WO2012037947A1

WO2012037947A1 - Method and system for fault detection on an electrical power transmission line

Info

Publication number: WO2012037947A1
Application number: PCT/EP2010/005742
Authority: WO
Inventors: Amr El-Hadidy; Christian Rehtanz
Original assignee: Technische Universität Dortmund
Priority date: 2010-09-20
Filing date: 2010-09-20
Publication date: 2012-03-29

Abstract

The inventions relates to a method and a system for detection of a fault, and in particular even high impedance faults, on at least one section (3) of an electrical power transmission line (2) and blocking of undesirable relay trips, in particular when load encroachment, power swings, or voltage instability occur in the system, the method comprising the steps of - measuring the voltage (V_A) and the current (I_A) at a first terminal (A) of the transmission line (2), the first terminal (A) being a power sending end of the section (3) of the transmission line (2), - measuring the voltage (V_B) and the current (I_B) at a second terminal (B) of the transmission line (2), the second terminal (B) being a power receiving end of the section (3) of the transmission line (2), - calculation of a fault resistance (R_f) to ground between the first terminal (A) and the second terminal (B) on the basis of the measured voltages (V_A, V_B), currents (I_A, I_B) and the impedance (Z_LT) between the first terminal (A) and the second terminal (B), and - identification of a fault on the basis of the calculated value of the fault resistance (R_f).

Description

Method and system for fault detection on an electrical power transmission line

The invention relates to a method and a system for detection of a fault on at least one section of an electrical power transmission line. The method is not affected by load encroachment, power swings, voltage instability and High Impedance Faults (HIF).

The technical field of the invention is in the area of protection of electric energy grids. The method and system according to the invention is a relaying concept that is expected to overcome many of the problems that face the setting and operation of distance protection like occurrence of: power swings, voltage instability, load encroachment, High Impedance Faults (HIF), over-reach and under-reach.

A High-impedance fault (HIF) is an undesirable electrical contact between a bare energized conductor and a non-conducting foreign object that has, due to its material, a high impedance. For example, a HIF occurs when a conducting overhead line physically breaks and falls on an non-conducting element like grass, sand or an asphalted road.

These problems were the reason for many of the previous power system blackouts worldwide.

To protect an electrical power system against faults, current-operating relays have been used as the main protection for a long time. This type of relays operates based on measuring the value of the current in a certain part of the system. The relay issues its trip command if the value of the electric current exceeds a certain threshold. However, the operation of such relays has shown to be unsatisfactory because they are dependent on the level of the fault current which caused the relays to lack sensitivity, selectivity and makes the setting of the relay depend on the system configuration. Moreover, these relays are affected by the occurrence of oscillations in the system which might lead to mal-tripping of the relay and affected also by the occurrence of HIF.

To improve the operation of power system protection a better operating principle was developed which is based on directly measuring the impedance to fault and not the electric current value. This operating theory has been called the "distance relaying". Application of this theory has shown better system performance with regard to the issues of sensitivity and selectivity. However, a problem like HIF could affect the selectivity of the distance relay. Other phenomena, so called virtual faults, for example power swings, voltage instability, load encroachment could cause the distance relay to mal-operate. Said phenomena are referred to as virtual faults because they are no faults in classical meaning but they appear to the relay as if there is a fault in the system. So, they are virtual from the relay's view point. This could lead to further deterioration of the system state and the occurrence of cascade tripping and finally collapse or blackout of the whole system as has been shown in previous worldwide blackout incidences.

According to the application of distance relays, the distance relay should have three zones, a first zone (zone 1), a second zone (zone 2) and a third zone (zone 3). Zone 1 is designed to protect 80% of the length of the transmission line at time delay of zero seconds. This means that if a failure occurs within 80% of the transmission line then the first distance relay should operate within zero seconds and disconnect a circuit breaker that separates the transmission line from the rest of the power distribution system. Zone 2 of this first relay is designed to protect 120% of the line at time delay of 0.5 seconds. 120% means 100% of the transmission line (TL1) and 20% of a directly following transmission line (TL2) which is normally protected by a second relay. The benefit of zone 2 is that if a fault occurs at TL2 and the second relay doesn't operate (in zero seconds) then the first relay protecting TL1 will be capable to stop the current flow into the fault by tripping circuit breaker and disconnecting TL1. If zone 2 of the first relay is given a time delay of zero seconds then the first relay will always disconnect the circuit breaker of TL1 whenever a fault occurs at TL2. This contradicts with the selectivity concept of protection system. The same thing is with zone 3 which has normally a time delay of 1 seconds but it protects 200% of the line, i.e. the total line TL1 and the total line TL2.

