Fast time domain distance protection method and system based on zero sequence current random correction

Abstract

The invention discloses a rapid time domain distance protection method and system based on zero sequence current random correction, and the method comprises the steps: carrying out the sampling of a new energy grid-connected line system, and obtaining a three-phase voltage, a three-phase current and a zero sequence current at one end of a line; if the zero-sequence current is larger than the threshold value, zero-sequence current compensation is carried out, and if the zero-sequence current is smaller than the threshold value, random correction is carried out on the zero-sequence current, and zero-sequence current compensation is not carried out at present; constructing a voltage / current information matrix in a unit data window by utilizing voltage / current sampling of the unit data window, calculating a single fitting distance result and a fitting error by utilizing a least square method, and calculating a weight; calculating a final accumulative weighted distance result; if the accumulated distance calculation result is continuously stable and smaller than a fixed value, judging that an internal fault occurs, and performing distance protection action tripping; otherwise, judging that an external fault occurs, and enabling distance protection not to act. Compared with the prior art, the weighted summation type fast time domain distance protection based on zero sequence current random correction is realized.

Description

Technical Field

[0001] The present invention relates to the field of relay protection of AC power transmission systems and new energy grid-connected systems, and in particular to a single-ended distance protection method. Background Art

[0002] In order to cope with environmental pollution and energy shortage, countries around the world are accelerating the process of replacing traditional fossil energy with renewable energy. As the main power source, new energy will become an inevitable trend in the development of new power systems in the future. However, compared with traditional synchronous power sources, new energy power sources have significant response characteristics such as weak feedback and frequency deviation, which seriously affects the performance of traditional protection principles based on power frequency quantities. Among them, distance protection has been widely used in actual power systems as the main protection or backup protection of some new energy grid-connected lines. However, the different fault response characteristics of new energy power sources will cause serious degradation of their operating performance. The high harmonics and frequency deviation characteristics of wind power systems make the extraction of power frequency distance protection phase quantities have significant errors in the initial fault period, which makes it unable to operate reliably.

[0003] Traditional time domain distance protection uses the inherent characteristics of the line model to design protection criteria, which is not affected by the back-side power supply in principle. However, the transient error of traditional time domain distance protection is large under the action of high harmonics. The fast time domain distance protection algorithm based on least squares fitting error improves the accuracy and convergence speed of the algorithm through weighted summation and cumulative calculation. However, since the actual protection device uses integer values ​​for storage and calculation, integer rounding deviation will occur. In particular, the zero-sequence current during normal operation is small and will be rounded to a constant value in the device, which will cause the weighted summation type fast time domain distance protection device to malfunction during normal operation. The prior art urgently needs a fast time domain distance protection algorithm that is not affected by rounding deviation. Summary of the invention

[0004] In view of the above-mentioned defects of the existing technology of single-ended distance protection, the present invention proposes a fast time domain distance protection method and system based on random correction of zero-sequence current, which realizes weighted summation type fast time domain distance protection based on random correction of zero-sequence current.

[0005] The present invention is achieved by the following technical solutions:

[0006] In a first aspect, the present invention provides a fast time domain distance protection method based on random correction of zero-sequence current, comprising:

[0007] Sampling the new energy grid-connected line system to obtain the three-phase voltage, three-phase current and zero-sequence current at one end of the line;

[0008] When the zero-sequence current is less than the threshold value, all zero-sequence currents affected by integer rounding errors are randomly corrected;

[0009] Using the voltage / current sampling of the unit data window, a voltage / current information matrix in the unit data window is constructed, and a single fitting distance result and a fitting error are calculated using the least squares method according to the obtained voltage / current information matrix in the unit data window, and a weight is obtained according to the fitting error;

[0010] Using the single fitting result and weight in the unit data window, the final cumulative weighted distance result is calculated by weighted summation in the cumulative data window;

[0011] Analyze the cumulative distance calculation results and determine the type of fault based on the cumulative distance calculation results.

