CN104914351A - Area power network fault positioning method based on optimal wave velocity - Google Patents

Abstract

The invention relates to a fault location method for power system automation, in particular to a regional power grid fault location method based on optimal wave velocity. Based on genetic algorithm and signal spectrum analysis, the method uses the refracted waves of branch lines related to fault lines in the power grid to locate faults. The method includes: (1) analysis of grid structure and data matching; (2) construction of objective function, Generate variable constraints; (3) Analyze the signal spectrum of the refracted wave of the branch line to determine the frequency component of the signal and the range of the frequency correction coefficient; (4) Use the genetic algorithm to solve the optimal solution of the objective function to obtain the optimal wave velocity; (5) Get the location of the fault point according to the optimal wave velocity, and output the result. This method fully considers the influence of the frequency and path of each branch line, and compared with the existing regional power grid fault location algorithm, the accuracy is significantly improved.

Description

A Fault Location Method for Regional Power Grid Based on Optimal Wave Velocity

technical field

The invention relates to a fault location method for power system automation, in particular to a regional power grid fault location method based on optimal wave velocity.

Background technique

After the transmission line fails, even if the reclosing is successful, line inspectors are required to find the fault point, and judge whether it can continue to operate or whether it needs to be powered off for maintenance according to the degree of damage caused by the fault, so as to eliminate hidden dangers. Therefore, quickly finding the fault point after a line fault (transmission line fault location technology is also called fault location technology) has become a key technology to ensure the safe and stable operation of the power grid.

Transmission line traveling wave fault location device (hereinafter referred to as traveling wave distance measuring device) can be divided into single-ended traveling wave method, double-ended traveling wave method and pulse method according to the different electrical quantities used. At present, the practical traveling wave distance measuring device mainly adopts the double-terminal traveling wave method, and its principle is as follows: the principle of the double-terminal traveling wave method is to use the first traveling wave head signal generated by the fault, and calculate the initial traveling wave of the fault to reach both ends of the line. The time difference is used to calculate the fault location, as shown in Figure 1, and the calculation formula is as follows:

$l_1=\frac{L^{\″}{(t_2-t_1)}^v}{2}$ ①;

In the above formula: l ₁ is the fault distance; t ₁ and t ₂ are the time for the initial traveling wave to reach both ends of the line, L'' is the total length of the line, and v is the propagation speed of the traveling wave. In the calculation of the double-ended traveling wave method, only The principle of identifying the initial wave head of the signal is simple and reliable, but it requires device data on both sides of the line, requires communication and GPS support, and the system configuration is relatively complicated.

At present, traveling wave fault location devices for transmission lines have been widely used in my country's power system, and domestic provinces such as Liaoning and Sichuan have established regional fault location systems consisting of a network of traveling wave distance measuring devices. As mentioned above, most of the existing traveling wave ranging devices use the double-ended traveling wave method, which involves multiple links such as sampling, GPS timing, and communication. Many intermediate links reduce the overall reliability of the system. If the side device fails, the system cannot work normally, which affects the overall reliability of the system.

After the transmission line fails, the transient traveling wave will be refracted to the branch line through the bus. The distance measuring device of the branch line is mainly designed to avoid refusal due to the algorithm and fixed value design. Therefore, the wave recording can be started in most fault situations. Regional power grid fault location creates conditions. Regional power grid fault location has the following engineering significance:

1) Improve system reliability, use multi-terminal data to complete ranging, and avoid fault location failure caused by device failure on one side;

2) To reduce system construction costs, the number of terminal configurations can be appropriately reduced.

At present, some achievements have been made in the fault location research of regional power grids, but in general, the existing regional power grid fault location methods mainly select the shortest path through the network structure diagram generation matrix, or use weight coefficients to refer to multiple sets of data for fault location. Location basically does not involve signal attenuation and waveform distortion during traveling wave transmission, which are important factors affecting the accuracy of fault location. In practical engineering applications, the accuracy of fault location in existing regional power grids is lower than that of the double-ended traveling wave method.

Contents of the invention

Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a regional power grid fault location method based on the optimal wave velocity. The method uses the refracted waves of the branch lines related to the fault line in the power grid to perform fault location calculations. The method fully considers Compared with the influence of the frequency and path of each branch line, the accuracy of the fault location algorithm of the existing regional power grid is significantly improved.

The object of the present invention is to adopt following technical scheme to realize:

The present invention provides a regional power grid fault location method based on optimal wave velocity. The improvement is that the method is based on genetic algorithm and signal spectrum analysis, and utilizes refracted waves of branch lines related to the faulty line in the power grid to perform fault location. Said method comprises the following steps:

(1) Grid structure analysis and data matching;

(2) Construct the objective function and generate variable constraints;

(3) Analyze the signal spectrum of the refracted wave of the branch line to determine the frequency component of the signal and the range of the frequency correction coefficient;

(4) Use the genetic algorithm to solve the optimal solution of the objective function to obtain the optimal wave velocity;

(5) Obtain the location of the fault point according to the optimal wave velocity, and output the result.