In a lot of power system incidences it occurs that the first relay disconnects its circuit breaker although there is no actual system fault. These incidences like power swings, voltage instability and load encroachment causes the first relay to measure the line length or directly line impedance, inside the zone 3 or zone2 reach of the relay causing it to interpret these measurements as a fault while in fact there is no fault. This was a reason for a lot of previous large scale blackouts worldwide.

As described above, a distance relay normally protects approximately 80 % of the length of a transmission line. However, in some cases it could be that, due to some factors, the distance relay protects less than the required percentage of the line. As an example zone 1 may be capable of protecting only 30% of the line. This fault is called under-reach. By other meaning the relay operates with higher time delays than zero for faults inside the 80%-section of the line. On the other hand, in some cases the distance relay could operate for faults outside its designated zone, e.g. that the relay operates at zero time delay for faults outside its reach of 80%. This fault is called over-reach.

Another relaying theory, so-called "differential relaying", is being used in power systems. Differential relays measure the currents at a sending and receiving end and issue its trip signal when the differential current, i.e. the difference between the two measured currents, exceeds a certain threshold. The theory of differential relaying is the best way to ensure the selectivity of the protection system. However, the occurrence of HIF could cause them not to discover the fault. Also these relays still measure a current value to make their decision and this way the relay is affected by the configuration and status of the power system, e.g. outage of line for maintenance, because normal over-current relays depends on estimation of the system condition or heuristic settings and this could be not accurate enough for proper setting or operation of the relay. Some of the prior art techniques proposed to detect virtual faults prefer not to use communication links to design a relaying algorithms and to base its decision on values of current and voltage at the relay location as described in publications "An adaptive scheme to prevent undesirable distance protection operation during voltage instability" by Jonsson, M. and Daalder, J.E. in IEEE Transactions on Power Delivery, vol.18, no.4, pp. 1174-1180, Oct. 2003; "A New Method to Prevent Undesirable Distance Relay Tripping During Voltage Collapse" by Ahmad Farid Abidin, Azah Mohamed and Afida Ayob in European Journal of Scientific Research, Vol. 31 No.1 (2009), pp.59-71 ; and "Blocking of Zone 3 Relays to Prevent Cascaded Events" by Seong-ll Lim, Chen-Ching Liu, Seung-Jae Lee, Myeon-Song Choi and Seong- Jeong Rim in IEEE Transactions on Power Systems, vol.23, no.2, pp.747-754, May 2008.

On the other hand some other techniques use new technologies based on remotely measured data like Synchronized Phasor Measurements (SPM) and fiber optic communications, as described in the publication "A study of synchronized sampling based fault location algorithm performance under power swing and out-of-step conditions" by Nan Zhang and Kezunovic, M. in Power Tech. 2005 IEEE Russia, vol., no., pp.1-7, 27-30 June 2005; and "Design and evaluation of a current differential protection scheme with enhanced sensitivity for high resistance in-zone faults on a heavily loaded line" by Villamagna, N. and Crossley, P.A. in Eighth IEE International Conference on Developments in Power System Protection, Vol.2, 5-8, pp. 410-413, April 2004.

Said publication "An adaptive scheme to prevent undesirable distance protection operation during voltage instability" suggest a new criterion to differentiate between fault and non fault case. The criterion is basically based on the monitoring of the rate of change of the voltage. Depending on the rate of change it is decided if a fault occurred in the system or not. However, the thresholds of dv/dt are defined based on the configuration of the system under study and should be studied for each

contingency in the system then set heuristically. This method also doesn't always result in right operation of protection system and mal-operation can still occur as described in publications "An adaptive scheme to prevent undesirable distance protection operation during voltage instability" and "A New Method to Prevent Undesirable Distance Relay Tripping During Voltage Collapse". The latter overcomes the occurrence of mal-operations by using a measure called Voltage Stability

Indicator (VSI) in addition to dv/dt. However, the threshold setting of VSI is

heuristically chosen. It is not stated if the threshold holds constant in all

configurations of the power system. The cited reference mainly overcome the mal- operation of zone 3 when voltage instability is occurring in the system. This method might not be suitable to prevent mal-operations of distance relays when other phenomena like power swing or load encroachment occur.

In said publication "Blocking of Zone 3 Relays to Prevent Cascaded Events" it is proposed to block the zone 3 operation based on comparing a predetermine value of power flow of the line and the actual measured value of the power flow. The decision to block zone 3 holds when the estimated and the measured value of power are identical. However, the estimation of the value of power flow of the line is not always guaranteed to be right due to the loop-flows described in publication "Power System Dynamics: Stability and Control" by Jan Machowski, Janusz W. Bialek and James R. Bumby, 2nd Edition, publisher John Wiley & Sons Ltd., 2008, pp.121 , which caused some previous blackouts. In this case the method described in "Blocking of Zone 3 Relays to Prevent Cascaded Events" could cause problems.