[0012] In some embodiments, if the zero-sequence current is greater than the threshold value, zero-sequence current compensation is performed, and the step of performing zero-sequence current compensation further includes:

[0013] According to the three-phase voltage, three-phase current and zero-sequence current, the measured current is compensated for zero-sequence current to obtain the compensated resistance current i mr and the compensated inductor current i ml , as shown below:

[0014] i mr =i m +3k r i0,i ml =i m +3k l i0

[0015] In the formula, k r =(r0–r1) / 3r1 is the zero-sequence resistance compensation coefficient, r0 and r1 are the zero-sequence resistance and positive-sequence resistance of the line per unit length, respectively, k l =(l0–l1) / 3l1 is the zero-sequence inductance compensation coefficient, l0 and l1 are the zero-sequence inductance and positive-sequence inductance of the line per unit length respectively; i mr Represents the sampling current after resistor zero-sequence compensation; i ml Indicates the sampled current after zero-sequence compensation of the inductor;

[0016] Thus, the compensated resistance current i at sampling point n is obtained. mr (n) and the compensated inductor current i ml (n).

[0017] In some implementations, the voltage and current are sampled using a unit data window determined by s preceding points of sampling point n to construct a voltage information matrix U(n) and a current information matrix I(n), as shown in the following formula:

[0018]

[0019] In the formula, pm () is the line voltage drop per unit length at sampling points n-s+2, n-s+3, ..., n-2, n-1, i0() is the zero-sequence current at sampling points n-s+2, n-s+3, ..., n-1, u m () is the sampling voltage of sampling points n-s+2, n-s+3, ..., n-1.

[0020] In some embodiments, a single fitting result β(n) is shown as follows:

[0021]

[0022] In the formula, and are the fault distance and equivalent transition resistance obtained by least square fitting, I(n) T is the current information transposed matrix;

[0023] The weight W(n) is the fitting error E rr The reciprocal of (n), W(n) = 1 / E rr (n);

[0024] Fitting error E rr (n) is shown in the following formula:

[0025]

[0026] In the formula, . / represents the division of the elements at corresponding positions in the vector. ||·||2 represents the 2-norm of the vector.

[0027] In some embodiments, using a single fitting result and a weight in a unit data window, calculating the final cumulative weighted distance result by weighted summation in the cumulative data window further includes: using a cumulative data window determined by q preceding points including the sampling point n to perform data sampling, and weighted distance As shown below:

[0028]

[0029] Where j is the sampling point number, from n-q+1 to n. W(j) and They represent the weight and fault distance calculated for the j-th unit data window respectively.

[0030] In a second aspect, the present invention provides a fast time-domain distance protection system based on random correction of zero-sequence current, comprising a sampling module, a zero-sequence current random correction module, a cumulative weighted distance calculation module and a distance protection analysis module; wherein the sampling module is used to sample the new energy grid-connected line system to obtain the three-phase voltage, three-phase current and zero-sequence current at one end of the line; the zero-sequence current random correction module is used to perform random correction on all zero-sequence currents affected by integer rounding deviation when the zero-sequence current is less than a threshold value; the cumulative weighted distance calculation module is used to construct a voltage / current information matrix within the unit data window using the voltage / current sampling of the unit data window, calculate a single fitting distance result and a fitting error using the least squares method based on the obtained voltage / current information matrix within the unit data window, and calculate the weight, and use the single fitting result and weight within the unit data window to calculate the final cumulative weighted distance result by weighted summation within the cumulative data window; the distance protection analysis module is used to analyze the cumulative distance calculation results and determine the type of fault according to the cumulative distance calculation results.

[0031] In some embodiments, the step of performing zero-sequence current compensation further includes:

[0032] According to the three-phase voltage, three-phase current and zero-sequence current, the measured current is compensated for zero-sequence current to obtain the compensated resistance current i mr and the compensated resistance current i ml , as shown below:

[0033] i mr =i m +3k r i0,i ml =i m +3k l i0

[0034] In the formula, k r =(r0–r1) / 3r1 is the zero-sequence resistance compensation coefficient, r0 and r1 are the zero-sequence resistance and positive-sequence resistance of the line per unit length, respectively, k l =(l0–l1) / 3l1 is the zero-sequence inductance compensation coefficient, l0 and l1 are the zero-sequence inductance and positive-sequence inductance of the line per unit length respectively; i mr Represents the sampling current after resistor zero-sequence compensation; i ml Represents the sampled current after inductor zero-sequence compensation.

