0 Introduction In the various protection principles of EHV lines, phase-division and longitudinal differential has outstanding advantages such as simple principle, reliable operation and good selectivity. With the complexity of the power grid structure and the development of communication technologies, digital phase-differential protections such as Mitsubishi, GEC, and ABB have been successfully put into operation and have been widely used since the 1980s. China's digital phase separation differential protection has also been developed from early introduction to later-stage self-development. There are also more than 10 years of on-site operational experience.
Although there are many current differential protection criteria, there are still some special issues that need to be considered in the protection of ultra-high voltage long lines. These issues include compensation of capacitive current, speed of exit tripping, and reflection of high resistance grounding faults. Capacity, reliability of phase-selective tripping, ability to withstand saturation of current transformers (TA), etc. Relying on a single differential criterion can not solve these problems, but must carry out a comprehensive analysis of the existing differential criteria, learn from each other and design a comprehensive protection program with the best overall performance.
Based on the research, this paper proposes a new comprehensive scheme and carries out ATP simulation of the scheme. Simulation results show:
The program has anti-time-limited operating characteristics, strong resistance to capacitive currents, and can reflect high-impedance faults.
1 longitudinal differential protection integrated principle scheme For the sake of research convenience, this paper studies the 500kV double-ended system model. The protection scheme consists of a model that is based on a combination of time-sharing and multi-criteria parallel processing. The entire program is divided into the following four parts.
a. Ultra-high speed differential criterion
Where: Edz(k) is the floating threshold; Edz is the fixed threshold; k1 is the proportional coefficient; △im, △in are the fault components in the sampling current across the protection line; tk ≥ 5ms.
The ultra-high-speed differential criterion only invests within 0 to 20 ms after the start of protection, where equation (1) constitutes the basic criterion, equation (2) is the differential criterion, and equation (3) is the auxiliary criterion. If formula (1) is established and any one of formulae (2) and (3) is established, it is judged as an internal fault.
b. Sensitive differential criterion
The sensitive differential criterion (ie, the fault component differential criterion) is calculated 20 ms to 40 ms after the start of protection, and equations (4) and (5) simultaneously satisfy the protection trip. The phasor value in the above criteria is determined by a full-period Fourier method from the value of the fault component after one cycle of protection activation.
c. Reserve differential criteria
The backup differential criterion (ie, the conventional phasor difference criterion) is put into operation 40 ms after the start of protection, and Equation (6) and Equation (7) are the relationship of “ANDâ€, and the input is until the protection trips or returns.
d. High resistance differential criterion
The high-impedance differential criterion (ie, the zero-sequence differential criterion) is put into use 40ms after the start of protection. Equations (8) and (9) are satisfied at the same time and continue to operate for 100ms.
2 Analysis of the characteristics of the principle scheme The ideal protection scheme should have inverse time characteristics, that is, the shorter the failure, the longer the removal time. The role of the ultra-high-speed differential criterion is to rapidly remove severe internal faults that pose a serious threat to system stability. This criterion uses the new principle of the related current differential proposed in [1], using the integral of the instantaneous value of the fault component. This criterion has the following outstanding advantages:
a. Anti-time-limited action characteristics. Because it uses the instantaneous value of the fault component current, the internal fault is more serious. The larger the fault component current is, the faster the fault is judged. This compensates for the phasor differential and cannot quickly export serious faults due to the calculation of phasors. The lack of.
b. With natural resistance to capacitive current capability, capacitive current compensation is not required during internal faults. From the fault component equivalence circuit in Figure 1, we can see that for the ultra-high voltage long-line internal fault, although the influence of the capacitor current can cause a large phase difference between the fault component currents at both ends of the line, the capacitive reactance of the line equivalent capacitance is much larger than the line. Reactance and system reactance, so the basic phase relationship of the fault component currents at both ends will not change, ie the phase difference will not exceed 90°. Therefore, the accumulation on the left side of equation (2) does not change the sign, and the capacitor current only affects the operation speed without causing a protection refusal. The difference between internal and external faults is obvious. For an internal fault, the final result of the integral is positive. For an external fault, the final result of the integral is negative. The longer the time, the more obvious this characteristic is.
c. Since the fault component current is used, it is not affected by the load current.
