CN112255493A - Multi-criterion comprehensive voting power distribution network fault line selection method - Google Patents

Abstract

The invention discloses a multi-criterion comprehensive voting power distribution network fault line selection method which specifically comprises the following steps: step 1: preprocessing zero-sequence current data: step 2: performing complementary set empirical mode decomposition; and step 3: construction of fault line selection criterion 1-IMF₁Component feature instantaneous phase criterion; and 4, step 4: construction of fault line selection criterion 2-IMF₁Component instantaneous energy relative entropy criterion; and 5: constructing a fault line selection criterion 3-a characteristic zero-sequence current value polarity criterion; step 6: according to the three fault line selection criteria in the steps 3, 4 and 5, the three fault line selection criteria are simultaneously calculated and judged, and the minority is adopted to carry out comprehensive voting according to the majority principle; if the judgment results of the 3 route selection criteria are inconsistent, the calculation is returned again, and then the voting is carried out. The invention is not influenced by initial phase angle, grounding resistance value, compensation degree, noise intensity and the like, and the accuracy of fault line selection is higher.

Description

Multi-criterion comprehensive voting power distribution network fault line selection method

Technical Field

The invention belongs to the technical field of relay protection of a power distribution network of a power system, and particularly relates to a multi-criterion comprehensive voting power distribution network fault line selection method.

Background

Most of Chinese 6 kV-66 kV power distribution networks are in a small current grounding mode, so the power distribution network systems are also called small current grounding systems, wherein 66kV and 35kV power grids are mainly grounded in a mode that neutral points are grounded through arc suppression coils; the 6 kV-10 kV power grid is partially grounded through a neutral point ungrounded mode, and partially grounded through an arc suppression coil. The power distribution network is densely distributed in urban, rural and mountain areas, is located outdoors in the whole year, is influenced by severe environments such as wind, rain, hail, thunder and lightning and increasingly severe environmental pollution, and has high probability of failure due to unpredictable human factors, especially an overhead line. Statistical data show that faults of the power grid mostly occur in the power distribution network, and 80% of faults of the power distribution network are single-phase earth faults. After the power distribution network is subjected to single-phase permanent grounding, the problems of weak fault signals, complex working condition and the like cause difficulty in line selection, so that great research development in the field has important and profound significance.

When single-phase earth faults occur, the non-fault phase-to-ground voltage rises, if intermittent arc grounding occurs, arc overvoltage can be caused, system insulation is threatened, the system insulation is easily expanded to be an interphase short circuit, a fault circuit is found as soon as possible, and faults are eliminated as soon as possible. Even if the constant resistance is grounded, the power frequency overvoltage can also generate loss on the equipment, and the loss can destroy the insulation performance of the equipment when accumulated to a certain degree. Although China's regulations for power distribution networks stipulate that single-phase grounding can be operated for 1-2 hours with faults, most power supply departments require that the fault line be cut off as soon as possible. At present, the problem of single-phase earth fault protection of a distribution network is still a technical problem in the operation of power enterprises, and the problem is not solved satisfactorily for a long time.

The research of fault line selection is mainly the identification and judgment of a fault line when a single-phase earth fault occurs in a low-current earth distribution network, at the moment, the fault current is weak, and particularly in an arc suppression coil earthing mode, and a satisfactory result is difficult to obtain only by using a conventional method of information such as the magnitude of the current amplitude and the like. In recent years, many scholars at home and abroad begin to apply modern signal processing technology to fault line selection of a low-current grounding system, and a large number of fault line selection methods emerge, and the methods mainly adopt wavelet transformation, S transformation, mathematical morphology, Hilbert-Huang transform (HHT), Prony algorithm and the like in summary.

Although a large number of line selection methods have been proposed and applied to the field, the practical effect is not ideal, and for the reason, the existing problems are not obvious in the aspect of fault. When the small-current grounding system is grounded in a single phase, the fault steady-state current is weak, and although the amplitude of a fault transient signal is larger than that of a steady-state signal, the fault transient signal has short duration. The second is the effect of an unstable fault arc. In single phase earth faults in the field, there is no stable earth current (including injected current) signature for arc grounding, particularly intermittent arc grounding. There are also random factors. The operation modes of the Chinese power distribution network are various, and the outgoing line length and the outgoing line number of the transformer substation are frequently changed; therefore, due to the influence of the factors, the accuracy of the fault line selection of the current line selection method needs to be further improved.