In publication "A study of synchronized sampling based fault location algorithm performance under power swing and out-of-step conditions", two measures are used, one to discover power swing and block the relay and another to calculate fault location and to allow the relay to trip. Thus, it is possible to differentiate between a fault and a power swing case. The use of two measures could be a reason for the relay operation to be unsatisfactory especially when a combination between an actual fault and a virtual fault occurs, e.g. a fault during a power swing.

The methods disclosed in the cited references are not verified to which limit they are capable to withstand the existence of HIF which might lead a relay not to differentiate between fault and non-fault case correctly or even not to estimate the fault location correctly. It is therefore an object of the present invention to provide a fault detection and protection method for transmission lines as well as a corresponding protection system that is capable to overcome all the previous problems, that is easy to realize and that offers complete selectivity and reliability in fault detection, in particular in case of real faults or virtual faults or combination of both, independent from the system configuration or system status.

This object is achieved by a method as described in claim 1 and a system as described in claimed 9. Advantages further developments of the method and the system are described in the respective dependant claims.

In accordance with the invention it is proposed a method for detection of a fault on at least one section of an electrical power transmission line, comprising the steps of

- measuring the voltage and the current at a first terminal of the transmission line, the first terminal being a power sending end of the section of the transmission line,

- measuring the voltage and the current at a second terminal of the transmission line, the second terminal being a power receiving end of the section of the transmission line,

- calculation of a fault resistance to ground between the first terminal and the second terminal on the basis of the measured voltages, currents and the line impedance between the first terminal and the second terminal, and

- identification of a fault on the basis of the calculated value of the fault

resistance.

The use of the value of the fault resistance R as a deterministic judging criteria enables to achieve discrimination of normal faults, from bolted faults to HIF with any value until several tens of Kiloohms, and "virtual faults" like power swings, voltage instability and load encroachment on the one hand, and to achieve a correction of the measured distance from a terminal to the fault in order to get an accurate fault location. Deterministic here means that it is not subject to a change of system configuration or heuristic assumptions like the case of over-current or differential relays which presents a very high sensitivity of the protection system. In addition to this it is not affected by virtual faults or tower foot or earth resistance like in case of conventional distance relays.

Changing the tripping decision of distance relay from being based on the differential current value to be based on the value of the fault resistance Rf improves the operation of the protection performance in many aspects and has many advantages in comparison to other used algorithms: Operation of distance relays under virtual faults is under control, i.e. distance relays can no more mal-operate. The proposed method is like the differential relaying in the sense that each equipment or

transmission line has its protection that is using the fault resistance information instead of the differential current information. This leads to a tripping decision that is not affected by the complete range of Rf from 0 through normal fault resistances until HIF and in this case the distance relay can no more under-reach. This saves the system selectivity and leads to discovery of any fault resistance and tripping in a suitable time. An easier protection system can be designed because there is no need for setting a biasing or restraining current. The sensitivity of the system is completely reserved.

The method according to the invention uses one judging criteria and one algorithm to prevent mal-operation of distance relays under all possible fault or virtual fault conditions. This saves the protection system from using several algorithms for each type of faults or virtual faults which adds to the complexity of the system and might be a source of errors. The approach used in this method is simple and this alleviates any complexities in the design of the relay software and leads to a better system/ relay security. The method could be also used in distribution networks where the values of fault resistances can be very high and hard to discover. This stretches the area of application of the method and makes it universal.

Another advantage is the capability to design the reach of zone 3 to any required value regardless of any N-x contingency in the system that might lead zone 3 to mal- operate. Thus, the new method according to the invention is suitable to be used as a backup function or as a monitor for the operation of existing backup functions like zone 2 or zone 3 of the distance relay. A protection system using the proposed method is not affected by the tower foot resistance or earth resistance because it will be added to the line resistance. In case of a fault (with or without a fault resistance), the method corrects the relay measured impedance to see accurately only the impedance to fault with no effect of fault resistance. The relay selectivity and tripping time delay will be always as required with no possibility to under-reach. An overreach fault as described in the preopening part of the description normally occurs due to the existence of fault resistance. As described in the following, the proposed method can't be affected by the fault resistance because it in principle uses the value of the fault resistance directly as an indication of the occurrence of fault or not.

