CN115313324A - Single-ended quantity protection method suitable for multi-ended flexible direct current system - Google Patents

Abstract

The invention relates to a single-end quantity protection method suitable for a multi-terminal flexible direct current system, analyzes the fault characteristics of the direct current line current and bridge arm current after the direct current line of the flexible direct current grid fails, and proposes a fault current direction criterion to select the pre-trip circuit breaker; the protection action time margin is set according to the action delay of the circuit breaker, so as to determine the reasonable protection action time according to the blocking time of the converter station; Accurate identification and rapid isolation, independent of line boundary elements, and strong resistance to transition resistance.

Description

A Single-Ended Quantitative Protection Method Applicable to Multi-terminal Flexible DC System

technical field

The invention belongs to the technical field of electric power system and automation, and in particular relates to a single-end quantity protection method suitable for a multi-terminal flexible DC system.

Background technique

When a fault occurs in the DC grid, the capacitors of each converter sub-module are violently discharged, and the fault current rises rapidly, which will seriously affect the safety of related electrical equipment in the system. In order to quickly remove the fault line to ensure the safe operation of the system, it is of great significance to study a fast and reliable protection scheme for the flexible HVDC transmission system.

Most of the main protection of the flexible DC grid chooses the protection principle based on single-ended electrical quantity to meet the requirements of the main protection of the flexible DC grid for the rapidity of protection. According to the division of principles, the existing single-ended quantity protection methods include voltage and current method, traveling wave method, boundary method and so on.

However, none of the above methods considers the cooperation between the action of the circuit breaker and the lockout of the converter station, and the lockout of the converter station may still occur during the fault isolation period. When the converter station is put into operation again, the capacitor needs to be recharged, which is not conducive to the rapid recovery of the grid power supply and greatly prolongs the power outage time.

Therefore, the present invention analyzes the fault characteristics of the DC line current and the bridge arm current after the DC line of the flexible DC grid fails, and on the basis of considering the action delay of the circuit breaker and the action protection time margin, proposes a method based on the converter station blocking Time-sensitive flexible DC power grid single-ended protection scheme to realize rapid removal of faulty lines before the converter station is blocked.

Contents of the invention

The purpose of the present invention is to overcome the deficiencies of the prior art, provide a single-ended protection method suitable for multi-terminal flexible DC systems, analyze the fault characteristics of the DC line current and the bridge arm current after the DC line of the flexible DC grid fails, and propose The fault current direction criterion is used to select the pre-trip circuit breaker; the protection action time margin is set according to the action delay of the circuit breaker, and the reasonable protection action time is determined by the block time of the converter station; this method can ensure that the converter station does not Under the premise of blocking, the accurate identification and rapid isolation of faulty lines can be realized, without relying on line boundary elements, and it has a strong ability to withstand transition resistance.

The present invention solves its technical problem and realizes through the following technical solutions:

A single-ended quantity protection method suitable for multi-terminal flexible DC systems, characterized in that: the steps of the method are:

S1. Analyze the directional characteristics of the DC line fault current and the fault characteristics of the bridge arm current after a fault occurs on the DC side of the flexible DC grid fault;

(1) DC line fault current direction characteristics

Based on the analysis of the four-terminal flexible DC system model, for the flexible DC grid with true bipolar connections, no matter if a bipolar short circuit, a pole-to-metal circuit short circuit or a single-pole grounding short circuit fault occurs on the DC side, the converter station will block Before, the fault pole MMC can form a fault circuit through the fault pole, metal return line or the earth on the DC side;

After a fault occurs in a multi-terminal flexible DC system using true bipolar wiring, the direction of the fault current variation on the DC side line of the converter station at any end is obtained. When a positive ground fault or a bipolar short-circuit fault occurs on the line, the positive line The direction of the change of the fault current is from the bus to the line; when a negative ground fault or a double-pole short-circuit fault occurs on the line, the direction of the change of the fault current on the negative line is from the line to the bus;

Therefore, it is assumed that the positive direction of the fault current change of the positive line is from the bus to the line, and the positive direction of the fault current change of the negative line is from the line to the bus. Therefore, no matter whether the positive line or the negative line is faulty, the two fault lines The fault current variation on both sides is positive;

In actual engineering, when the length difference of each section of the DC power grid is not very large, the impedance difference of each section of the line will not be very large. For any busbar, if a fault occurs on one of the outgoing lines of this busbar, the busbar will not appear. The change direction of the fault current on the two outgoing lines of the bus is the same; for the converter station connected to the bus, the direction of the change of the fault current on the fault line measured must be positive, and the fault current on the other line must be positive. The direction of current variation must be negative; therefore, when the direction of fault current variation on two outgoing lines of a bus is the same, it can be considered that the line connected to the bus must not be a fault line;