Further, in the step (1), after the fault of the transmission line, the transient traveling wave is refracted to each branch line through the bus, and the refracted wave signals of the branch lines are collected for each substation in the regional power grid. In practice, the transient traveling wave The influence of signal dispersion, select the data of the adjacent substation of the fault line;

The traveling wave distance measuring device of the transmission line adopts the double-terminal traveling wave method: the principle of the double-terminal traveling wave method is to use the first traveling wave head signal generated by the fault, and calculate the fault location by calculating the time difference between the fault initial traveling wave arriving at the two ends of the line, the expression as follows:

$l_1=\frac{L^{\″}{(t_2-t_1)}^v}{2}$ ①;

In the above formula: l ₁ is the fault distance; t ₁ and t ₂ are the time for the initial traveling wave to reach both ends of the line respectively, L'' is the total length of the line, and v is the propagation speed of the traveling wave;

Let the regional power grid include substations M, N and substations S ₁ , S ₂ ... S _n ; the substation M is connected to the substation N through the line L'; the substation S ₁ is connected to the substation M through the branch line 1; the substation S ₂ is connected to substation M through branch line 2; said substation S _n is connected to substation M through branch line n; branch line 1, branch line 2 and branch line n are connected in parallel.

Further, in the step (2), when the faulty line is the MN section, after the transient traveling wave reaches the M terminal of the substation, it is refracted to the branch lines 1~n through the bus bar of the M terminal of the substation, and the branch line of the S terminal of the substation is selected The refracted wave and the initial traveling wave at the N terminal of the substation form a double-terminal ranging; the S terminal of the substation includes the S ₁ , S ₂ and S _n terminals of the substation;

The calculation formulas for the double-terminal distance measurement composed of the substation S-terminal and substation N-terminal data are as follows:

$\left\{\begin{array}{c}{d}_1=\frac{L+L_1-(t_1^{\′}-t_2)*v_1}{2}\\ d_2=\frac{L+L_2-(t_2^{\′}-t_2)*v_2}{2}\\ ......\\ d_{\mathrm{no}}=\frac{L+L_{\mathrm{no}}-(t_{\mathrm{no}}^{\′}-t_2)*v_{\mathrm{no}}}{2}\end{array}\right.$ ②;

Among them, t′ ₁ , t' ₂ , and t' _n conform to the following formula:

t' _n ＝t ₁ +(L _n /v _n ) ③;

In formulas ② and ③, L, L ₁ , L ₂ , and L _n are the line lengths of the faulty line and branch lines 1, 2, and n respectively; t1, t2 are the moments when the initial traveling wave arrives at terminals M and N; t′ _{1 . _ _ _ _ _ _ _ _ _} v _n is the traveling wave velocity on branch lines 1, 2, and n; since d ₁ , d ₂ , and d _n are theoretically equal, the following conditions are satisfied:

$t_1^{\′}-t_2^{\′}-t_{\mathrm{no}}^{\′}-\frac{L_1}{v_0\×k_{f1}\×k_{\mathrm{the\; s}1}}+\frac{L_2}{v_0\×k_{f2}\×k_{\mathrm{the\; s}2}}+\frac{L_{\mathrm{no}}}{v_0\×k_{\mathrm{fn}}\×k_{\mathrm{sn}}}=0$ ④;

$t_1=t_1^{\′}-\frac{L_1}{v_0\×k_{f1}\×k_{\mathrm{the\; s}1}}$ ⑤;

In formula ④, v ₀ is the assumed reference wave velocity; k _f1 , k _f2 , k _fn are frequency correction coefficients, which are used to correct the difference in wave velocity caused by different frequency signals of refracted waves in each branch line; k _s1 , k _s2 , k _sn are respectively The path correction coefficient is used to correct the wave speed difference caused by the different transmission paths of the refracted waves of each branch line;

Build the objective function as follows:

$f(k_{f1},k_{f2},k_{\mathrm{fn}},k_{\mathrm{the\; s}1},k_{\mathrm{the\; s}2},k_{\mathrm{sn}},v_0)=t_1^{\′}-t_2^{\′}-t_{\mathrm{no}}^{\′}-\frac{L_1}{v_0\×k_{f1}\×k_{\mathrm{the\; s}1}}+\frac{L_2}{v_0\×k_{f2}\×k_{\mathrm{the\; s}2}}+\frac{L_{\mathrm{no}}}{v_0\×k_{\mathrm{fn}}\×k_{\mathrm{sn}}}$ ⑥;

The variable constraints of the objective function are as follows:

$\left\{\begin{array}{c}{v}_{\min }<v_0<v_{\max }\\ k_{f\min }<k_{\mathrm{fn}}<k_{f\max }\\ k_{\mathrm{the\; s}\min }<k_{\mathrm{sn}}<k_{\mathrm{the\; s}\max }\end{array}\right.$ ⑦;

In formula ⑦, v _max and v _min are the upper and lower limits of the wave velocity; k _fmax and k _fmin are the upper and lower limits of the frequency correction coefficient respectively; k _smax and k _smin are the upper and lower limits of the path correction coefficient respectively.