All the mentioned points would in turn lead to a higher reliability of the protection system and prevent power system collapse or blackouts.

The invention can be used for detection of three phase faults as well as for single phase faults by calculating R_f for each transmission line. In normal system operation the three fault resistances of the three lines (R_fa, Rf_b and Rf_C) will be very high. Once a fault occurs on one of the lines its calculated value of Rf, e.g. Rf_a) will be within its practical range and the line will be tipped.

The invention is described hereinafter by means of the attached figures in which a schematic illustration of a transmission line and protection components is given. In the figures:

Fig. 1 Transmission line during fault (a) and no fault (b)

Fig. 2 Change of R_f depending on different fault conditions of line

Fig. 3 Schematic system for calculation of Rf

Fig. 4 Change of Rf in a simulation

Fig. 5 The new method in a block and monitor scheme

Fig. 6 Protection scheme for two transmission lines

Fig. 7 The new method as a standalone application

Fig. 8 Medium Length Transmission Line pi model

Fig. 9 At fault location 10%

Fig. 10: At fault location 30%

Fig. 11 : At fault location 50%

Fig. 12: Comparison of Rf and Abs Idiff Fig 13 Comparison to Conventional Differential Relays

Fig 14 Test system for load encroachment

Fig 15 Load encroachment

Fig 16 Change of Rf during load encroachment

Fig 17 HIF during a power swing (out of step condition)

Fig 18 Rf in case of no fault on the line

The main idea of the invention is to monitor the value of a fault resistance R_f. In case a certain transmission line of the system is having a fault as depicted in Fig. 1a, the value of the fault resistance R_f will be in between a known practical fault range, in particular from zero Ohms to few tens of KiloOhm. This range covers all expected fault resistances R_f over the whole voltage levels encountered in power systems from EHV to distribution. When the system is healthy the fault resistance R_f will not exist as shown in Fig. 1 b but we can imagine that it exists and that it has a very high value (see Fig. 1 b) compared to its value when there is a fault. The high value is theoretically infinity but practically a few egaOhm.

The problems that face the operation of distance relays can be divided into three categories. A first category comprises phenomena like power swings, voltage instability and load encroachment which are not considered as faults. The power system will be as in Fig. 1 b). However their existence could be interpreted by a distance relay as if there is a fault in the system, namely virtual fault. This could cause the relay to mal-operate. As mentioned in the last paragraph, the value of the fault resistance R_f in these cases will be very high and outside of their practical range which is a sign for the relay that there is no fault in the system and the relay operation could be blocked in this case. A second category relates to the existence of a fault resistance which causes the relay to over-reach or a HIF that is in many cases hard to be discovered by relays. These cases are presenting a real fault in the system as shown in Fig. 1a and in this case fault resistance R_f will be located inside its practical range in between 0 to few tens of KiloOhm. A third category comprises the

occurrence of the first two categories at the same time, e.g. a power swing and a HIF. This case should be interpreted by the relaying systems as fault. The value of the fault resistance R_f is still the judging criteria and so long its value in this case is in between its practical range, the relay would issue its tripping signal. Fig. 2 depicts the calculated value of fault resistance R_f in any of the three previous cases. As can be seen in Fig. 2 the value of fault resistance R_f is very high, i.e.

considerably higher than 100 kQ, in particular a few MegaOhm, in case that there is no fault on the transmission line or if the relay is experiencing a virtual fault, and on the other hand the value of R_f will be relatively low, i.e. less than 100 kQ, in particular a few tens of KiloOhm, in case of an actual fault. Furthermore, the value of fault resistance R_f will be approximately 0 Ohm in case of a bolted fault, i.e. if the transmission line has a ground fault.

On the basis of these findings and according to an aspect of the invention, a blocking signal is issued and transmitted to a relay if the calculated fault resistance R_f is outside a predetermined fault range, the relay being able to interrupt voltage supply over the transmission line in case of a fault but being blocked in operation if it receives the blocking signal. This ensures that a virtual fault will not trip the relay. Thus, mal-functioning in case of a virtual fault, i.e. power swing of HIF, can be prevented. In case of a zone 3 monitoring a blocking signal is issued and transmitted to a relay if a calculated fault resistance value of another section of the transmission line immediately following the first section is outside a predetermined fault range, the relay being able to interrupt voltage supply over the transmission line in case of a fault but being blocked in operation if it receives the blocking signal.

Preferably, the relay is unblocked if the calculated resistance value is in between the predetermined fault range. This means that the relay may be tripped in case of a real fault that is indicated by a comparatively low fault resistance value. Furthermore, in case of such a fault, a circuit breaker is tripped to disconnect the transmission line if the calculated fault resistance value R_f is in between the predetermined fault range.