(2) Bridge arm current fault characteristics

In a typical half-bridge MMC converter topology, the half-bridge MMC converter consists of three symmetrical phase units, each phase includes upper and lower bridge arms, and each bridge arm consists of several identical sub-modules and bridge arms Reactors are connected in series;

The MMC bridge arm current is composed of three parts during normal operation, namely the AC side current, the DC side current and the bridge arm circulation current. The upper and lower bridge arm currents i _jp and i _jn (j=a,b,c) can be expressed as :

In the formula: I _ac is the amplitude of the AC side current, and I _cir is the amplitude of the bridge arm circulating current;

After the fault, the sub-module capacitors in each converter station quickly discharge to the fault point, causing the fault current of the DC line to rise sharply, and the DC component of the bridge arm current increases rapidly. At the same time, the discharge of the sub-module capacitors causes the equivalent voltage of the bridge arm to change, and then Cause the injection current of the AC side to change;

After the DC side fault occurs, the change of the current of each bridge arm of the converter station is mainly caused by the change of the DC component and the change of the injection current of the AC side; The proportion in is relatively small, and its amplitude changes little in a short period of time before and after the fault, so it can be approximately considered that the amplitude of the bridge arm circulation does not change within a few milliseconds after the fault occurs;

After a fault occurs on the DC side, the rectifier station and the converter station show different fault characteristics: the magnitudes of the DC current and the AC current after the rectifier station fault continue to increase, and the DC current and the AC current after the inverter station fault The amplitudes of all have to go through the process of first decreasing and then reversely increasing;

S2. For the DC line faults that occur in the multi-terminal flexible DC grid, a single-terminal protection scheme based on the blocking time of the converter station is proposed;

First, select the pre-trip circuit breaker according to the direction of the fault current change; second, determine the action time of the protection according to the blocking time of the converter station. On the basis of correctly selecting the pre-trip circuit breaker, the protection of each section of the line can be activated by Cooperate in time to complete the removal of the fault line;

(1) Fault current direction criterion and selection of pre-trip circuit breaker

The current change is used to reflect the change of the fault current. The selection of the positive direction is related to the polarity of the current flowing through the line, that is, the positive direction of the current change of the positive line is from the bus to the line, and the positive direction of the current change of the negative line is from the line. flow to the busbar, define the rate of change of current as follows:

In the formula: i _k and i _k-1 represent the adjacent current sampling values within the sampling interval ΔT;

Define the current variation direction index S, which represents the direction of the fault current variation measured at the end of the DC line;

S=1 means that the direction of the current change is positive, and the result of formula (2) is greater than 0 at this time;

S=-1 means that the direction of current variation is negative, and the result of formula (2) is less than 0;

S=0 represents that no current variation is detected, and the result of formula (2) is equal to 0 at this moment;

Considering the normal fluctuation of the current on the DC line during the normal operation of the system, when the absolute value of the detected current change rate exceeds the set value k _set , the direction index changes, and the following criterion for the direction of the fault current change is proposed:

Combined with the analysis of the fault current characteristics of the DC line, each converter station can select the pre-trip circuit breaker according to the direction index S of the fault current variation measured at the end of the line, and the pre-trip circuit breaker selection criteria S _im and S _in are respectively is the direction index of the fault current variation detected at the circuit breakers CB _im and CB _in , and the selection criterion of the pre-trip circuit breaker shown in formula (4) is proposed for each converter station:

(2) Calculation of protection action time based on converter station blocking time

On the basis of determining the pre-trip circuit breaker, the protection action time of each converter station depends on the blocking time of the converter station. Under the premise of ensuring that the converter station is not blocked, this scheme considers the circuit breaker action delay and action In the case of time margin, the protection action time is determined, and the specific calculation method is shown in formula (5):

t _act ＝t _block -t _cb -t _yd (5)

In the formula: t _block is the block time of the converter station;

_tcb is the breaker delay time;

t _yd is the protection action time margin;

Considering the action delay of the circuit breaker and the cooperation of the circuit breakers on both sides of the line, setting t _yd equal to t _cb can ensure the protection cooperation of each section to the greatest extent;