Further, the path correction coefficient is obtained by analyzing the line structure, and the frequency correction coefficient is obtained by analyzing the signal spectrum, that is, the frequency correction coefficient based on the signal spectrum analysis and the determination method of the correction coefficient based on the line structure;

The method of determining the correction coefficient based on the line structure includes: according to the line parameters such as the distance between the ground wires of each branch line, the structure of the tower, and the length, and comparing with the reference branch line, the path correction coefficient of each branch line is obtained;

The details are as follows: Calculate the characteristic impedance, attenuation coefficient and phase shift coefficient of each branch line according to the line parameters, and the expressions are as follows:

$\mathrm{\&alpha;}=\sqrt{[R_m{G}_m-{\mathrm{\&omega;}}^2{L}_m{C}_m+\sqrt{(R_m^2+{\mathrm{\&omega;}}^2{L}_m^2)(G_m^2+{\mathrm{\&omega;}}^2{C}_m^2)}]/2}$ ⑩;

$\mathrm{\&beta;}=\sqrt{[{\mathrm{\&omega;}}^2{L}_m{C}_m-R_m{G}_m+\sqrt{(R_m^2+{\mathrm{\&omega;}}^2{L}_m^2)(G_m^2+{\mathrm{\&omega;}}^2{C}_m^2)}]/2}$ ;

In the above formula, α is the attenuation coefficient, β is the phase shift coefficient; R _m , G _m , ω, and L _m respectively correspond to the mold resistance, inductance, conductance and capacitance of the line per unit length; when the phase shift coefficient of the line is obtained, the line The speed of wave propagation is as follows:

$v=\frac{\mathrm{\&omega;}}{\mathrm{\&beta;}}$ .

Taking the reference branch line as the standard, calculate the difference in transmission wave speed of the same frequency signal due to the different transmission paths of each branch line, and convert it into a relative value in the actual project, that is, the path correction coefficient.

Further, in the step (3), the Hilbert-Huang HHT transform marginal spectrum is used to analyze the spectrum of the refracted waves of each branch line, and the frequency of the high-frequency component of the refracted wave is calculated, and then compared with the reference branch line to obtain Frequency correction coefficient of each branch line;

The Hilbert-Huang HHT transform marginal spectrum is based on the decomposition basis intrinsic mode function IMF, and multiple decomposition basis intrinsic mode functions IMF are obtained through empirical mode decomposition EMD. The final result of the transformation is as follows:

$\mathrm{the\; s}\left(t\right)=\underset{k=1}{\overset{\mathrm{no}}{\mathrm{\&Sigma;}}}{C}_k+r$ ⑧;

In the above formula: s(t) is the original signal, r is the residual component, C _k is the decomposition basis intrinsic mode function IMF, and the Hilbert-Huang HHT transform marginal spectrum is defined as follows:

$h\left(w\right)={\&Integral;}_0^{T}h(w,t)\mathrm{dt}$ ⑨;

The Hilbert-Huang HHT transformation marginal spectrum represents the cumulative distribution on the frequency point, that is, the energy distribution. Taking the reference branch line as the standard, calculate the frequency difference of the frequency component of the refracted wave of each branch line, and convert it into a relative value, that is, the frequency Correction factor.

Further, in the step (4), the branch line length, wave velocity, frequency correction coefficient and path correction coefficient are substituted into the formula ⑥ to calculate the objective function;

In the genetic algorithm: the coding method adopts binary coding; the fitness function is the objective function; the genetic operation and control parameters are used for calculation; the optimal wave velocity is obtained through the formula ⑥, and the initial wave head reaches the bus at the M terminal of the substation by substituting into the formula ⑤. Finally, substituting the initial time of the M and N busbars of the substation into the formula ① to obtain the location of the fault point.

Compared with prior art, the beneficial effect that the present invention reaches is:

(1) The regional fault location algorithm is basically not affected by factors such as terminal faults, GPS, and communication interruptions. The overall reliability of the fault location system is significantly improved compared with existing devices.

(2) The accuracy of fault location is high. The algorithm of the invention fully considers the influence of the frequency and path of each branch line, and the accuracy of the fault location algorithm of the existing regional power grid is significantly improved. After testing the data of Liaoning Province from 2009 to 2011, the accuracy of the regional fault location algorithm basically reaches the accuracy of the existing double-ended traveling wave method.