It should be understood that a relay can't switch on or off a transmission line directly because a relay contact can't carry the high currents flowing in the transmission line. Thus, the relay only sends a signal to a circuit breaker in case of a detected fault. A circuit breaker has the capability to switch off a transmission line. In a preferred embodiment of the invention the fault range is 0 Ohm to a few tens of KiloOhm, thus comprising all values corresponding to an actual fault on the monitored transmission line.

The operation of the proposed method is based on the synchronous measurement of the voltages and currents at the different terminals. In a preferred embodiment of the invention a measuring technology called Synchronized Phasor Measurements (SPM) is used. This measuring technology allows to synchronize and measure any required variable in the system, in particular voltage and current, and to utilize the measured values in establishment of new control and protection function. More on this technology could be found in "Snchronized Phasor Measurements and Their

Applications", A.G. Phadke, J.S. Thorp, published by Springer, 2008.

In the following it will be explained how the fault resistance value Rf can be

calculated. Fig.3 shows a single transmission line 2 that is used only to explain how to calculate the value of fault resistance Rf. It should be noted that in actual applications and in systems with more line sections, Rf will be calculated for each transmission line, and for each section respectively, then protection schemes using blocking and monitoring of zone 2 and 3 could be achieved as will be explained below.

From Fig. 3, the voltages V_A and V_B and currents terminals A and B during a fault with fault resistance Rf can be described by the following equations (1 ) and (2):

By adding formulas (1) and (2) your get:

Where Z_A + Z_B is equivalent to the total impedance of the protected transmission line Z_TL . From formula (3) and considering I = I_A + I_B , the fault resistance Rf is equivalent to:

It is clear from formula (1 ) that the measured impedance

' Measured at A

of the relay in case of a fault resistance Rf will not be equivalent to Z_A but equals z Measured at A = z A ^•R 7.

This is the reason why relay could under-reach in case of faults with Rf.

Pursuant to these findings, in a preferred embodiment of the method according to the invention the fault resistance Rf is calculated by means of the formula

wherein

Rf is the fault resistance to be calculated,

V_A is the voltage at the first terminal (A)

V_B is the voltage at the second terminal (B) IA is the current flowing into the section (3, 4) of the transmission line (2) at the first terminal (A)

IB is the current flowing into the section (3, 4) of the transmission line (2) at the second terminal (B) and

ZTL is the impedance of the transmission line (2) between the first terminal

(A) and the second terminal (B)

ZTL is a constant that is found in the specifications of the transmission line. All power system utilities have information and specifications about any impedance of any transmission line in the system. Therefore, the impedance ZTL is either known or is given per length unit, i.e. per meter, so that it can be determined on the basis of the known length of the section of transmission line from the first terminal to the second terminal. However, if ZTL is not available for any reason then there are techniques to measure it directly using the voltages and currents at both ends of the transmission line. It is not required at all to calculate Z_A or Z_B.

The formula of fault resistance R_f given in (4) uses the value of the fault current If in a way that translates this value to an equivalent impedance R_f. Because R_f has known practical values, then it is advantageous to use it in the judgment of occurring of a fault or not (virtual fault). The results will be more accurate than using fault current l_f, i.e. in case of conventional differential protection which is not accurate enough in case of a High Impedance Fault and requires the setting of biasing current which is dependent on the system configuration and other estimations. The detection of virtual faults or the judgment of occurrence of an actual fault according to this invention is based on only one measure, namely the fault resistance R_f, so there will be no need to use two different measures to differentiate between a fault and a virtual fault.

Returning back to Fig. 2, the calculated value of R_f from (4) will have a very high value in case that there is no fault or relay seeing a virtual fault in the system. In this case blocking of operation of the relay, and consequently of zone 3 or zone 2, could be done. When a fault takes place in the system fault resistance Rf would change to fall in between the predetermined fault range and in this case the method according to the invention will be indicating that there is a fault in the system, the blocking signal is removed and zones 3 or 2 allowed to operate normally. If the fault resistance is high enough, e.g. in case of HIF, so that zone 3 or 2 underreaches, in this case the circuit breaker will be operated.

Fig. 4 depicts the change of R_f during no fault or virtual fault and during an actual fault. Fig. 4 only gives how the change of R_f calculated for a certain transmission line will look like during no fault and during fault.