Since equation (2) does not reflect airdrop failures, single power failures, etc., these special faults can be determined as an auxiliary criterion (3).
The sensitive differential criterion is the fault component differential criterion [2]. It uses a fault component current phasor from 0 to 40 ms to make a differential judgment, since it is calculated using a full-period Fourier method etc. 20 ms after the start of protection. , and use the voltage phasor to compensate for the capacitive current, so it has the following advantages:
a. High sensitivity and easy to set. Due to the use of the fault component current, the influence of the load current is removed from the principle [3], and the protection sensitivity is improved; the ratio of the differential momentum and the braking quantity is determined by the system impedance and the position of the fault point, and the zone fault and outside the zone. There is a clear interval between failures. In the case of internal faults, the ratio of differential to braking is greater than 1, the external fault is much less than 1, the action boundary is clear, and the setting of the proportional coefficient is relatively large.
b. For out-of-zone faults, more accurate capacitive current compensation can be performed, which can reduce the current threshold setting and further increase the sensitivity of the protection to internal high-impedance faults. In the case of internal faults, the fault component differential is less affected by the component of the capacitive current fault. The phasor quantity calculation, the transient effect of capacitive current discharge is also greatly weakened, and the capacitive current compensation also increases the amount of internal fault action.
c. The simulation results show that the fault component differential can reflect most of the internal faults except some high-impedance faults, and basically meet the requirements of EHV long-line differential protection. Based on the above advantages, it is reasonable to use Equations (4) and (5) as the sensitive differential criteria of this solution.
The backup differential criterion is the phasor differential criterion. Since the fault component cannot be used for a long period of time, this scheme uses the conventional phasor differential criterion as a backup criterion for the entire scheme. Its role is to remove the criterion from (1) to (5) after exiting the exclusion zone. The conversion failure and the new failure occurred when the protection did not return.
The high-resistance grounding differential criterion and the backup differential criterion are simultaneously input. The high-resistance grounding criterion adopted in this scheme is the zero-sequence differential criterion for delaying 100ms of operation. The criterion is mainly to reflect the high-resistance grounding fault under heavy load. Since the zero-sequence current is a fault component, it has a higher sensitivity. The research results in [4] show that: 1 For external faults, non-full-phase operation and system oscillation, the zero-sequence current flowing through the protected circuit is through, and the phase difference between the currents on both sides approaches 180°. 9) Will not malfunction; 2 For the internal fault, when the line length is 300km, the operating angle is 60 °, it can reflect the transition resistance up to 550n. The time delay 100ms used in this scheme can ensure that the zero-sequence current criterion is not affected by TA transients.
3 Capacitive Current Compensation Capacitance current is the main factor affecting the sensitivity of EHV long-line differential protection. Capacitive current compensation must be performed. At present, there are two kinds of capacitive current compensation schemes: half compensation and full compensation [5]. Under normal conditions and external fault conditions, the full compensation error is relatively large, and half compensation is a better method. However, for no-load closing, if TY is connected to the bus, half compensation can only compensate for all capacitor currents. 1/2. When the circuit switching condition can be obtained, it is better to adopt a switching compensation method, that is, the three-phase disconnection of the opposite end of the line adopts the full compensation mode, and switches to the half compensation mode after the opposite side is closed. This scheme uses a half-compensation method. details as follows:
For the fault component differential criterion, the voltage and current of the fault component at both ends of the line are also calculated according to equations (10) and (11). In the actual calculation, each sequence compensating current needs to be calculated according to the sequence network diagram, and then the compensating current of each phase is synthesized. Considering that the actual line is an EHV long line with uniform distribution parameters, the order capacitance is calculated as follows: 
When a fault occurs outside the protected line, the transient effects after the fault are not taken into consideration. After the compensation, the currents at both ends are added to 0, which means that the compensation can be fully compensated. When the line is switched to a no-fault line without load, if TY is on the line side, it can still be fully compensated by the sum of the compensation currents at both ends.