Considering that the bridge arm current of the inverter station has different fault characteristics from that of the rectifier station, a unified processing method is proposed for the inverter station and the virtual blocking time of the inverter station is defined, which can well reflect the severity of the fault; The amplitude of the circulating current of the bridge arm does not change, and the AC and DC components of the bridge arm current of the inverter station are analyzed and processed separately;

In order to make the inverter station show the effect that the received active power continues to increase after the fault, the change of the AC current after the fault is reversed, that is, the value after the fault is subtracted by twice the normal operating value, and the direction is changed. The processing expression is shown in formula (6):

i _ac '＝2i _ac0 -i _ac (6)

In the formula: i _ac ' is the virtual alternating current;

i _ac0 is the AC current during normal operation;

i _ac is the actual AC current after the fault;

The magnitude of the fault current on the virtual AC side of the inverter station obtained after the direction change process continues to increase after the fault, which is consistent with the change trend of the AC side current of the rectifier station under the same initial current;

Considering the transmission of power on the DC side, in order to make the DC current of the inverter station show the same continuous increase effect as that of the rectifier station, the processing method of i _DC is the same as that of i _AC , and the expression of direction change processing is as follows: 7) as shown:

i _dc '＝2i _dc0 -i _dc (7)

In the formula: i _dc 'is the modified DC current;

i _dc0 is the DC current before the fault;

S3. Based on the waveform fitting prediction of the moving data window, the converter station can predict the blocking time of the converter station in real time through the fault data;

Based on the analysis of the fault characteristics of the bridge arm current, it can be seen that after the DC side fault occurs, the change of each bridge arm current is mainly caused by the DC component and the AC component, and the change of the bridge arm circulating current is small, that is, I _cir can be considered to be related to the fault The same as before; therefore, the DC current i _dc and the AC current I _ac sin(ωt+θ ₁ ) are respectively fitted and predicted, and the bridge arm current waveform after the fault can be obtained;

After the DC side fault occurs, the DC current at the outlet of the converter station can be directly measured, so for the DC component of the bridge arm current, the fitting prediction can be made directly based on the measured value of the DC current at the outlet of the converter station;

After a fault occurs on the DC side, the change of the injected current on the _AC side is not only the change of the amplitude Iac, but also its frequency ω and phase _θ1 will also change to a certain extent; in the control link of the converter, the AC current is controlled by Converted to the d, q axis components to participate in the control, the change of the d, q axis components is similar to the change of the DC current; therefore, in the fitting prediction of the AC current, the measured AC current can be transformed into The form of the d and q axis components of the AC current, and then respectively carry out fitting predictions on i _d and i _q , and finally obtain the fitting prediction results of the AC side current through inverse Parker transformation;

In the initial period of time after the DC side fault occurs, the DC current i _dc and the d, q axis components of the AC current i _d , i _{q do} not increase strictly linearly, and their slopes are constantly changing;

The second-order polynomial fitting method is used to fit and predict the DC current component and the AC current component in the fault current of the bridge arm respectively. Since the change of the fault current growth rate is consistent, a moving data window is used for fault current fitting prediction;

Firstly, curve fitting is performed according to the sampling data of the latest T _nh window length to obtain the expression of the fault current, and then the subsequent fault current value with a length of t _yc is obtained according to this expression; combined with the protection action time calculation formula proposed in formula (5), Set the fault current waveform prediction duration t _yc as t _yc =t _cb +t _yd , that is, each fitting prediction obtains the bridge arm current value with a time length of t _yc after the current moment; if the predicted current value reaches the converter station Blocking condition, the trip signal should be issued immediately; if the predicted waveform does not meet the blocking condition of the converter station, the trip signal will not be issued, and the next prediction will be made after the next sampling data is obtained.

Advantage of the present invention and beneficial effect are:

The present invention is applicable to the single-ended quantity protection method of the multi-terminal flexible DC system. Compared with the prior art, the protection action time is calculated based on the blocking time of the converter station, and each protection can realize selective protection through the cooperation of the action time. Cut off the faulty line; in addition, the present invention also fully considers the cooperation between the protection action and the blockout of the converter station to ensure that there will be no blockage of the converter station during the fault isolation period, which is conducive to the rapid recovery of the power grid after the fault is cleared.

Description of drawings

Figure 1 is a schematic diagram of the four-terminal DC grid structure;

Figure 2 is a schematic diagram of the direction of DC line fault current variation under different fault types;

Figure 3 is a schematic diagram of the topology of the half-bridge MMC converter;

Fig. 4 is a schematic diagram of the selection criteria of the pre-trip circuit breaker;

Fig. 5 is a schematic diagram of fault current prediction principle based on moving data window;

Figure 6 is a comparison chart between the predicted value and the actual value of the bridge arm fault current;

Fig. 7 is a flowchart of the protection logic based on the blocking time of the converter station.