Description of drawings

Figure 1 is a schematic diagram of double-terminal traveling wave ranging;

Fig. 2 is a schematic diagram of the regional fault location algorithm provided by the present invention;

Fig. 3 is the flow chart of the regional fault ranging algorithm based on optimal wave velocity provided by the present invention;

Fig. 4 is a network structure diagram of a specific embodiment provided by the present invention;

Fig. 5 is a graph of the iterative convergence curve of the genetic algorithm provided by the present invention.

Detailed ways

The specific implementation manners of the present invention will be further described in detail below in conjunction with the accompanying drawings.

The present invention provides a regional power grid fault location method based on optimal wave velocity. The method is based on genetic algorithm and signal spectrum analysis, and uses refracted waves of branch lines related to faulty lines in the power grid to perform fault location. The method includes the following steps:

(1) Grid structure analysis and data matching;

After the transmission line is faulty, the transient traveling wave will be refracted to each branch line through the busbar. In theory, each substation in the regional power grid can collect the refracted wave signal of the branch line. However, in practical engineering applications, considering the transient traveling wave signal For dispersion, the data of adjacent substations of fault lines are mainly selected.

The traveling wave distance measuring device of the transmission line adopts the double-terminal traveling wave method: the principle of the double-terminal traveling wave method is to use the first traveling wave head signal generated by the fault, and calculate the fault location by calculating the time difference between the fault initial traveling wave arriving at the two ends of the line, the expression as follows:

$l_1=\frac{L^{\&Prime;}{(t_2-t_1)}^v}{2}$ ①;

In the above formula: l ₁ is the fault distance; t ₁ and t ₂ are the time for the initial traveling wave to reach both ends of the line respectively, L'' is the total length of the line, and v is the propagation speed of the traveling wave;

The principle diagram of the regional fault location algorithm is shown in Figure 2. The regional power grid includes substations M, N and substations S ₁ , S ₂ ... S _n ; the substation M is connected to the substation N through a branch line L'; the substation S ₁ The substation S is connected to the substation M through the branch line 1; the substation S2 is connected to the substation M through the branch line ₂ ; the substation S _n is connected to the substation M through the branch line n; the branch line 1, the branch line 2 and the branch line n are connected in parallel.

(2) Construct the objective function and generate variable constraints;

When the faulty line is the MN section, after the transient traveling wave reaches the M terminal, it can be refracted to the branch lines 1~n through the M terminal bus. end ranging. Considering that there may be multiple substations at the S terminal (defined as S ₁ , S ₂ , and S _n terminals), the calculation formula for the double-terminal distance measurement composed of the data of the S terminal and the N terminal is as follows:

$\left\{\begin{array}{c}{d}_1=\frac{L+L_1-(t_1^{\&prime;}-t_2)*v_1}{2}\\ d_2=\frac{L+L_2-(t_2^{\&prime;}-t_2)*v_2}{2}\\ ......\\ d_{\mathrm{no}}=\frac{L+L_{\mathrm{no}}-(t_{\mathrm{no}}^{\&prime;}-t_2)*v_{\mathrm{no}}}{2}\end{array}\right.$ ②;

Among them, t′ ₁ , t' ₂ , and t' _n conform to the following formula:

t' _n ＝t ₁ +(L _n /v _n ) ③;

In formulas ② and ③, L, L ₁ , L ₂ , and L _n are the line lengths of the faulty line and branch lines 1, 2, and n respectively; t1, t2 are the moments when the initial traveling wave arrives at terminals M and N; t′ _{1 . _ _ _ _ _ _ _ _ _} v _n is the traveling wave velocity on branch lines 1, 2, and n; since d ₁ , d ₂ , and d _n are theoretically equal, the following conditions are satisfied:

$t_1^{\&prime;}-t_2^{\&prime;}-t_{\mathrm{no}}^{\&prime;}-\frac{L_1}{v_0\&times;k_{f1}\&times;k_{\mathrm{the\; s}1}}+\frac{L_2}{v_0\&times;k_{f2}\&times;k_{\mathrm{the\; s}2}}+\frac{L_{\mathrm{no}}}{v_0\&times;k_{\mathrm{fn}}\&times;k_{\mathrm{sn}}}=0$ ④;

$t_1=t_1^{\&prime;}-\frac{L_1}{v_0\&times;k_{f1}\&times;k_{\mathrm{the\; s}1}}$ ⑤;

In formula ④, v ₀ is the assumed reference wave velocity; k _f1 , k _f2 , k _fn are frequency correction coefficients, which are used to correct the difference in wave velocity caused by different frequency signals of refracted waves in each branch line; k _s1 , k _s2 , k _sn are respectively The path correction coefficient is used to correct the wave velocity difference caused by the different transmission paths of the refracted waves of each branch line; by solving the formula ④, and substituting the result into the formula ⑤ to get t ₁ is the final M-terminal initial wave head time, which can be combined with the N-terminal data Complete fault location. However, the frequency and path correction coefficients are affected by climate conditions and have unstable values, and there are often errors in actual engineering. Therefore, accurate calculation of wave velocity and correction coefficients is difficult to achieve in engineering.