4 Simulation Analysis This paper uses the ATP electromagnetic transient simulation program to establish a 340km long 500kV line model according to the distribution parameters, as shown in Figure 2. According to the principle that the overvoltage coefficient of the no-load line power frequency exceeds 1.4, three kinds of system impedance are considered: large, medium, and small, and the power potential angle difference considers four kinds of uniform distribution from 0° to 60°. For the ground fault, 10 kinds of transition resistances from 0Ω to 300Ω are considered. The fault modes include single-phase grounding, two-phase short circuit, two-phase ground, and three-phase short circuit. There are 10 types of faults. An external fault point is set from the 5 bus to the 0 bus, and 3 internal fault points are set at the beginning of the MS line, the middle of the line, and the end of the line. For each operating condition, there are 400 fault conditions calculated for one short circuit point. A total of 5,101 fault conditions have been calculated. For the moment, the model does not consider the case of series capacitors and parallel reactances in the line, and does not consider the case of a failover.
The specific parameters are as follows: System impedance 1: Zm1=1.2675+j100.23350Ω, Zmo=0.5595+j33.6055Ω; System impedance 2: Zm1=0.44225+j33.4117Ω, Zmo=0.1865+j11.2017Ω; System impedance 3: Zm1= 2.374+j187.73Ω, Zmo=1.047+j62.939Ω.
The line distribution parameters are as follows: positive sequence: R1=0.027Ω/km, X1=0.2783Ω/km, C1=0.0127μF/km; zero sequence: Ro=0.195Ω/km, Xo=0.6946Ω/km, Co=8.98nF /km. The sample rate is 120 points per 20 ms.
Simulation results show:
a. Ultra-high-speed differential criteria can reflect most internal faults, and only two-phase earth faults and single-phase earth faults with a ground fault resistance of 300Ω cannot be reflected. And for the grounding resistance is less than 100Ω fault can guarantee 5ms exit, for the grounding resistance of 100Ω ~ 300Ω fault can guarantee about 10ms exit. On the basis of appropriate tuning (see [1] for specific tuning methods), high speed, inverse timing, and reliable operating characteristics can be guaranteed. Fig. 3 and Fig. 4 show the results of the product of the current sampling values ​​of the out-of-zone fault and the end-of-zone fault, respectively, when the load is zero before the fault and the grounding resistance is 300Ω. It can be clearly seen from the figure that for an out-of-zone fault, the product of the current is negative except for the beginning of a very short time, and the other is negative, so the 10ms integral is a large negative value; for the internal fault, the integral result is 10ms. For larger positive values, the difference between the two is very obvious.
b. Sensitive differential criteria can reflect all faults in the area, with high reliability and large set margins. Taking system impedance 2 as an example, when the sensitive differential criterion is not used for capacitive current compensation, the minimum brake current for faults or misoperations outside the zone is 393.2 A, and the maximum setting current for faults within the zone is 612. 28A, can meet the line protection requirements; the use of current compensation, the area outside the fault does not move the minimum braking current becomes 136A, the maximum set current in the area does not resist failure is 501.2A, internal and external faults are significantly different.
c. The backup differential criterion cannot meet the line protection requirements if the capacitor current compensation is not used; after the compensation is adopted, the performance is obviously improved and the line protection requirements can be met. However, regardless of the fault, the fault component differential criterion has higher sensitivity and selectivity than the phasor differential criterion.
d. Simulation shows that the low-pass filter has a great influence on the correct operation of the protection, and after adopting filtering measures, the influence of the transient current can be significantly reduced, and the reliability of the protection action can be improved.
e. The single-sided power line is vacant to the fault. The above criteria can meet the protection action requirements. No-load closing, the charging current contains high-frequency oscillation components, and the transient current peak is high. The influence of non-mains frequency components must be reduced by the digital low-pass filter and differential filter, and the operating current is reduced by the capacitive current compensation. On this basis, the influence of the current threshold can be neglected by proper selection.
f. When zero-sequence differential protection is used, it needs to be used in conjunction with an independent phase selection element in order to meet the needs of split-phase tripping and reclosing. Phase selection elements also need further study.