Detailed ways

The present invention will be further described in detail below through the specific examples, the following examples are only descriptive, not restrictive, and cannot limit the protection scope of the present invention with this.

Aiming at the multi-terminal flexible DC power grid constructed by the modularized multi-level converter, the present invention proposes a single-ended quantity protection method for the multi-terminal flexible DC system, and analyzes the DC line current and bridge current after the DC line of the flexible DC power grid fails. According to the fault characteristics of the arm current, the fault current direction criterion is proposed to select the pre-trip circuit breaker; the protection action time margin is set according to the action delay of the circuit breaker, and the reasonable protection action time is determined by the blocking time of the converter station. This method can realize accurate identification and rapid isolation of faulty lines on the premise of ensuring that the converter station is not blocked, does not rely on line boundary elements, and has a strong ability to withstand transition resistance. Include the following steps:

S1. Analyze the directional characteristics of the fault current of the DC line and the fault characteristics of the bridge arm current after the fault occurs on the DC side of the flexible DC grid fault

(1) DC line fault current direction characteristics

Based on the analysis of the four-terminal flexible DC system model shown in Figure 1, for a flexible DC grid with true bipolar connections, no matter if a bipolar short circuit, a pole-to-metal loop short circuit or a single-pole grounding short circuit fault occurs on the DC side, Before the converter station is blocked, the fault pole MMC can form a fault circuit through the fault pole, metal return line or the ground on the DC side.

After a fault occurs in a multi-terminal flexible DC system using a true bipolar connection, the direction of the fault current variation on the DC side of the converter station at any end is shown in Figure 2. When a positive ground fault or a double-pole short-circuit fault occurs on the line, the direction of the fault current change on the positive line is from the busbar to the line; when a negative ground fault or a double-pole short-circuit fault occurs on the line, the fault current on the negative line The direction of current variation is from the line to the bus. Therefore, it is assumed that the positive direction of the fault current change of the positive line is from the bus to the line, and the positive direction of the fault current change of the negative line is from the line to the bus. Therefore, no matter whether the positive line or the negative line is faulty, the two fault lines The fault current variation on both sides is positive.

In actual engineering, when the length difference of each section of the DC power grid is not very large, the difference in impedance of each section of the line will not be very large. For any busbar, if a fault occurs on one of the outgoing lines of the busbar, the direction of the change of fault current on the two outgoing lines of the busbar will not appear in the same direction. For the converter station connected to the busbar, the measured direction of the fault current change on the fault line must be positive, and the direction of the fault current change on the other line must be negative. Therefore, when the direction of change of the fault current on the two outgoing lines of a certain bus is the same, it can be considered that the line connected to the bus must not be the fault line.

(2) Bridge arm current fault characteristics

A typical half-bridge MMC converter topology is shown in Figure 3. The half-bridge MMC converter consists of three symmetrical phase units, each phase includes upper and lower bridge arms, and each bridge arm consists of several identical sub-phase units. The module and the bridge arm reactor are connected in series.

Existing studies have shown that the current of the MMC bridge arm is composed of three parts during normal operation, namely the AC side current, the DC side current and the bridge arm circulating current. The upper and lower bridge arm currents i _jp , i _jn (j=a,b,c) can be expressed as:

In the formula: I _ac is the amplitude of the AC side current, and I _cir is the amplitude of the circulating current of the bridge arm.

After the fault, the sub-module capacitors in each converter station quickly discharge to the fault point, causing the fault current of the DC line to rise sharply, and the DC component of the bridge arm current increases rapidly. At the same time, the capacitor discharge of the sub-module causes the equivalent voltage of the bridge arm to change, which in turn causes the injection current of the AC side to change.

After the DC side fault occurs, the current change of each bridge arm of the converter station is mainly caused by the change of the DC component and the change of the injection current of the AC side. Compared with the DC current and the AC side current, the bridge arm circulating current accounts for a small proportion in the bridge arm current, and its amplitude changes little in a short period of time before and after the fault, so it can be detected within a few milliseconds after the fault occurs. It is approximately considered that the amplitude of the bridge arm circulation does not change.

After a fault occurs on the DC side, the rectifier station and the converter station show different fault characteristics: the magnitudes of the DC current and the AC current after the rectifier station fault continue to increase, and the DC current and the AC current after the inverter station fault The amplitudes of all have to go through the process of first decreasing and then reversely increasing.