According to the invention it is further proposed a system for detection of a fault on at least one section of an electrical power transmission line, comprising

- a first measurement unit for measuring the voltage and the current at a first terminal of the transmission line, the first terminal being a power sending end of the section of the transmission line,

- a second measurement unit for measuring the voltage and the current at a second terminal of the transmission line, the second terminal being a power receiving end of the section of the transmission line,

- a communication line for transmitting the measured data to a central control device and

- the central control device for calculation of a fault resistance to ground

between the first terminal and the second terminal on the basis of the measured voltages, currents, and the impedance between the first terminal and the second terminal, and for identification of a fault on the basis of the calculated value of the fault resistance.

The new method and aforementioned system according to the invention could be used in different arts. It could be used to aid for a better operation of existing distance relays with zone 3 or it could be used as a standalone application, in particular in distribution systems.

Example of using the method and system to aid with existing zone 3:

The flow chart in Fig. 5 shows how the protection of the system shown in Fig. 6 operates using the new method. In a first step 21 , the values of voltage V_A, V_B, V_c, and currents l_A, IB, I_C at the terminals A, B, C of all sections 3, 4 of a transmission line 2 are measured using any devices capable of making synchronized phasor measurements like Phasor

Measurement Units (PMU) as indicated in Fig. 6 by PMU1 , PMU2, PMU3. After the measurement process the PMUs send their measurement data to a central control device 8, in particular computer, using a communication line 7, in particular a flexible communication structure. At the central computer 8 the values of fault resistances R_f for all transmission line sections 3, 4 on the system 1 will be calculated. To illustrate the steps of the method the second transmission line section 4 in Fig.6 will be used. This second section 4 is part of zone 3 of relay R1 , i.e. from the point of view of relay R1 located at the beginning of the first section 3, i.e. at terminal A.

The fault resistance R_f2 on the second section 4 will be calculated using formula (4) and if it is found to be outside of its fault range, step 22, then blocking of zone 3 by blocking relay R1 takes place, step 23, wherein relay R1 protects both the first section 3 and the second section 4 of transmission line 2. Otherwise, zone 3 is unblocked, step 24. Because there is a possibility for zone 3 not to trip due to the existence of a fault resistance, the operation of zone 3, i.e. the status of the circuit breaker CB1 will still be monitored, and if the circuit breaker CB1 is not opened after the allotted time of 1 to1.5 seconds has passed, see step 25 in Fig. 5, e.g. due to HIF, then the circuit breaker CB1 will be opened, step 26. The same algorithm will be applied for the first section 3 of transmission line 2.

The protection system as shown in Fig. 6 comprises a relay R1 , R2, R3, R4 and a corresponding circuit breaker CB1 , CB2, CB3, CB4 at the beginning and at the end of each section 2, 3 of the transmission line 2. The arrows above the relays R1 , R2, R3, R4 indicate the direction of the monitored power flow, i.e. R1 protects the first and the second section 3, 4, relay R2 protects only the first section 2 in opposite direction, relay R3 protects only the second section 4 in the same direction as relay R1 and relay R4 protects the first and the second section 3, 4 in opposite direction.

According to Figure 6, zone 3 of relay R4 will be blocked so long the value of Rn, i.e. calculated fault resistance of the first section 3 transmission line 2, is very high indicating that there is no fault in the system. If Rn changes to be within its practical range which indicates a fault on the first section 3 then zone 3 of circuit breaker CB4 is unblocked. However, before zone 3 of circuit breaker CB 4 could operate, a chance should be given to relay R2 to operate and disconnect transmission line 2 by tripping circuit breaker CB2 because the fault is at the first section 3 of transmission line 2, for which relay R2 is responsible. If for any reason circuit breaker CB2 doesn't disconnect the circuit (breaker failure) or R2 has not operated, then zone 3 of R4 will trip the breaker CB4 to stop the flow of current into the fault on the first section 3.

For systems with more transmission lines, the values of Rf for all transmission lines will be calculated and monitored and as long as their measured values are

comparatively high, it is then indicating that all zone 3 in the substation don't need to operate and will be blocked. Once any of the measured fault resistances Rf of any transmission line section 3, 4 drops to be within its practical range, i.e. its fault range, then the designated zone 3 of that line will be unblocked and monitored.

Example of using the method and system as standalone application

Nowadays, the communication systems are very reliable in a comparison to many decades ago and it is possible to design the operation of zone 3 to be based on new technologies like SPM. In this context, the new method proposed here could be adapted to operate as a standalone zone 3 or zone 2 protection as shown in the flow chart of Fig. 7.