5 Conclusions Based on the study of the existing differential protection criteria, combined with the characteristics of the phase separation of the EHV transmission line, this paper proposes a differential principle based on the overall performance of the optimal target. simulation Research. The simulation results show that the scheme has good anti-time-limit action characteristics, strong resistance to capacitive currents, and can reflect high-impedance faults. It is a kind of EHV line differential protection scheme with better overall performance.
Although there are many current differential protection criteria, there are still some special issues that need to be considered in the protection of ultra-high voltage long lines. These issues include compensation of capacitive current, speed of exit tripping, and reflection of high resistance grounding faults. Capacity, reliability of phase-selective tripping, ability to withstand saturation of current transformers (TA), etc. Relying on a single differential criterion can not solve these problems, but must carry out a comprehensive analysis of the existing differential criteria, learn from each other and design a comprehensive protection program with the best overall performance.
Based on the research, this paper proposes a new comprehensive scheme and carries out ATP simulation of the scheme. Simulation results show:
The program has anti-time-limited operating characteristics, strong resistance to capacitive currents, and can reflect high-impedance faults.
1 longitudinal differential protection integrated principle scheme For the sake of research convenience, this paper studies the 500kV double-ended system model. The protection scheme consists of a model that is based on a combination of time-sharing and multi-criteria parallel processing. The entire program is divided into the following four parts.
a. Ultra-high speed differential criterion
Where: Edz(k) is the floating threshold; Edz is the fixed threshold; k1 is the proportional coefficient; △im, △in are the fault components in the sampling current across the protection line; tk ≥ 5ms.
The ultra-high-speed differential criterion only invests within 0 to 20 ms after the start of protection, where equation (1) constitutes the basic criterion, equation (2) is the differential criterion, and equation (3) is the auxiliary criterion. If formula (1) is established and any one of formulae (2) and (3) is established, it is judged as an internal fault.
b. Sensitive differential criterion
c. Reserve differential criteria
The backup differential criterion (ie, the conventional phasor difference criterion) is put into operation 40 ms after the start of protection, and Equation (6) and Equation (7) are the relationship of “ANDâ€, and the input is until the protection trips or returns.
d. High resistance differential criterion
The high-impedance differential criterion (ie, the zero-sequence differential criterion) is put into use 40ms after the start of protection. Equations (8) and (9) are satisfied at the same time and continue to operate for 100ms.
2 Analysis of the characteristics of the principle scheme The ideal protection scheme should have inverse time characteristics, that is, the shorter the failure, the longer the removal time. The role of the ultra-high-speed differential criterion is to rapidly remove severe internal faults that pose a serious threat to system stability. This criterion uses the new principle of the related current differential proposed in [1], using the integral of the instantaneous value of the fault component. This criterion has the following outstanding advantages:
a. Anti-time-limited action characteristics. Because it uses the instantaneous value of the fault component current, the internal fault is more serious. The larger the fault component current is, the faster the fault is judged. This compensates for the phasor differential and cannot quickly export serious faults due to the calculation of phasors. The lack of.
b. With natural resistance to capacitive current capability, capacitive current compensation is not required during internal faults. From the fault component equivalence circuit in Figure 1, we can see that for the ultra-high voltage long-line internal fault, although the influence of the capacitor current can cause a large phase difference between the fault component currents at both ends of the line, the capacitive reactance of the line equivalent capacitance is much larger than the line. Reactance and system reactance, so the basic phase relationship of the fault component currents at both ends will not change, ie the phase difference will not exceed 90°. Therefore, the accumulation on the left side of equation (2) does not change the sign, and the capacitor current only affects the operation speed without causing a protection refusal. The difference between internal and external faults is obvious. For an internal fault, the final result of the integral is positive. For an external fault, the final result of the integral is negative. The longer the time, the more obvious this characteristic is.
c. Since the fault component current is used, it is not affected by the load current.