S2. Aiming at the DC line faults of the multi-terminal flexible DC grid, a single-terminal protection scheme based on the blocking time of the converter station is proposed

First, the pre-trip circuit breaker is selected according to the direction of the fault current variation; second, the action time of the protection is determined according to the blocking time of the converter station. On the basis of correctly selecting the pre-trip circuit breaker, the protection of each section of the line can complete the removal of the faulty line through the coordination of the action time.

(1) Fault current direction criterion and selection of pre-trip circuit breaker

The current change is used to reflect the change of the fault current. The selection of the positive direction is related to the polarity of the current flowing through the line, that is, the positive direction of the current change of the positive line is from the bus to the line, and the positive direction of the current change of the negative line is from the line. flow to the busbar. Define the rate of change of current as follows:

In the formula, i _k and i _k-1 represent the adjacent current sampling values within the sampling interval ΔT.

A current variation direction index S is defined, which represents the direction of the fault current variation measured at the end of the DC line.

S=1 means that the direction of the current change is positive, and the result of formula (2) is greater than 0 at this time;

S=-1 means that the direction of current variation is negative, and the result of formula (2) is less than 0;

S=0 means no current variation is detected, and the result of formula (2) is equal to 0 at this time.

Considering the normal fluctuation of the current on the DC line during the normal operation of the system, when the absolute value of the detected current change rate exceeds the set value k _set , the direction index changes, so the following criterion for the direction of the fault current change is proposed:

Combined with the analysis of the fault current characteristics of the DC line, each converter station can select the pre-trip circuit breaker according to the direction index S of the fault current variation measured at the end of the line. The schematic diagram of the pre-trip circuit breaker selection criterion is shown in Figure 4 . In Fig. 4, S _im and S _in are the direction indicators of the fault current variation detected at the circuit breakers CB _im and CB _in respectively, and the selection criterion of the pre-trip circuit breaker shown in formula (4) is proposed for each converter station :

(2) Calculation of protection action time based on converter station blocking time

On the basis of determining the pre-trip circuit breaker, the protection action time of each converter station depends on the blocking time of the converter station. On the premise of ensuring that the converter station is not blocked, this protection algorithm determines the protection action time in consideration of the circuit breaker action delay and action time margin. The specific calculation method is shown in formula (5):

t _act ＝t _block -t _cb -t _yd (5)

In the formula: t _block is the blocking time of the converter station, t _cb is the circuit breaker opening delay, t _yd is the protection action time margin.

Considering the action delay of the circuit breaker and the cooperation of the circuit breakers on both sides of the line, setting t _yd equal to t _cb can ensure the protection cooperation of each section to the greatest extent.

Considering that the bridge arm current of the inverter station has different fault characteristics from that of the rectifier station, a unified processing method is proposed for the inverter station and the virtual blocking time of the inverter station is defined, which can well reflect the severity of the fault. Assuming that the amplitude of the bridge arm circulation does not change after a fault occurs, the AC and DC components of the bridge arm current in the inverter station are analyzed and processed separately.

In order to make the inverter station show the effect that the received active power continues to increase after the fault, the change of the AC current after the fault is reversed, that is, the value after the fault is subtracted by twice the normal operating value, and the direction is changed. The processing expression is shown in formula (6):

i _ac '＝2i _ac0 -i _ac (6)

In the formula: i _ac ' is the virtual alternating current, i _ac0 is the alternating current during normal operation, and i _ac is the actual alternating current after the fault.

The magnitude of the fault current on the virtual AC side of the inverter station obtained after the direction change process continues to increase after the fault, which is consistent with the change trend of the AC side current of the rectifier station under the same initial current.

Considering the transmission of power on the DC side, in order to make the DC current of the inverter station show the same continuous increase effect as that of the rectifier station, the processing method of i _DC is the same as that of i _AC , and the expression of direction change processing is as follows: 7) as shown:

i _dc '＝2i _dc0 -i _dc (7)

Where: i _dc 'is the modified DC current, and i _dc0 is the DC current before the fault.

S3. Based on the waveform fitting prediction of the moving data window, each converter station predicts the blocking time of the converter station in real time through the fault data

Based on the analysis of the fault characteristics of the bridge arm current, it can be seen that after the DC side fault occurs, the change of each bridge arm current is mainly caused by the DC component and the AC component, and the change of the bridge arm circulating current is small, that is, Icir can be considered to be the same as that before the fault. same. Therefore, the DC current idc and the AC current Iacsin (ωt+θ1) are respectively fitted and predicted to obtain the bridge arm current waveform after the fault.