Disclosure of Invention

The invention aims to provide a multi-criterion comprehensive voting power distribution network fault line selection method, which solves the problem that the accuracy of fault line selection in the conventional line selection method needs to be further improved.

The technical scheme adopted by the invention is that,

a multi-criterion comprehensive voting power distribution network fault line selection method specifically comprises the following steps:

step 1: preprocessing zero-sequence current data: the specific expression is as follows formula (9):

i_0m(t)＝i_0m[(t_f-T/2):(t_f+T/2)]-i_0m[(t_f-3T/2):(t_f-T/2)] (9)，

wherein i_0m(t) is zero sequence current after data preprocessing, m is a feeder line number, and t is 1,2, …, k; t is t_fIs the time of failure; t is a power frequency period; i.e. i_0m[(t_f-T/2):(t_f+T/2)]Is i_0mFrom the moment of failure t_fBefore T/2 to fault time T_fValues between the last T/2; i.e. i_0m[(t_f-3T/2):(t_f-T/2)]Is i_0mFrom the moment of failure t_fFirst 3T/2 to fault time T_fValues between the first T/2;

step 2: performing complementary set empirical mode decomposition;

and step 3: construction of fault line selection criterion 1-IMF₁Component feature instantaneous phase criterion;

and 4, step 4: construction of fault line selection criterion 2-IMF₁Component instantaneous energy relative entropy criterion;

and 5: constructing a fault line selection criterion 3-a characteristic zero-sequence current value polarity criterion;

step 6: according to the three fault line selection criteria in the steps 3, 4 and 5, when a single-phase earth fault occurs in the power distribution network, the three fault line selection criteria are calculated and judged at the same time, the principle that minority obeys majority is adopted for comprehensive voting, and when 2 or more criteria judge that a branch feeder line has a fault, the branch feeder line is voted for the fault; when the bus is judged to have a fault by 2 or more criteria, the voting bus has a fault; if the judgment results of the 3 route selection criteria are inconsistent, the calculation is returned again, and then the voting is carried out.

The present invention is also characterized in that,

in step 2, the complementary set empirical mode decomposition specifically comprises:

step 2.1: to i_0m(k) Performing CEEMDAN decomposition to obtain each eigenmode component, and recording the obtained eigenmode component as IMF, wherein the IMF component is a time-amplitude matrix IMF (i)_0m) The rows of the matrix correspond to the amplitudes and the columns correspond to the sampling points, the 1 st IMF component is adopted and is recorded as IMF₁To IMF₁The components are subjected to Hilbert transform to obtain Hi_0mThe specific expression is as follows:

step 2.2: to Hi_0mCalculating a modulus and a phase for each element in theAngle, the formula (10) and (11) below:

die:

phase angle: theta [ Hi ]_0m(l,k)]＝arctan[y_m(l,k)/x_m(l,k)] (11)，

Wherein, Hi_0m(l,k)＝x_m(l,k)+jy_m(l, k), j is an imaginary unit; l is the number of the feeder line, and l is 1,2, …, m; k is 1,2, …, n; a (-) modulo operation; theta (-) is a phase angle calculation, and the calculated mode matrix and phase angle matrix respectively represent IMF₁The instantaneous amplitude and instantaneous phase of the component, denoted as a_{_Hi0m}And theta_{_Hi0m}；

Step 2.3: for IMF₁Instantaneous amplitude a of the component_{_Hi0m}The phase angles corresponding to the instantaneous amplitude values of the top 3 are called characteristic instantaneous phase positions in a descending order, and the instantaneous amplitude values are sequentially marked as a_{1_Hi0m}、 a_{2_Hi0m}、a_{3_Hi0m}The characteristic instantaneous phase is sequentially marked as theta_{1_Hi0m}、θ_{2_Hi0m}、θ_{3_Hi0m}。