In a first step 27 the voltage values V_Al V_B, V_c, and currents l_A, IB, IC at the terminals A, B, C of all sections 3, 4 of a transmission line 2 are measured using Phasor Measurement Units PMU1 , PMU2, PMU3 and the values of fault resistances Rf for all transmission line sections 3, 4 on the system 1 are calculated. In a next step it is determined if the calculated value of fault resistance Rf is in between the fault range, lower than 100 kQ, step 28. If this is the case circuit breaker CB1 is tripped by relay R1 either immediately if the fault has been detected on the first section 3, or after one second if the fault has been detected on the second section 4 (zone3), step 29.

Otherwise, relay R1 is blocked so that no mal-functioning of the protection system 1 can occur, step 30. In the following results of an application of the method according to the invention in different cases of virtual faults and in HIF are presented.

Results in case of non-real faults

This section is intended to show the results of the application of the proposed method when the power system is exposed to any of the non-real faults like load encroachment, power swings, transient voltage instability or normal faults with low or high impedance .

Simulation

The formula of Rf as described in (4) was tested on the power system shown in Fig. 8 using a medium length transmission line represented by its pi model. Basically two tests have been conducted. The first test is performed under no- fault (e.g. load increase). The second test is a fault case taking into consideration the distribution of transmission line capacitance when the fault occurs at different locations on the line. The test has been conducted at different values of Rf and power angles. The model parameters are as given in the following table 1.

Parameter Value

Volt Level 345 kV

ZG1 (Ohm) 4+j25

ZG2 (Ohm) 4+j50

Z T.L.(Ohm/km) 0.037+j0.3

Y T.L.(ps/km) 3,76

Line Length 160 km

Table 1

1) No - Fault Case

Table.2 shows the calculated value of Rf in case the system is experiencing no- or non-real fault. The values of Rf in this case are calculated using (4) at different power angles corresponding to load increase. Delta 10 30 60 90

Rf 3^*106 2.2^*106 1.8*106 1.9^*106

TABLE 2: CALCULATED Rf WHEN IN VIRTUAL- OR NO- FAULT

The values of Rf in this case are very high because under no- or virtual- fault no actual Rf exists. It has been found that the value of Rf in this case is related to the reciprocal of Yc of the transmission line which is in the range of Mega Ohms. So long the system is in normal state or experiencing a non-real fault, the value of Rf is expected to keep so high indicating that there is no need for zone 3 distance relay to trip.

2) Fault Case

Once a fault occurs in the system, the value of Rf will have a change from very high values as given in Table.2 to very low values representing the value of the fault resistance (Rf), Fig. 4 showing a drop of Rf when a fault occurs.

In test 2, the value of Rf has been changed in the simulation from 0 to 10 kOhms then formula (4) has been used to retrieve back the value of Rf from the measured currents and voltages. Fig. 9 to Fig. 11 show the calculated values of Rf when a fault occurs at different locations and different power angles. At high power angles (e.g. delta=60) the deviation between the actual and the calculated value of Rf increases for faults near the line ends (e.g. 10%). At faults near the middle of the line the calculated values of Rf are enhanced. Regardless of how much exactly Rf is, the value of the calculated Rf has dropped from its very high value (e.g. during normal load increase or no fault condition as in table 2) into the range that is indicating that a fault has occurred between the monitored line terminals.

Comparing the use of the differential current and the proposed method in case of HIF, Fig. 12 shows that at HIF (e.g. beginning from 400 Ohms) the difference between IA and IB is getting less and in this case there is a possibility for the relay to trip and there is a possibility for it not to trip depending on the value of the relay setting and the bias current which is in itself depends on system configuration and estimation of the errors in measurements, Fig. 13. Rf on the other hand is having a range from 0 to 10 kOhms indicating that definitely there is a fault.

The mentioned errors can affect also the calculation of Rf in the proposed method, however the effect of these errors on calculation of Rf can be ignored because it will not drastically change the calculated Rf from few kOhms to some MegaOhms. The relation between the calculated value of Rf and its real value is linear while the differential current is suffering a drastic nonlinear change.

Load Encroachment as a case of virtual(non-real) fault

The system shown in Fig. 14 was used to test the relay performance under load encroachment. The two transmission lines have the same parameters as in Table 1. Fig. 15 traces the change of the measured impedance by the relay at Bus1 when the load is increasing. Under this load increase, normal zone 3 distance relays at Bus1 are expected to mal-trip. On the other hand, the value of the calculated Rf (between A and B) keeps very high (in MegaOhms) indicating the that distance relay at Bus1 (which is supposed to protect TL2 by its zone 3) is not required to trip due to the non existence of a fault at the TL2; Fig. 16.

System under power swing and transient voltage instability

The system in Fig. 8 has been used to check the operation of the algorithm under power swing. The test has been conducted under worst case of power swing (out of step).