Since equation (2) does not reflect airdrop failures, single power failures, etc., these special faults can be determined as an auxiliary criterion (3).
The sensitive differential criterion is the fault component differential criterion [2]. It uses a fault component current phasor from 0 to 40 ms to make a differential judgment, since it is calculated using a full-period Fourier method etc. 20 ms after the start of protection. , and use the voltage phasor to compensate for the capacitive current, so it has the following advantages:
a. High sensitivity and easy to set. Due to the use of the fault component current, the influence of the load current is removed from the principle [3], and the protection sensitivity is improved; the ratio of the differential momentum and the braking quantity is determined by the system impedance and the position of the fault point, and the zone fault and outside the zone. There is a clear interval between failures. In the case of internal faults, the ratio of differential to braking is greater than 1, the external fault is much less than 1, the action boundary is clear, and the setting of the proportional coefficient is relatively large.
b. For out-of-zone faults, more accurate capacitive current compensation can be performed, which can reduce the current threshold setting and further increase the sensitivity of the protection to internal high-impedance faults. In the case of internal faults, the fault component differential is less affected by the component of the capacitive current fault. The phasor quantity calculation, the transient effect of capacitive current discharge is also greatly weakened, and the capacitive current compensation also increases the amount of internal fault action.
c. The simulation results show that the fault component differential can reflect most of the internal faults except some high-impedance faults, and basically meet the requirements of EHV long-line differential protection. Based on the above advantages, it is reasonable to use Equations (4) and (5) as the sensitive differential criteria of this solution.
The backup differential criterion is the phasor differential criterion. Since the fault component cannot be used for a long period of time, this scheme uses the conventional phasor differential criterion as a backup criterion for the entire scheme. Its role is to remove the criterion from (1) to (5) after exiting the exclusion zone. The conversion failure and the new failure occurred when the protection did not return.
The high-resistance grounding differential criterion and the backup differential criterion are simultaneously input. The high-resistance grounding criterion adopted in this scheme is the zero-sequence differential criterion for delaying 100ms of operation. The criterion is mainly to reflect the high-resistance grounding fault under heavy load. Since the zero-sequence current is a fault component, it has a higher sensitivity. The research results in [4] show that: 1 For external faults, non-full-phase operation and system oscillation, the zero-sequence current flowing through the protected circuit is through, and the phase difference between the currents on both sides approaches 180°. 9) Will not malfunction; 2 For the internal fault, when the line length is 300km, the operating angle is 60 °, it can reflect the transition resistance up to 550n. The time delay 100ms used in this scheme can ensure that the zero-sequence current criterion is not affected by TA transients.
3 Capacitive Current Compensation Capacitance current is the main factor affecting the sensitivity of EHV long-line differential protection. Capacitive current compensation must be performed. At present, there are two kinds of capacitive current compensation schemes: half compensation and full compensation [5]. Under normal conditions and external fault conditions, the full compensation error is relatively large, and half compensation is a better method. However, for no-load closing, if TY is connected to the bus, half compensation can only compensate for all capacitor currents. 1/2. When the circuit switching condition can be obtained, it is better to adopt a switching compensation method, that is, the three-phase disconnection of the opposite end of the line adopts the full compensation mode, and switches to the half compensation mode after the opposite side is closed. This scheme uses a half-compensation method. details as follows:
When a fault occurs outside the protected line, the transient effects after the fault are not taken into consideration. After the compensation, the currents at both ends are added to 0, which means that the compensation can be fully compensated. When the line is switched to a no-fault line without load, if TY is on the line side, it can still be fully compensated by the sum of the compensation currents at both ends.