[0035] Thus, the compensated resistance current i at sampling point n is obtained. mr (n) and the compensated inductor current i ml (n).

[0036] In some implementations, the voltage and current are sampled using a unit data window determined by s preceding points of sampling point n to construct a voltage information matrix U(n) and a current information matrix I(n), as shown in the following formula:

[0037]

[0038] In the formula, p m () is the line voltage drop per unit length at sampling points n-s+2, n-s+3, ..., n-2, n-1, i0() is the zero-sequence current at sampling points n-s+2, n-s+3, ..., n-1, u m () is the sampling voltage of sampling points n-s+2, n-s+3, ..., n-1.

[0039] In some embodiments, a single fitting result β(n) is shown as follows:

[0040]

[0041] In the formula, and are the fault distance and equivalent transition resistance obtained by least square fitting, I(n) T is the current information transposed matrix;

[0042] The weight W(n) is the fitting error E rr The reciprocal of (n), W(n) = 1 / E rr (n);

[0043] Fitting error E rr (n) is shown in the following formula:

[0044]

[0045] In the formula, . / represents the division of the elements at corresponding positions in the vector. ||·||2 represents the 2-norm of the vector.

[0046] In some embodiments, using a single fitting result and a weight in a unit data window, calculating the final cumulative weighted distance result by weighted summation in the cumulative data window further includes: using a cumulative data window determined by q preceding points including the sampling point n to perform data sampling, and weighted distance As shown below:

[0047]

[0048] Where j is the sampling point number, from n-q+1 to n. W(j) and They represent the weight and fault distance calculated for the j-th unit data window respectively.

[0049] Compared with the prior art, the advantages and positive technical effects achieved by the present invention are as follows:

[0050] 1) The zero-sequence current compensation procedure is designed to ensure that the zero-sequence current is not affected by rounding deviation, thereby realizing a fast time-domain distance protection algorithm based on the compensation current;

[0051] 2) Through random correction of zero-sequence current, it is ensured that all zero-sequence currents affected by integer rounding deviation can be randomly corrected to ensure that the corrected zero-sequence current is within a reasonable range, thereby ensuring the efficient and reliable operation of fast time domain distance protection in actual relay protection devices;

[0052] 3) Using the single fitting result and weight within the unit data window, the final cumulative weighted distance result is calculated by weighted summation within the cumulative data window, which provides an important criterion for the judgment analysis of fast time domain distance protection analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 It is an overall flow chart of a fast time domain distance protection method based on random correction of zero-sequence current of the present invention;

[0054] Figure 2 It is a module diagram of a fast time domain distance protection system based on random correction of zero-sequence current of the present invention;

[0055] Figure 3 It is a specific implementation flow chart of a fast time domain distance protection method based on random correction of zero-sequence current of the present invention;

[0056] Figure 4 is a schematic diagram of a test system according to an embodiment of the present invention;

[0057] Figure 5 It is a schematic diagram of the zero-sequence current rounding deviation result and the zero-sequence current random correction effect;

[0058] Figure 6 It is the recording of the protection device and the tripping situation when a metallic grounding fault of phase A occurs 8km away from the M end of the line. DETAILED DESCRIPTION

[0059] The technical solution will be clearly described below in conjunction with the accompanying drawings and embodiments.

[0060] like Figure 1 As shown, the overall process of a fast time domain distance protection method based on zero-sequence current random correction of the present invention is as follows:

[0061] S1. Sampling the new energy grid-connected line system to obtain the three-phase voltage, three-phase current and zero-sequence current at one end of the line;

[0062] S2. If the zero-sequence current is greater than the threshold value, zero-sequence current compensation is performed. If the zero-sequence current is less than the threshold value, the zero-sequence current is randomly corrected and zero-sequence current compensation is not performed at present.

[0063] S3, using the voltage / current sampling of the unit data window, constructing the voltage / current information matrix in the unit data window, calculating the single fitting distance result and the fitting error using the least squares method according to the obtained voltage / current information matrix in the unit data window, and calculating the weight according to the fitting error;

[0064] S4, using the single fitting result and weight in the unit data window, calculating the final cumulative weighted distance result by weighted summation in the cumulative data window;

[0065] S5. If the cumulative distance calculation result is stably less than the set value for 5 milliseconds, it is judged as an internal fault and the distance protection will trip; otherwise, it is judged as an external fault and the distance protection will not operate.