In the method of the present invention, the optimal wave velocity is used to replace the traditional preset wave velocity, and the frequency and path correction coefficients are converted into relative values for calculation. Based on a single branch line, given the value range of the coefficients between each branch line, the objective function is constructed, and the optimal solution of the objective function (that is, the optimal wave velocity) is obtained by using the genetic algorithm for optimization calculation, as shown in the following formula:

$f(k_{f1},k_{f2},k_{\mathrm{fn}},k_{\mathrm{the\; s}1},k_{\mathrm{the\; s}2},k_{\mathrm{sn}},v_0)=t_1^{\&prime;}-t_2^{\&prime;}-t_{\mathrm{no}}^{\&prime;}-\frac{L_1}{v_0\&times;k_{f1}\&times;k_{\mathrm{the\; s}1}}+\frac{L_2}{v_0\&times;k_{f2}\&times;k_{\mathrm{the\; s}2}}+\frac{L_{\mathrm{no}}}{v_0\&times;k_{\mathrm{fn}}\&times;k_{\mathrm{sn}}}$ ⑥;

In order to ensure the rationality of the solution, the following variable constraints of formula ⑥ were also proposed in the study:

$\left\{\begin{array}{c}{v}_{\min }<v_0<v_{\max }\\ k_{f\min }<k_{\mathrm{fn}}<k_{f\max }\\ k_{\mathrm{the\; s}\min }<k_{\mathrm{sn}}<k_{\mathrm{the\; s}\max }\end{array}\right.$ ⑦;

In formula ⑦, v _max and v _min are the upper and lower limits of the wave velocity; k _fmax and k _fmin are the upper and lower limits of the frequency correction coefficient respectively; k _smax and k _smin are the upper and lower limits of the path correction coefficient respectively. In actual engineering, the path correction coefficient can be obtained through the analysis of the line structure, while the frequency correction coefficient needs to be obtained through the spectrum analysis of the signal, that is, the frequency correction coefficient based on the signal spectrum analysis and the correction coefficient determination method based on the line structure;

The method of determining the correction coefficient based on the line structure includes: according to the line parameters such as the distance between the ground wires of each branch line, the structure of the tower, and the length, and comparing with the reference branch line, the path correction coefficient of each branch line is obtained.

The specific method is as follows: Calculate the characteristic impedance, attenuation coefficient and phase shift coefficient of each branch line according to the line parameters, and the expressions are as follows:

$\mathrm{\&alpha;}=\sqrt{[R_m{G}_m-{\mathrm{\&omega;}}^2{L}_m{C}_m+\sqrt{(R_m^2+{\mathrm{\&omega;}}^2{L}_m^2)(G_m^2+{\mathrm{\&omega;}}^2{C}_m^2)}]/2}$ ⑩;

$\mathrm{\&beta;}=\sqrt{[{\mathrm{\&omega;}}^2{L}_m{C}_m-R_m{G}_m+\sqrt{(R_m^2+{\mathrm{\&omega;}}^2{L}_m^2)(G_m^2+{\mathrm{\&omega;}}^2{C}_m^2)}]/2}$

In the above formula, α is the attenuation coefficient, β is the phase shift coefficient; R _m , G _m , ω, and L _m respectively correspond to the mold resistance, inductance, conductance and capacitance of the line per unit length; when the phase shift coefficient of the line is obtained, the line The speed of wave propagation is as follows:

$v=\frac{\mathrm{\&omega;}}{\mathrm{\&beta;}}$

Taking the reference branch line as the standard, calculate the difference in transmission wave speed of the same frequency signal due to the different transmission paths of each branch line, and convert it into a relative value in the actual project, that is, the path correction coefficient.

(3) Analyze the signal spectrum of the refracted wave of the branch line to determine the frequency component of the signal and the range of the frequency correction coefficient;

Use the Hilbert-Huang HHT transformation marginal spectrum to analyze the frequency spectrum of the refracted wave of each branch line, calculate the frequency of the high frequency component of the refracted wave, and then compare it with the reference branch line to obtain the frequency correction coefficient of each branch line;

The Hilbert-Huang HHT transform marginal spectrum is based on the decomposition basis intrinsic mode function IMF, and multiple decomposition basis intrinsic mode functions IMF are obtained through empirical mode decomposition EMD. The final result of the transformation is as follows:

$\mathrm{the\; s}\left(t\right)=\underset{k=1}{\overset{\mathrm{no}}{\mathrm{\&Sigma;}}}{C}_k+r$ ⑧;

In the above formula: s(t) is the original signal, r is the residual component, C _k is the decomposition basis intrinsic mode function IMF, and the Hilbert-Huang HHT transform marginal spectrum is defined as follows:

$h\left(w\right)={\&Integral;}_0^{T}h(w,t)\mathrm{dt}$ ⑨;

It can be seen from Equation 9 that the Hilbert-Huang HHT transform marginal spectrum represents the cumulative distribution at the frequency points, that is, the energy distribution, so it is more suitable for analyzing the frequency components of traveling waves. The value range of the frequency correction coefficient can be obtained by analyzing the signal spectrum of the refracted wave of each branch line.