In step 3, a fault route selection criterion 1-IMF is established₁The component characteristic instantaneous phase criterion specifically comprises the following steps:

step 3.1: using instantaneous amplitude a_{1_Hi0m}、a_{2_Hi0m}、a_{3_Hi0m}Characterizing instantaneous phase θ_{1_Hi0m}、θ_{2_Hi0m}、θ_{3_Hi0m}Accuracy, a modified confidence calculation is constructed, as shown in equations (12), (13) and (14):

wherein λ is_{1_Hi0m}，λ_{2_Hi0m}，λ_{3_Hi0m}The larger the value, the greater the confidence;

step 3.2: in order to construct a line selection criterion 1 with quantitative characterization confidence, 3 maximum instantaneous amplitude values a of a feeder line m are measured_{1_Hi0m}、a_{2_Hi0m}、a_{3_Hi0m}Corresponding 3 characteristic instantaneous phases theta_{1_Hi0m}、θ_{2_Hi0m}、θ_{3_Hi0m}Correcting to obtain corrected characteristic instantaneous phase

The specific correction formula is as follows (15), (16) and (17):

step 3.3: by IMF₁Taking the corrected characteristic instantaneous phase corresponding to the maximum value of the instantaneous amplitude of the component and the single-phase earth fault of the feeder 1 as examples, the following formula (18) is satisfied:

i.e. lambda_{1_Hi01}＝λ_{1_Hi02}＝…λ_{1_Hi 0m}Before correction, the characteristic instantaneous phase theta_{1_Hi01}，θ_{1_Hi02}，θ_{1_Hi03}，……，θ_{1_Hi0m}Satisfies the following inequality (19):

and constructing a corrected fault line selection criterion of the branch feeder line 1, wherein the following inequality (20) is as follows:

when the inequality is satisfied, judging that the feeder line 1 has a single-phase earth fault; the line selection criterion of whether the other branch feeders have single-phase earth faults is similar to the criterion of the branch feeder 1;

step 3.4: when the bus is in fault and before correction, the characteristic instantaneous phase theta_{1_Hi01}，θ_{1_Hi02}，θ_{1_Hi03}，…，θ_{1_Hi0m}Satisfies inequality (21):

and constructing a corrected bus fault line selection criterion, wherein the following inequality (22) is as follows:

and when the inequality is satisfied, judging that the bus has single-phase earth fault.

In step 4, a fault route selection criterion 2-IMF is established₁The criterion of the relative entropy of the component instantaneous energy is specifically as follows:

step 4.1: IMF for each feeder₁Maximum 3 instantaneous amplitudes a of the components_{1_Hi0m}、a_{2_Hi0m}、 a_{3_Hi0m}Calculating the energy values respectively according to the following formula (27):

E_{1_Hi0m}＝|a_{1_Hi0m}|²，E_{2_Hi0m}＝|a_{2_Hi0m}|²，E_{3_Hi0m}＝|a_{3_Hi0m}|² (27)，

further, 3 energy sums were constructed, respectively,

next, the instantaneous energy ratio is calculated as follows (28):

step 4.2: calculating the instantaneous energy relative entropy of each feeder relative to the other feeders, as in equation (29):

step 4.3: calculating the comprehensive relative energy entropy value of each feeder line, wherein IMF₁The comprehensive relative energy entropy corresponding to the maximum value of the instantaneous amplitude of the component is as follows:

the comprehensive relative energy entropy corresponding to the second largest value is:

the relative energy entropy corresponding to the 3 rd maximum value is:

step 4.4: and correcting the comprehensive relative energy entropy of each feeder line by adopting the following calculation formula, wherein the correction calculation formula of the maximum instantaneous amplitude, the second maximum value and the 3 rd maximum value is as the following formula (30):

step 4.5: calculating the average value of the corrected comprehensive relative energy entropy, wherein the calculation formula is as the following formula (31):

i.e. one for each feeder line

Therefore, a fault route selection criterion 2 is constructed, and specifically comprises the following steps: the largest 3 are selected

In the order of magnitude are

When it is satisfied with

When it is determined that the maximum value is

And the corresponding feeder line is a fault feeder line, otherwise, the fault feeder line is judged to be a bus fault.