[0066] like Figure 2 As shown, a fast time domain distance protection system based on random correction of zero-sequence current of the present invention includes a sampling module, a zero-sequence current random correction module, a cumulative weighted distance calculation module and a distance protection analysis module. Among them, the sampling module is used to sample the new energy grid-connected line system to obtain the three-phase voltage, three-phase current and zero-sequence current at one end of the line. The zero-sequence current random correction module is used to perform zero-sequence current compensation if the zero-sequence current is greater than the threshold value, and if the zero-sequence current is less than the threshold value, the zero-sequence current is randomly corrected, and zero-sequence current compensation is not performed at present. The cumulative weighted distance calculation module is used to construct the voltage / current information matrix in the unit data window by using the voltage / current sampling of the unit data window, and calculate the single fitting distance result and fitting error and calculate the weight according to the voltage / current information matrix in the unit data window by using the least squares method, and calculate the final cumulative weighted distance result by weighted summation in the cumulative data window using the single fitting result and weight in the unit data window. The distance protection analysis module is used to analyze that if the cumulative distance calculation result is stably less than a set value for 5 milliseconds, it is determined that an internal fault has occurred and the distance protection will trip; otherwise, it is determined that an external fault has occurred and the distance protection will not operate.

[0067] like Figure 3 As shown, the implementation process of a weighted summation type fast time domain distance protection method based on zero-sequence current random correction of the present invention is specifically described as follows:

[0068] Step 1: Sample the new energy grid-connected line system to obtain the three-phase voltage u at each sampling point n ma (n),u mb (n),umc (n) and the three-phase current i ma (n), i mb (n), i mc (n), calculate the zero-sequence current i0(n) at sampling point n, as shown in the following formula:

[0069] i0(n)=i ma (n)+i mb (n)+i mc (n)

[0070] Step 2: Compare the absolute value of the zero-sequence current i0(n) at the sampling point n with the threshold value γ. If the absolute value is greater than γ, execute step 3 to perform zero-sequence current compensation; if the absolute value is less than or equal to γ, perform random correction on the zero-sequence current: keep the sign unchanged, change the absolute value to a random value in the range of 0 to Γ, skip step 3, and execute step 4; wherein, the threshold value γ is selected to be greater than the zero-sequence current that is uniformly rounded to an integer value in the actual relay protection device, so as to ensure that all zero-sequence currents affected by integer rounding deviations can be randomly corrected, and can be specifically selected as one percent of the rated current amplitude during normal operation of the system; the random correction upper limit value Γ used for random correction of zero-sequence current can be selected as the maximum value of the zero-sequence current after adding 20dB noise during normal operation, so as to ensure that the corrected zero-sequence current is within a reasonable range;

[0071] Step 3: Perform zero-sequence current compensation on the measured current according to the three-phase voltage, three-phase current and zero-sequence current in step 1 to obtain the compensated resistance current i mr and the compensated inductor current i ml , as shown below:

[0072] i mr =i m +3k r i0,i ml =i m +3k l i0

[0073] In the formula, k r =(r0–r1) / 3r1 is the zero-sequence resistance compensation coefficient, r0 and r1 are the zero-sequence resistance and positive-sequence resistance of the line per unit length, respectively, k l =(l0–l1) / 3l1 is the zero-sequence inductance compensation coefficient, l0 and l1 are the zero-sequence inductance and positive-sequence inductance of the line per unit length respectively; i mr Represents the sampling current after resistor zero-sequence compensation; i ml Indicates the sampled current after zero-sequence compensation of the inductor;

[0074] Thus, the compensated resistance current i at sampling point n is obtained. mr (n) and the compensated inductor current iml (n);

[0075] Step 4: Use the unit data window determined by the s preceding points of the sampling point n to sample the voltage and current to construct the voltage information matrix U(n) and the current information matrix I(n), as shown in the following formula:

[0076]

[0077] In the formula, p m () is the line voltage drop per unit length at sampling points n-s+2, n-s+3, ..., n-2, n-1, i0() is the zero-sequence current at sampling points n-s+2, n-s+3, ..., n-1, u m () is the sampling voltage of sampling points n-s+2, n-s+3, ..., n-1, p m (n) is shown in the following formula:

[0078]

[0079] Where Δt is the sampling interval, which is the sampling frequency f s The reciprocal of Δt = 1 / f s ;

[0080] Step 5: Using the voltage and current information matrices U(n) and I(n) described in step 4, use the least squares algorithm to calculate the single fitting result β(n) and the corresponding fitting error E rr (n). The number of data window points s and the fixed value unit data window length t s Related, s = t s / Δt; where

[0081] The single fitting result β(n) is shown as follows:

[0082]

[0083] In the formula, and are the fault distance and equivalent transition resistance obtained by least square fitting, I(n) T is the current information transposed matrix;

[0084] Fitting error E rr (n) is shown in the following formula:

[0085]

[0086] In the formula, the . / symbol represents the division of the elements at corresponding positions in the vector, and ||·||2 represents the 2-norm of the vector;

[0087] Step 6: Use the accumulated data window determined by q previous points including sampling point n to sample data and calculate the weight W(n) and weighted distance

[0088] The weight W(n) is the fitting error E described in step 5 rr The reciprocal of (n), W(n) = 1 / E rr (n);

[0089] Weighted distance As shown below:

[0090]

[0091] Where j is the sampling point number, from n-q+1 to n. W(j) and They represent the weight and fault distance calculated for the j-th unit data window respectively;

[0092] Specifically, the cumulative data window number q and the fixed value cumulative data window length t q Regarding, q = t q / D t ;

[0093] Step 7: The weighted distance obtained in step 6 Less than the fixed value x within 5ms set , and the difference between the maximum and minimum values ​​is less than 0.1 km, the distance protection trips and the fault handling procedure ends; otherwise, let n = n + 1, return to step 1, and repeat the above process using n + 1 sampling points until the weighted distance Meet the distance protection action criterion; the distance protection action criterion is as follows:

[0094] Continuous within 5ms:

[0095] In summary, the present invention realizes a weighted summation type fast time domain distance protection based on random correction of zero-sequence current, performs random correction on zero-sequence current less than the threshold value and locks its zero-sequence compensation. According to the information matrix composed of sampling values, the single fitting distance and the cumulative weighted distance are calculated, and the protection criterion is constructed according to the relationship between the weighted distance and the setting distance, realizing a single-ended time domain distance protection strategy that is not affected by zero-sequence current rounding deviation and new energy characteristics.

[0096] like Figure 4As shown, it is a test system of an embodiment of the present invention, specifically a wind farm grid-connected system via a 110kV line established through a real-time digital simulation platform (RTDS). The wind farm consists of 20 double-fed wind turbines with a rated capacity of 3MW and a total capacity of 60MW. After two-stage boosting, it is sent to the AC main grid by a 110kV line, and the total length of the line is 15km. The line protection M on the wind farm side is taken as the observation object, and the sampling rate is 4kHz. The parameters of the line are: line length 15km, positive and negative sequence resistance: 0.1111Ω / m; positive and negative sequence inductance is 1.2685mH / km; positive and negative sequence capacitance is: 0.01629F / km; zero sequence resistance: 0.3334Ω / m; zero sequence inductance is 3.552mH / km; zero sequence capacitance is: 5.2F / km. The unit fitting data window length is selected as 2.5ms, and the cumulative data window length is selected as 10ms. The rated current amplitude during normal system operation is about 0.5 kA, so the threshold value for random correction of zero-sequence current is g = 1% * 0.5 = 5 * 10 -3 kA. The upper limit of random correction is 10 -1 kA. 80% of the total length of the protection line of section I, i.e., the fixed value of the distance of section I x set =12km.

[0097] like Figure 5 As shown in the figure, taking the A-phase metallic grounding fault 8km away from the M terminal as an example, the actual zero-sequence current i of the system during normal operation before the fault is 0re , zero sequence current i affected by the rounding deviation of the protection device 0err and the randomly corrected zero-sequence current i 0rd Respectively Figure 5 (a) Figure 5 (b) and Figure 5 (c) It can be seen that the effect of zero-sequence current correction can change the constant zero-sequence current affected by the rounding error into a fluctuating zero-sequence current similar to that during normal operation.