(4) Use the genetic algorithm to solve the optimal solution of the objective function to obtain the optimal wave velocity;

Substitute the branch line length, wave velocity, frequency correction coefficient and path correction coefficient into formula ⑥ to calculate the objective function;

In the genetic algorithm: the coding method adopts binary coding; the fitness function is the objective function; the genetic operation and control parameters are used for calculation; the optimal wave velocity is obtained through the formula ⑥, and the initial wave head reaches the bus at the M terminal of the substation by substituting into the formula ⑤. Finally, substituting the initial time of the M and N busbars of the substation into the formula ① to obtain the location of the fault point. The flowchart of regional fault location algorithm based on optimal wave velocity is shown in Fig. 3.

(5) Obtain the location of the fault point according to the optimal wave velocity, and output the result.

Example

Let’s take a set of actual fault records as an example: In August 2011, Qingchang Line B in Liaoning was faulty. Due to the damage of the distance measuring device on the side of Qinghe Power Plant on Qinghe Line B, the distance measurement results could not be given in time. The network structure diagram of the specific embodiment is shown in the figure As shown in 4, the calculation process of the method of the present invention is as follows:

(1) Select the adjacent stations of Qinghe Power Plant: Shendong, Niugang, Tieling, Hushitai, and Changtu stations to locate regional power grid faults.

(2) Construction of the objective function and parameter initialization, based on the initial wave head time of each branch line to pre-calculate the initial wave head time of the M terminal. Since the measurement error generally exhibits the principle of normal distribution, the average value should be close to the exact value in the case of multiple measurements. Therefore, the branch line closest to the average value is used as the reference line, and its wave velocity, frequency, and transmission path are used as reference reference values.

Table 1 Actual failure analysis data

Remarks: In the initial value calculation, it is assumed that the wave speed of each branch line is consistent, which is only used as a reference for subsequent analysis.

The initial wave head time at M terminal calculated based on each branch line (t ₁ , the time when the initial fault wave head arrives at Qinghe Power Plant) is shown in Table 2, and the average =100725.26us, considering that the average error (σ) of the distance measuring device is about 500 meters (3~4us), when the difference from the average value is more than 3σ, it is considered to be bad measurement data and will be eliminated in the calculation. After eliminating Qinghu Line B, the calculated average value of each branch line is 100716.16us, and the error between the actual M-terminal initial wave head time of 100706.49us is about 9.67us. In subsequent calculations, the branch line 1 closest to the average value is used as the reference value . Similarly, the wave velocity is calculated according to the average error ±σ (3~4us), and the value range of the reference wave velocity is 284.9~300m/us.

(3) Analyze the signal spectrum of the refracted wave of the branch line to determine the frequency component of the signal and the value range of the frequency correction coefficient.

Table 2 Frequency Correction Coefficient and Path Correction Coefficient

Remarks: Considering that the spacing between conductors and ground wires of the same voltage level is basically the same, the path correction factor mainly considers the influence of line structure and length.

(4) According to the formula 4, generate the objective function required by the genetic algorithm calculation, determine the parameters required by the genetic algorithm, solve the optimal approximate solution (that is, find the most reasonable wave velocity), and substitute the formula 5 and 1 to get the final fault point location .

See Table 2 for the values of the frequency and path correction coefficients in actual fault data analysis. Taking Qingchang Line B as the reference value, the value range of frequency and path correction coefficients in engineering applications is appropriately expanded in consideration of simplifying calculations and improving accuracy. Substituting parameters such as length, wave velocity, and correction coefficient into Equation 6, the objective function required for calculation can be obtained.

In this fault, after multiple iterations, the optimal solution of the equation (that is, the optimal wave velocity of Qingchang Line B) is 285.54m/us, corresponding to the M-side bus time of 100710.3us. Compared with the branch line 1 and the initial calculated average error, the error is reduced by 3us, and the difference from the actual fault time is 3.81us. After deducting the 0.5us additional error brought by the bus in the station, the error is 3.31us, and the corresponding fault location error of the transmission line is about 484m (assuming that the wave velocity is 293m/us), which is close to the accuracy of the existing double-terminal traveling wave ranging. The iterative convergence curve of the algorithm is shown in Figure 5.

Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the above embodiments, those of ordinary skill in the art should understand that: the present invention can still be Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention shall be covered by the scope of the claims of the present invention.