After the DC side fault occurs, the DC current at the outlet of the converter station can be directly measured. Therefore, for the DC component of the bridge arm current, the fitting prediction can be made directly based on the measured value of the DC current at the outlet of the converter station.

After a fault occurs on the DC side, the change of the injected current on the AC side is not only the change of the amplitude Iac, but also its frequency ω and phase θ1 will also change to a certain extent. In the control link of the converter, the AC current is converted into d and q axis components to participate in the control, and the changes of the d and q axis components are similar to the changes of the DC current. Therefore, in the fitting prediction of the AC current, the measured AC current can be converted into the form of the d and q axis components of the AC current through the Parker transformation, and then the id and iq are respectively fitted and predicted, and finally obtained by the Parker inverse transformation The fitting prediction results of the AC side current.

In the initial period of time after the DC side fault occurs, the DC current idc and the d, q axis components id and iq of the AC current do not increase strictly linearly, and their slopes are constantly changing. Therefore, the second-order polynomial fitting method is used to fit and predict the DC current component and AC current component in the bridge arm fault current. Since the change of the fault current growth rate is consistent, the moving data window is used to predict the fault current fitting.

The basic principle of this method is to adjust the basic data window used for prediction in real time as the sampling data is updated, perform subsequent waveform fitting prediction according to the fault current data in the latest data window, and then judge the blocking condition to determine the protection action time .

A schematic diagram of this method is shown in Figure 5. Firstly, according to the sampling data of the latest T _nh window length, curve fitting is performed to obtain the expression of the fault current, and then according to this expression, the value of the fault current with a subsequent length of t _yc is obtained. Combined with the formula (5) for calculating the protection action time, the fault current waveform prediction duration t _yc is set as t _yc = t _cb + t _yd , that is, the bridge with a time length of t _yc after each fitting prediction obtains the current moment Arm current value. If the predicted current value meets the blocking condition of the converter station, a trip signal should be issued immediately; if the predicted waveform does not meet the blocking condition of the converter station, the trip signal will not be issued, and the next prediction will be made after the next sampling data is obtained.

Figure 6a), b) and c) show the comparison between the predicted value of the bridge arm current and the real value obtained by using the prediction method based on the moving data window at 3ms, 5ms and 7ms after the fault respectively. It can be seen from the figure that when the predicted time is closer to the blocking time, the predicted value of the bridge arm current is more accurate. Compared with the method of long-term prediction with fixed data window, the prediction method based on moving data window has higher accuracy, and can obtain more accurate blocking time and protection action time. And when the fault current growth rate changes due to the action of the circuit breaker, this fault current prediction method based on the moving data window can effectively predict according to the latest fault current growth rate.

Combining the fault current waveform fitting prediction method based on the changing data window with the previous protection scheme based on the block time of the converter station and the processing method of the fault current waveform of the inverter station, the protection logic flow chart is shown in Figure 7.

Although the embodiments and drawings of the present invention are disclosed for the purpose of illustration, those skilled in the art can understand that various replacements, changes and modifications are possible without departing from the spirit and scope of the present invention and the appended claims Therefore, the scope of the present invention is not limited to what is disclosed in the embodiments and drawings.

Claims (1)

1. A single-ended quantity protection method suitable for a multi-ended flexible direct current system is characterized by comprising the following steps: the method comprises the following steps:

s1, analyzing the directional characteristic of a direct current line fault current and the fault characteristic of a bridge arm current after a fault occurs on a direct current side of a flexible direct current power grid fault;

(1) DC line fault current direction characteristic

The method comprises the steps that a four-terminal flexible direct current system model is used as a basis for analysis, and for a flexible direct current power grid of a true bipolar wiring line, no matter a bipolar short circuit, a pole-to-metal return line short circuit or a unipolar grounding short circuit fault occurs on a direct current side, a fault pole MMC can form a fault loop on the direct current side through a fault pole, the metal return line or the ground before a converter station is locked;

when a positive grounding fault or a bipolar short-circuit fault occurs on the line, the direction of the fault current variation on the positive line is from the bus to the line; when a negative earth fault or a bipolar short-circuit fault occurs on the line, the direction of the fault current variation on the negative line is from the line to the bus;

therefore, the positive direction of the fault current variation of the positive electrode line is from the bus to the line, and the positive direction of the fault current variation of the negative electrode line is from the line to the bus, so that the fault current variation of both sides of the fault line is positive no matter the positive electrode line or the negative electrode line has a fault;

in practical engineering, when the length difference of each section of the line of the direct-current power grid is not very large, the impedance difference of each section of the line is not very large, and for any bus, if a fault occurs on one outgoing line of the bus, the situation that the fault current variation directions on two outgoing lines of the bus are the same cannot occur; for the converter station connected with the bus, the direction of the fault current variation on the fault line measured by the converter station is determined to be positive, and the direction of the fault current variation on the other line is determined to be negative; therefore, when the directions of the fault current variation quantities on the two outgoing lines of a certain bus are the same, the line connected with the bus is not necessarily a fault line;