[0098] like Figure 6 As shown, the fault time is 0ms, and the voltage and current sampled by the M side protection from 50ms before the fault to 50ms after the fault are as follows: Figure 6 (a) and Figure 6 The tripping signals of the conventional time domain distance protection algorithm that does not use the random zero-sequence current and the algorithm of the present invention are trip0 and trip1 respectively. c ,like Figure 6 It can be seen that the algorithm of the present invention can solve the problem of false operation during normal operation of protection caused by the rounding deviation of zero-sequence current of actual device, and can correctly judge the actual line fault.

[0099] In addition, based on similar inventive concepts, an embodiment of the present invention also provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the above method when executing the computer program.

[0100] It should be noted that, although the technical solutions have been shown and described with reference to specific exemplary embodiments of the present invention, those skilled in the art should understand that the present invention is not limited to the above-mentioned implementation modes, and any replacements, improvements and supplements made without departing from the spirit of the present invention and without creative labor shall fall within the scope of protection of the present invention.

[0101] Based on the embodiments of the present invention, all other embodiments and technical replacements and modifications of the embodiments obtained by ordinary technicians in the field without departing from the spirit of the present invention and without making creative work shall fall into the protection scope of the present invention.

Claims

1. A fast time domain distance protection method based on random correction of zero-sequence current, characterized in that: The following steps are involved: Sampling the new energy grid-connected line system to obtain the three-phase voltage, three-phase current and zero-sequence current at one end of the line; When the zero-sequence current is less than the threshold value, all zero-sequence currents affected by integer rounding errors are randomly corrected; Using the voltage / current sampling of the unit data window, a voltage / current information matrix in the unit data window is constructed, and a single fitting distance result and a fitting error are calculated using the least squares method according to the obtained voltage / current information matrix in the unit data window, and a weight is obtained according to the fitting error; Using the single fitting result and weight in the unit data window, the final cumulative weighted distance result is calculated by weighted summation in the cumulative data window; Analyze the cumulative distance calculation results and determine the type of fault based on the cumulative distance calculation results.

2. A fast time domain distance protection method based on zero-sequence current random correction according to claim 1, characterized in that: If the zero-sequence current is greater than the threshold value, zero-sequence current compensation is performed, and the steps of zero-sequence current compensation further include: According to the three-phase voltage, three-phase current and zero-sequence current, the measured current is compensated for zero-sequence current to obtain the compensated resistance current i mr and the compensated inductor current i ml , as shown below: i mr =i m +3k r i0,i ml =i m +3k l i0 In the formula, k r =(r0–r1) / 3r1 is the zero-sequence resistance compensation coefficient, r0 and r1 are the zero-sequence resistance and positive-sequence resistance of the line per unit length, respectively, k l =(l0–l1) / 3l1 is the zero-sequence inductance compensation coefficient, l0 and l1 are the zero-sequence inductance and positive-sequence inductance of the line per unit length respectively; i mr Represents the sampling current after resistor zero-sequence compensation; i ml Indicates the sampled current after zero-sequence compensation of the inductor; Thus, the compensated resistance current i at sampling point n is obtained. mr (n) and the compensated inductor current i ml (n).

3. The fast time domain distance protection method based on zero-sequence current random correction according to claim 1 is characterized in that: The voltage and current are sampled using the unit data window determined by the s preceding points of the sampling point n to construct the voltage information matrix U(n) and the current information matrix I(n), as shown in the following formula: In the formula, p m () is the line voltage drop per unit length at sampling points n-s+2, n-s+3, ..., n-2, n-1, i0() is the zero-sequence current at sampling points n-s+2, n-s+3, ..., n-1, u m () is the sampling voltage of sampling points n-s+2, n-s+3, ..., n-1.

4. A fast time domain distance protection method based on zero-sequence current random correction according to claim 1, characterized in that: The single fitting result β(n) is shown as follows: In the formula, and are the fault distance and equivalent transition resistance obtained by least square fitting, I(n) T is the current information transposed matrix; The weight W(n) is the fitting error E rr The reciprocal of (n), W(n) = 1 / E rr (n), fitting error E rr (n) is shown in the following formula: In the formula, . / represents the division of the elements at corresponding positions in the vector. ||·||2 represents the 2-norm of the vector.