Fig. 17 depicts the calculated value of Rf in case of HIF of 10000 Ohm and 1000 Ohm during a power swing. The shown noise in Fig. 17b could be overcome using a moving average window. Fig.18 on the other hand depicts the value of Rf in case of power swing with no fault. Less noise in both Fig.17 and Fig. 18 could also be obtained by calculating ZTL using equation (6) and using online measurements which will result in better results.

Claims

1. Method for detection of a fault or virtual fault on at least one section (3) of an electrical power transmission line (2), characterized in that it comprises the steps of

- measuring the voltage (VA) and the current (IA) at a first terminal (A) of the transmission line (2), the first terminal (A) being a power sending end of the section (3) of the transmission line (2),

- measuring the voltage (V_B) and the current (IB) at a second terminal (B) of the transmission line (2), the second terminal (B) being a power receiving end of the section (3) of the transmission line (2),

- calculation of a fault resistance (R_f) to ground between the first terminal (A) and the second terminal (B) on the basis of the measured voltages (V_A, V_B), currents (l_A, I_B) and the impedance (Z_LT) between the first terminal (A) and the second terminal (B), and

- identification of a fault on the basis of the calculated value of the fault resistance (R_f).

2. Method according to claim 1 , characterized in that the fault resistance (Rf) is calculated by means of the formula

wherein

R_f is the fault resistance to be calculated,

V_A is the voltage at the first terminal (A)

VB is the voltage at the second terminal (B) l_A is the current flowing into the section (3, 4) of the transmission line (2) at the first terminal (A)

l_B is the current flowing into the section (3, 4) of the transmission line (2) at the second terminal (B) and

Z_TL is the impedance of the transmission line (2) between the first terminal

(A) and the second terminal (B)

Method according to claim 1 or 2, characterized^" in that a blocking signal is issued and transmitted to a relay (R1 ) if the calculated fault resistance (R_f) is outside a predetermined fault range, the relay (R1) being able to interrupt voltage supply over the transmission line (2) in case of a fault but being blocked in operation if it receives the blocking signal, in particular when load encroachment, power swings or voltage instability occur.

Method according to claim 1 , 2 or 3, characterized in that a blocking signal is issued and transmitted to a relay (R1 ) if a calculated fault resistance (R_f) value of another section (4) of the transmission line (2) immediately following the first section (3) is outside a predetermined fault range, the relay (R1 ) being able to interrupt voltage supply over the transmission line (2) in case of a fault but being blocked in operation if it receives the blocking signal, in particular when load encroachment, power swings or voltage instability occur.

Method according to claim 3 or 4, characterized in that the relay (R1 ) is unblocked if the calculated resistance value (R_f) is in between the

predetermined fault range.

Method according to one of the preceding claims, characterized in that a circuit breaker (7) is tripped if the calculated resistance value (R_f) is in between a predetermined fault range.

Method according to one of the claims 3 to 6, characterized in that the predetermined fault range is 0 Ohm to a few tens of KiloOhm.

8. Method according to one of the preceding claims, characterized in that the voltages (V_A, V_B) and currents (l_A, IB) at the different terminals (A, B) are measured synchronously by means of Synchronized Phasor Measurements.

9. System for detection of a fault or virtual fault on at least one section (3) of an electrical power transmission line (2), characterized by

- a first measurement unit (PMU1) for measuring the voltage (V_A) and the current (l_A) at a first terminal (A) of the transmission line (2), the first terminal (A) being a power sending end of the section (3) of the transmission line (2),

- a second measurement unit (PMU2) for measuring the voltage (VB) and the current (IB) at a second terminal (B) of the transmission line (2), the second terminal (B) being a power receiving end of the section (3) of the transmission line (2),

- a communication line (7) for transmitting the measured data to a central control device (8) and

- the central control device (8) for calculation of a fault resistance (R_f) to ground between the first terminal (A) and the second terminal (B) on the basis of the measured voltages (V_A, V_B), currents (l_A, l_B), and the impedance (Z_LT) between the first terminal (A) and the second terminal (B), and for identification of a fault on the basis of the calculated value of the fault resistance (R_f).

10. System according to claim 9, characterized in that it further comprises a relay (R1) that is able to interrupt voltage supply over the transmission line (2) in case of a fault, the relay (R1) being connected to the central control device (8) via a communication line (7) and being able to receive a blocking signal via said communication line (7) blocking its operation.

11.System according to claim 10, characterized in that it further comprises a circuit breaker (CB1 ) to interrupt the power transmission over the transmission line (2), the circuit breaker (CB1) being tripable by the relay (R1).