(2) Bridge arm current fault characteristic

In a typical half-bridge MMC converter topological structure, a half-bridge MMC converter consists of three symmetrical phase units, each phase comprises an upper bridge arm and a lower bridge arm, and each bridge arm is formed by connecting a plurality of identical sub-modules and bridge arm reactors in series;

the MMC bridge arm current consists of three parts, namely alternating current side current, direct current side current and bridge arm circulating current, and upper and lower bridge arm currents i _jp 、i _jn (j = a, b, c) may be respectively expressed as:

in the formula: i is _ac Is the amplitude of the AC side current, I _cir Is the bridge arm circulating current amplitude;

after a fault, sub-module capacitors in each converter station quickly discharge to a fault point, so that fault current of a direct-current line rapidly rises, direct-current components of bridge arm current rapidly increase, and meanwhile, the equivalent voltage of a bridge arm is changed due to discharge of the sub-module capacitors, so that the alternating-current side injection current is changed;

after the fault of the direct current side occurs, the change of the current of each bridge arm of the converter station is mainly caused by the change of a direct current component and the change of an injection current of the alternating current side; the proportion of the bridge arm circulating current in the bridge arm current is smaller than that of the direct current and the alternating current, and the amplitude of the bridge arm circulating current is not changed greatly in a short time before and after a fault, so that the amplitude of the bridge arm circulating current can be approximately considered to be not changed in a few milliseconds after the fault occurs;

after a fault occurs on the direct current side, the rectifier station and the converter station show different fault characteristics: the amplitude values of the direct current and the alternating current of the rectifying station after the fault are continuously increased, and the amplitude values of the direct current and the alternating current of the inverter station after the fault are subjected to the processes of firstly reducing and then reversely increasing;

s2, aiming at the direct-current line fault generated by the multi-terminal flexible direct-current power grid, a single-terminal quantity protection scheme based on the converter station blocking time is provided;

firstly, selecting a pre-trip circuit breaker according to the fault current variation direction; secondly, determining the action time of protection according to the locking time of the converter station, and on the basis of correctly selecting a pre-trip circuit breaker, protecting each section of circuit to complete the removal of a fault circuit by the cooperation of the action time;

(1) Fault current direction criterion and pre-trip circuit breaker selection

Adopt the change situation of current variation volume reaction fault current, its positive direction is selected and is related to with the circuit polarity that the electric current flows through, and the current variation volume positive direction of anodal circuit is from the bus flow direction circuit promptly, and the current variation volume positive direction of negative pole circuit is from the circuit flow direction bus, and the definition electric current rate of change is as follows:

in the formula: i.e. i _k And i _k-1 Representing adjacent current sample values within a sample interval Δ T;

defining a current variation direction index S which represents the fault current variation direction measured by the end part of the direct current circuit;

s =1 represents that the direction of the current variation is positive, and the result of equation (2) is greater than 0;

s = -1 represents that the direction of the current change amount is negative, and the result of the formula (2) is smaller than 0;

s =0 represents that no current change is detected, when the result of equation (2) equals 0;

considering the normal fluctuation condition of the current on the direct current line when the system normally operates, when the absolute value of the detected current change rate exceeds the setting value k _set And in time, the direction index changes, and the following fault current change quantity direction criterion is provided:

by combining the analysis of the fault current characteristics of the direct current line, each converter station can select the pre-trip circuit breaker according to the fault current variation direction index S measured at the end part of the line, and the pre-trip circuit breaker selection criterion S _im 、S _in Are respectively a circuit breaker CB _im 、CB _in And (3) extracting the selection criterion of the pre-trip circuit breaker shown in the formula (4) for each converter station according to the detected fault current variation direction index:

(2) Protection action time calculation based on converter station lockout time

On the basis of determining the pre-tripping circuit breaker, the protection action time of each converter station depends on the locking time of the converter station, on the premise of ensuring that the converter stations are not locked, the protection action time is determined under the condition of considering the action delay and the action time margin of the circuit breaker, and the specific calculation method is as shown in formula (5):

t _act ＝t _block -t _cb -t _yd (5)

in the formula: t is t _block Is the lock-out time of the converter station;

t _cb delaying the opening of the circuit breaker;

t _yd a protection action time margin;

considering the action delay of the circuit breaker and the matching problem of the circuit breakers at two sides of the circuit, t is _yd Is set to be and _cb equality, the coordination of each section of protection can be ensured to the maximum extent;

considering that bridge arm current of the inverter station has different fault characteristics from those of the rectifier station, a unified processing method is provided for the inverter station, virtual locking time of the inverter station is defined, and the fault severity can be well reflected; assuming that the circulating current amplitude of the bridge arm is not changed after the fault occurs, respectively analyzing and processing the AC and DC components of the bridge arm current of the inverter station;

in order to enable the inverter station to show the effect of continuously increasing the received active power after the fault, the variable quantity of the alternating current after the fault is processed in a reverse way, namely the value after the fault is subtracted from the double normal operation value, and the expression of the direction change processing is shown as the formula (6):

i _ac '＝2i _ac0 -i _ac (6)

in the formula: i.e. i _ac ' is a virtual alternating current;

i _ac0 alternating current for normal operation;

i _ac actual alternating current after the fault;

the amplitude of the fault current on the virtual alternating current side of the inverter station obtained after the direction change processing after the fault is continuously increased and is consistent with the change trend of the current on the alternating current side of the rectifier station under the same initial current;

in order to make the direct current of the inverter station show the same effect of continuous increase as that of the rectifier station in consideration of the transmission condition of the direct-current side power, the direct-current side power converter is used for the i _dc And i _ac Similarly, the direction-changing processing expression is shown as formula (7):

i _dc '＝2i _dc0 -i _dc (7)

in the formula: i.e. i _dc ' is a modified direct current;

i _dc0 the direct current is the direct current before the fault;

s3, predicting waveform fitting based on the mobile data window, and predicting the locking time of the convertor station in real time through fault data by the convertor station;

based on the analysis of the fault characteristics of the bridge arm currents, after the direct-current side fault occurs, the change of each bridge arm current is mainly caused by a direct-current component and an alternating-current component, the change of the bridge arm circulation current is small, namely I can be considered as _cir Same as before the failure; thus to the direct current i _dc And an alternating current I _ac sin(ωt+θ ₁ ) Respectively carrying out fitting prediction to obtain the current waveforms of the bridge arms after the faults;

after the direct current side fault occurs, the direct current at the outlet of the converter station can be directly measured, so that the direct current component of the bridge arm current can be directly subjected to fitting prediction according to the direct current measurement value at the outlet of the converter station;

after the fault of the direct current side occurs, the change of the injection current of the alternating current side is not only the amplitude I _ac Of frequency ω and phase θ ₁ Some degree of variation may also occur; in the control link of the converter, alternating current is converted into d and q axis components to participate in control, and the change conditions of the d and q axis components are similar to the change conditions of direct current; therefore, in the fitting prediction of the alternating current, the measured alternating current can be converted into the form of the d-axis component and the q-axis component of the alternating current through park transformation, and then the fitting prediction of the alternating current is carried out on i _d 、i _q Respectively carrying out fitting prediction, and finally obtaining a fitting prediction result of the alternating current side current through park inverse transformation;

during the first period of time after the DC side fault occurs, the DC current i _dc And d, q-axis components i of the alternating current _d 、i _q The slope of the linear gradient is not strictly linear increase, and is continuously changed;

a second-order polynomial fitting method is adopted to respectively carry out fitting prediction on a direct current component and an alternating current component in the fault current of the bridge arm, and because the change of the fault current acceleration rate has continuity, a mobile data window is adopted to carry out the fitting prediction on the fault current;

first according to the latest T _nh Carrying out curve fitting on the sampled data of the window length to obtain an expression of fault current, and then obtaining a subsequent length t according to the expression _yc The fault current value of (a); the time length t of the fault current waveform is predicted by combining a protection action time calculation formula provided by the formula (5) _yc Is set to t _yc ＝t _cb +t _yd I.e. the time length t after the current moment is obtained by fitting prediction every time _yc The bridge arm current value of (1); if the predicted current value reaches the converter station locking condition, a tripping signal is sent out immediately; and if the predicted waveform does not reach the converter station locking condition, not sending a trip signal, and performing next prediction after the next sampling data is obtained.

Images