian

Standards. Only some of the 2 NER cases have TRV response times (as well as maximum TRV value) exceeding the Australian Standards. 6. CONCLUSIONS In conclusion, this study resulted in the following key observations:

the use of a 8Ω common NER is favourable because it is evident from the transient simulation that it provides an effective natural damping to the high oscillation frequency components of TRV and therefore presents the lowest TRV level across all the three zone substations studied. In addition to lower TRV levels, the 8Ω common NER provides additional consumer benefits in terms of improved voltage quality and supply at 415V LV level. Although the temporary overvoltage (TOV) appears to be higher, it is not considered a threat because the TOV level is within the circuit breaker TOV rating.

the 1Ω neutral earthing reactors will neither provide effective limitation of earth fault current nor provide adequate benefits to quality of supply at the 415V level. Furthermore, it should be noted that in a zone substation with capacitor bank installations, neutral earthing reactors could increase the risk of resonance situations that may be difficult to predict.

The installation of 2Ω neutral earth reactor will result in high TRV levels that are above the rating of the existing 24kV circuit breakers. Although substation C’s TRV levels are considered satisfactory with the existing zone substation configuration, the future installation of capacitor banks (2 X 12MVAr) will cause the TRV levels to rise above the acceptable rating. Therefore, the use of 2Ω NER is not recommended.

7. REFERENCE

D. Mc Nabb and al., June 2001, 'Transient Design Studies for the TransMantaro Series-Compensated Transmission System', IPST'01 Proceedings, Rio de Janeiro.

D. Braun and G. Koeppl, 2003, ‘Transient Recovery Voltages During the Switching Under Out-of Phase Conditions’, International Conference on Power System Transients IPST , New Orleans, USA.

http://www.emtp.com/services/pdf/3.pdf

http://www.emtp.com/services/pdf/3.pdf



8. APPENDICES Definition of Terms TOV – Temporary Overvoltage at Fundamental Frequency TRV – Transient Recovery Overvoltage NER – Neutral Earthing Resistor NEX – Neutral Earthing Reactor NEI – Neutral Earthing Impedance EMTP – Electromagnetic Transient Program CB – Circuit Breaker LV – Low Voltage

[NER]Fault Reduction Strategy by NER

Documents

Transcript of [NER]Fault Reduction Strategy by NER