4 Simulation Analysis This paper uses the ATP electromagnetic transient simulation program to establish a 340km long 500kV line model according to the distribution parameters, as shown in Figure 2. According to the principle that the overvoltage coefficient of the no-load line power frequency exceeds 1.4, three kinds of system impedance are considered: large, medium, and small, and the power potential angle difference considers four kinds of uniform distribution from 0° to 60°. For the ground fault, 10 kinds of transition resistances from 0Ω to 300Ω are considered. The fault modes include single-phase grounding, two-phase short circuit, two-phase ground, and three-phase short circuit. There are 10 types of faults. An external fault point is set from the 5 bus to the 0 bus, and 3 internal fault points are set at the beginning of the MS line, the middle of the line, and the end of the line. For each operating condition, there are 400 fault conditions calculated for one short circuit point. A total of 5,101 fault conditions have been calculated. For the moment, the model does not consider the case of series capacitors and parallel reactances in the line, and does not consider the case of a failover.
The line distribution parameters are as follows: positive sequence: R1=0.027Ω/km, X1=0.2783Ω/km, C1=0.0127μF/km; zero sequence: Ro=0.195Ω/km, Xo=0.6946Ω/km, Co=8.98nF /km. The sample rate is 120 points per 20 ms.
Simulation results show:
a. Ultra-high-speed differential criteria can reflect most internal faults, and only two-phase earth faults and single-phase earth faults with a ground fault resistance of 300Ω cannot be reflected. And for the grounding resistance is less than 100Ω fault can guarantee 5ms exit, for the grounding resistance of 100Ω ~ 300Ω fault can guarantee about 10ms exit. On the basis of appropriate tuning (see [1] for specific tuning methods), high speed, inverse timing, and reliable operating characteristics can be guaranteed. Fig. 3 and Fig. 4 show the results of the product of the current sampling values ​​of the out-of-zone fault and the end-of-zone fault, respectively, when the load is zero before the fault and the grounding resistance is 300Ω. It can be clearly seen from the figure that for an out-of-zone fault, the product of the current is negative except for the beginning of a very short time, and the other is negative, so the 10ms integral is a large negative value; for the internal fault, the integral result is 10ms. For larger positive values, the difference between the two is very obvious.
b. Sensitive differential criteria can reflect all faults in the area, with high reliability and large set margins. Taking system impedance 2 as an example, when the sensitive differential criterion is not used for capacitive current compensation, the minimum brake current for faults or misoperations outside the zone is 393.2 A, and the maximum setting current for faults within the zone is 612. 28A, can meet the line protection requirements; the use of current compensation, the area outside the fault does not move the minimum braking current becomes 136A, the maximum set current in the area does not resist failure is 501.2A, internal and external faults are significantly different.
c. The backup differential criterion cannot meet the line protection requirements if the capacitor current compensation is not used; after the compensation is adopted, the performance is obviously improved and the line protection requirements can be met. However, regardless of the fault, the fault component differential criterion has higher sensitivity and selectivity than the phasor differential criterion.
d. Simulation shows that the low-pass filter has a great influence on the correct operation of the protection, and after adopting filtering measures, the influence of the transient current can be significantly reduced, and the reliability of the protection action can be improved.
e. The single-sided power line is vacant to the fault. The above criteria can meet the protection action requirements. No-load closing, the charging current contains high-frequency oscillation components, and the transient current peak is high. The influence of non-mains frequency components must be reduced by the digital low-pass filter and differential filter, and the operating current is reduced by the capacitive current compensation. On this basis, the influence of the current threshold can be neglected by proper selection.
f. When zero-sequence differential protection is used, it needs to be used in conjunction with an independent phase selection element in order to meet the needs of split-phase tripping and reclosing. Phase selection elements also need further study.
5 Conclusions Based on the study of the existing differential protection criteria, combined with the characteristics of the phase separation of the EHV transmission line, this paper proposes a differential principle based on the overall performance of the optimal target. simulation Research. The simulation results show that the scheme has good anti-time-limit action characteristics, strong resistance to capacitive currents, and can reflect high-impedance faults. It is a kind of EHV line differential protection scheme with better overall performance.
ADF
ADFADF
WEIHAI XINYUE GARMENT CO.LTD. , https://www.whxinyue.cn