Claims (6)

1. A regional power grid fault location method based on optimal wave velocity, it is characterized in that, described method is based on genetic algorithm and signal spectrum analysis, utilizes the branch line refraction wave relevant with fault line in the power grid to carry out fault location, described method comprises Follow the steps below: (1) Grid structure analysis and data matching; (2) Construct the objective function and generate variable constraints; (3) Analyze the signal spectrum of the refracted wave of the branch line to determine the frequency component of the signal and the range of the frequency correction coefficient; (4) Use the genetic algorithm to solve the optimal solution of the objective function to obtain the optimal wave velocity; (5) Obtain the location of the fault point according to the optimal wave velocity, and output the result. 2. The regional power grid fault location method according to claim 1, characterized in that in the step (1), the transient traveling wave after the transmission line fault is refracted to each branch line through the busbar, and the substations in the regional power grid The refracted wave signals of the branch lines are all collected. In practice, considering the influence of the dispersion of the transient traveling wave signal, the data of the adjacent substation of the faulty line are selected; The traveling wave distance measuring device of the transmission line adopts the double-terminal traveling wave method: the principle of the double-terminal traveling wave method is to use the first traveling wave head signal generated by the fault, and calculate the fault location by calculating the time difference between the fault initial traveling wave arriving at the two ends of the line, the expression as follows: $l_1=\frac{L^{\&Prime;}{(t_2-t_1)}^v}{2}$ ①; In the above formula: l ₁ is the fault distance; t ₁ and t ₂ are the time for the initial traveling wave to reach both ends of the line respectively, L'' is the total length of the line, and v is the propagation speed of the traveling wave; Let the regional power grid include substations M, N and substations S ₁ , S ₂ ... S _n ; the substation M is connected to the substation N through the line L'; the substation S ₁ is connected to the substation M through the branch line 1; the substation S ₂ is connected to substation M through branch line 2; said substation S _n is connected to substation M through branch line n; branch line 1, branch line 2 and branch line n are connected in parallel. 3. The regional power grid fault location method according to claim 1, characterized in that, in the step (2), when the fault line is the MN segment, after the transient traveling wave reaches the M terminal of the substation, it passes through the M terminal of the substation The busbar is refracted to the branch lines 1~n, and the refracted wave of the branch line at the S end of the substation and the initial traveling wave at the N end of the substation are selected to form a double-end distance measurement; the S end of the substation includes the S ₁ , S ₂ and S _n ends of the substation; The calculation formulas for the double-terminal distance measurement composed of the substation S-terminal and substation N-terminal data are as follows: $\left\{\begin{array}{c}{d}_1=\frac{L+L_1-(t_1^{\&prime;}-t_2)*v_1}{2}\\ d_2=\frac{L+L_2-(t_2^{\&prime;}-t_2)*v_2}{2}\\ ......\\ d_{\mathrm{no}}=\frac{L+L_{\mathrm{no}}-(t_{\mathrm{no}}^{\&prime;}-t_2)*v_{\mathrm{no}}}{2}\end{array}\right.$ ②; Among them, t′ ₁ , t' ₂ , and t' _n conform to the following formula: t' _n ＝t ₁ +(L _n /v _n ) ③; In formulas ② and ③, L, L ₁ , L ₂ , and L _n are the line lengths of the faulty line and branch lines 1, 2, and n respectively; t1, t2 are the moments when the initial traveling wave arrives at terminals M and N; t′ _{1 . _ _ _ _ _ _ _ _ _} v _n is the traveling wave velocity on branch lines 1, 2, and n; since d ₁ , d ₂ , and d _n are theoretically equal, the following conditions are satisfied: $t_1^{\&prime;}-t_2^{\&prime;}-t_{\mathrm{no}}^{\&prime;}-\frac{L_1}{v_0\&times;k_{f1}\&times;k_{\mathrm{the\; s}1}}+\frac{L_2}{v_0\&times;k_{f2}\&times;k_{\mathrm{the\; s}2}}+\frac{L_{\mathrm{no}}}{v_0\&times;k_{\mathrm{fn}}\&times;k_{\mathrm{sn}}}=0$ ④; $t_1=t_1^{\&prime;}-\frac{L_1}{v_0\&times;k_{f1}\&times;k_{\mathrm{the\; s}1}}$ ⑤; In formula ④, v ₀ is the assumed reference wave velocity; k _f1 , k _f2 , k _fn are frequency correction coefficients, which are used to correct the difference in wave velocity caused by different frequency signals of refracted waves in each branch line; k _s1 , k _s2 , k _sn are respectively The path correction coefficient is used to correct the wave speed difference caused by the different transmission paths of the refracted waves of each branch line; Build the objective function as follows: $f(k_{f1},k_{f2},k_{\mathrm{fn}},k_{\mathrm{the\; s}1},k_{\mathrm{the\; s}2},k_{\mathrm{sn}},v_0)=t_1^{\&prime;}-t_2^{\&prime;}-t_{\mathrm{no}}^{\&prime;}-\frac{L_1}{v_0\&times;k_{f1}\&times;k_{\mathrm{the\; s}1}}+\frac{L_2}{v_0\&times;k_{f2}\&times;k_{\mathrm{the\; s}2}}+\frac{L_{\mathrm{no}}}{v_0\&times;k_{\mathrm{fn}}\&times;k_{\mathrm{sn}}}$ ⑥; The variable constraints of the objective function are as follows: $\left\{\begin{array}{c}{v}_{\min }<v_0<v_{\max }\\ k_{f\min }<k_{\mathrm{fn}}<k_{f\max }\\ k_{\mathrm{the\; s}\min }<k_{\mathrm{sn}}<k_{\mathrm{the\; s}\max }\end{array}\right.