5. The fast time domain distance protection method based on zero-sequence current random correction according to claim 1 is characterized in that: Using the single fitting result and weight in the unit data window, the final cumulative weighted distance result is calculated by weighted summation in the cumulative data window, further comprising: using the cumulative data window determined by q preceding points including the sampling point n to perform data sampling, and calculating the weighted distance As shown below: Where j is the sampling point number, which ranges from n-q+1 to n. W(j) and They represent the weight and fault distance calculated for the j-th unit data window respectively.

6. A fast time domain distance protection system based on random correction of zero-sequence current, characterized in that: It includes a sampling module, a zero-sequence current random correction module, a cumulative weighted distance calculation module and a distance protection analysis module; wherein the sampling module is used to sample the new energy grid-connected line system to obtain the three-phase voltage, three-phase current and zero-sequence current at one end of the line; the zero-sequence current random correction module is used to perform random correction on all zero-sequence currents affected by integer rounding deviation when the zero-sequence current is less than a threshold value; the cumulative weighted distance calculation module is used to construct a voltage / current information matrix in the unit data window by using the voltage / current sampling of the unit data window, calculate the single fitting distance result and the fitting error by using the least squares method according to the obtained voltage / current information matrix in the unit data window, and calculate the weight, and use the single fitting result and weight in the unit data window to calculate the final cumulative weighted distance result by weighted summation in the cumulative data window; the distance protection analysis module is used to analyze the cumulative distance calculation result and determine the type of fault according to the cumulative distance calculation result.

7. A fast time domain distance protection system based on zero-sequence current random correction according to claim 6, characterized in that: The steps of performing zero-sequence current compensation further include: According to the three-phase voltage, three-phase current and zero-sequence current, the measured current is compensated for zero-sequence current to obtain the compensated resistance current i mr and the compensated inductor current i ml , as shown below: i mr =i m +3k r i0,i ml =i m +3k l i0 In the formula, k r =(r0–r1) / 3r1 is the zero-sequence resistance compensation coefficient, r0 and r1 are the zero-sequence resistance and positive-sequence resistance of the line per unit length, respectively, k l =(l0–l1) / 3l1 is the zero-sequence inductance compensation coefficient, l0 and l1 are the zero-sequence inductance and positive-sequence inductance of the line per unit length respectively; i mr Represents the sampling current after resistor zero-sequence compensation; i ml Represents the sampled current after inductor zero-sequence compensation. Thus, the compensated resistance current i at sampling point n is obtained. mr (n) and the compensated inductor current i ml (n).

8. A fast time domain distance protection system based on zero-sequence current random correction according to claim 6, characterized in that: The voltage and current are sampled using the unit data window determined by the s preceding points of the sampling point n to construct the voltage information matrix U(n) and the current information matrix I(n), as shown in the following formula: In the formula, p m () is the line voltage drop per unit length at sampling points n-s+2, n-s+3, ..., n-2, n-1, i0() is the zero-sequence current at sampling points n-s+2, n-s+3, ..., n-1, u m () is the sampling voltage of sampling points n-s+2, n-s+3, ..., n-1.

9. A fast time domain distance protection system based on zero-sequence current random correction according to claim 7, characterized in that: The single fitting result β(n) is shown as follows: In the formula, and are the fault distance and equivalent transition resistance obtained by least square fitting, I(n) T is the current information transposed matrix. The weight W(n) is the fitting error E rr The reciprocal of (n), W(n) = 1 / E rr (n), fitting error E rr (n) is shown in the following formula: Here, the . / symbol represents the division of the elements at corresponding positions in the vector. ||·||2 represents the 2-norm of the vector.

10. A fast time domain distance protection system based on zero sequence current random correction according to claim 7, characterized in that: Using the single fitting result and weight in the unit data window, the final cumulative weighted distance result is calculated by weighted summation in the cumulative data window, further comprising: using the cumulative data window determined by q preceding points including the sampling point n to perform data sampling, and calculating the weighted distance As shown below: Where j is the sampling point number, from n-q+1 to n. W(j) and They represent the weight and fault distance calculated for the j-th unit data window respectively.

Images