$ ⑦; In formula ⑦, v _max and v _min are the upper and lower limits of the wave velocity; k _fmax and k _fmin are the upper and lower limits of the frequency correction coefficient respectively; k _smax and k _smin are the upper and lower limits of the path correction coefficient respectively. 4. The regional power grid fault location method as claimed in claim 3, wherein the path correction coefficient is obtained by analyzing the line structure, and the frequency correction coefficient is obtained by analyzing the signal spectrum, that is, the frequency correction coefficient based on the signal spectrum analysis and Correction coefficient determination method based on line structure; The method of determining the correction coefficient based on the line structure includes: according to the line parameters such as the distance between the ground wires of each branch line, the structure of the tower, and the length, and comparing with the reference branch line, the path correction coefficient of each branch line is obtained; The details are as follows: Calculate the characteristic impedance, attenuation coefficient and phase shift coefficient of each branch line according to the line parameters, and the expressions are as follows: $\mathrm{\&alpha;}=\sqrt{[R_m{G}_m-{\mathrm{\&omega;}}^2{L}_m{C}_m+\sqrt{(R_m^2+{\mathrm{\&omega;}}^2{L}_m^2)(G_m^2+{\mathrm{\&omega;}}^2{C}_m^2)}]/2}$ ⑩; $\mathrm{\&beta;}=\sqrt{[{\mathrm{\&omega;}}^2{L}_m{C}_m-R_m{G}_m+\sqrt{(R_m^2+{\mathrm{\&omega;}}^2{L}_m^2)(G_m^2+{\mathrm{\&omega;}}^2{C}_m^2)}]/2}$ In the above formula, α is the attenuation coefficient, β is the phase shift coefficient; R _m , G _m , ω, and L _m respectively correspond to the mold resistance, inductance, conductance and capacitance of the line per unit length; when the phase shift coefficient of the line is obtained, the line The speed of wave propagation is as follows: $v=\frac{\mathrm{\&omega;}}{\mathrm{\&beta;}}$ Taking the reference branch line as the standard, calculate the difference in transmission wave speed of the same frequency signal due to the different transmission paths of each branch line, and convert it into a relative value in the actual project, that is, the path correction coefficient. 5. The regional power grid fault location method according to claim 1, characterized in that, in the step (3), the spectral analysis of the refracted waves of each branch line is carried out by using the Hilbert-Huang HHT transform marginal spectrum, and the calculated The frequency of the high-frequency component of the refracted wave is compared with the reference branch line to obtain the frequency correction coefficient of each branch line; The Hilbert-Huang HHT transform marginal spectrum is based on the decomposition basis intrinsic mode function IMF, and multiple decomposition basis intrinsic mode functions IMF are obtained through empirical mode decomposition EMD. The final result of the transformation is as follows: $\mathrm{the\; s}\left(t\right)=\underset{k=1}{\overset{\mathrm{no}}{\mathrm{\&Sigma;}}}{C}_k+r$ ⑧; In the above formula: s(t) is the original signal, r is the residual component, C _k is the decomposition basis intrinsic mode function IMF, and the Hilbert-Huang HHT transform marginal spectrum is defined as follows: $h\left(w\right)={\&Integral;}_0^{T}h(w,t)\mathrm{dt}$ ⑨; The Hilbert-Huang HHT transformation marginal spectrum represents the cumulative distribution on the frequency point, that is, the energy distribution. Taking the reference branch line as the standard, calculate the frequency difference of the frequency component of the refracted wave of each branch line, and convert it into a relative value, that is, the frequency Correction factor. 6. The regional power grid fault location method according to claim 1, characterized in that in the step (4), the branch line length, wave velocity, frequency correction coefficient and path correction coefficient are substituted into the formula ⑥ to calculate the objective function; In the genetic algorithm: the coding method adopts binary coding; the fitness function is the objective function; the genetic operation and control parameters are used for calculation; the optimal wave velocity is obtained through the formula ⑥, and the initial wave head reaches the bus at the M terminal of the substation by substituting into the formula ⑤. Finally, substituting the initial time of the M and N busbars of the substation into the formula ① to obtain the location of the fault point.