In step 5, a fault line selection criterion 3-a characteristic zero sequence current value polarity criterion is constructed, which specifically comprises the following steps:

step 5.1: according to the selected 3 characteristic instantaneous amplitude values a_{1_Hi0m}、a_{2_Hi0m}、a_{3_Hi0m}Finding out 3 zero-sequence current amplitude points corresponding to the respective zero-sequence currents of each feeder line, sequentially defining the amplitude points as a maximum characteristic zero-sequence current value, a second maximum characteristic zero-sequence current value and a 3 rd maximum characteristic zero-sequence current value, and respectively representing the values as follows: d_{1_l}，d_{2_l}，d_{3_l}(ii) a And (3) constructing a polarity calculation formula of 3 characteristic zero sequence current values of each feeder line, wherein the formula (32) is as follows:

wherein, when k is 1,2, 3, d_{1_l}，d_{2_l}，d_{3_l}Respectively representing zero sequence current waves of the feeder line lMaximum, next largest, 3 rd largest, r_ljDenotes d_{k_l}And d_{k_j}Polarity of r between_ljIs +1, represents d_{k_l}And d_{k_j}The phases are the same; r is_ljIs-1, represents d_{k_l}And d_{k_j}The phases are opposite, and therefore, a fault line selection matrix R is obtained, as shown in the following formula (33):

wherein, diagonal elements in the matrix R are all 1, and each non-diagonal element is a comparison result of the polarities of each two of the zero sequence current amplitudes of the feeder line characteristics;

step 5.2: constructing a fault line selection criterion 3, which specifically comprises the following steps: comparing the number of the elements of each row in the matrix R, which are '-1', to judge the fault feeder; if the number of '-1' in the l-th line is m-1, the feeder line l is judged to be a fault feeder line, and meanwhile, other feeder lines are judged to be sound feeder lines; and if each row element in the matrix R is '1', judging that the bus is a fault feeder line, and simultaneously judging that other feeder lines are sound feeder lines.

The method has the beneficial effects that the multi-criterion comprehensive voting power distribution network fault line selection method is based on complementary set empirical mode decomposition and Hilbert transformation theory, and provides a multi-criterion comprehensive voting fault line selection method based on 1 st intrinsic mode component characteristic instantaneous phase criterion, instantaneous energy relative entropy criterion and characteristic zero sequence current value polarity criterion. A large number of simulations and field experiments show that the method is not influenced by a primary phase angle, a grounding resistance value, a compensation degree, noise intensity and the like to a certain extent, and the accuracy of fault line selection is high.

Drawings

Fig. 1 is a 10kV radial distribution network simulation model in an embodiment of the multi-criterion comprehensive voting distribution network fault line selection method of the invention.

Detailed Description

The method for selecting the fault line of the multi-criterion comprehensive voting power distribution network is described in detail below with reference to the accompanying drawings and the detailed description.

As shown in fig. 1, a multi-criterion comprehensive voting power distribution network fault line selection method specifically includes the following steps:

step 1: preprocessing zero-sequence current data: the specific expression is as follows formula (9):

i_0m(t)＝i_0m[(t_f-T/2):(t_f+T/2)]-i_0m[(t_f-3T/2):(t_f-T/2)] (9)，

wherein i_0m(t) is zero sequence current after data preprocessing, m is a feeder line number, and t is 1,2, …, k; t is t_fIs the time of failure; t is a power frequency period; i.e. i_0m[(t_f-T/2):(t_f+T/2)]Is i_0mFrom the moment of failure t_fBefore T/2 to fault time T_fValues between the last T/2; i.e. i_0m[(t_f-3T/2):(t_f-T/2)]Is i_0mFrom the moment of failure t_fFirst 3T/2 to fault time T_fValues between the first T/2;

step 2: complementary Ensemble Empirical Mode Decomposition with Adaptive Noise, CEEMDAN) was performed:

step 2.1: first, for i_0m(k) Performing CEEMDAN decomposition to obtain each Intrinsic Mode component (IMF), wherein the IMF component is a time-amplitude matrix IMF (i)_0m) The rows of the matrix correspond to the amplitude values and the columns correspond to the sampling points. Considering the compensation function of the arc suppression coil, in order to avoid the influence of the arc suppression coil on the fault line selection method in the dynamic adjustment process, the invention adopts the 1 st IMF component (IMF)₁) Further to IMF₁The components are subjected to Hilbert transform to obtain Hi_0mThe concrete expression is

Step 2.2: to Hi_0mThe modulus and phase angle are calculated for each element, the formula (10) and (11) are calculated as follows:

die:

phase angle: theta [ Hi ]_0m(l,k)]＝arctan[y_m(l,k)/x_m(l,k)] (11)，

Wherein, Hi_0m(l,k)＝x_m(l,k)+jy_m(l, k), j is an imaginary unit; l is the number of the feeder line, and l is 1,2, …, m; k is 1,2, …, n; a (-) modulo operation; theta (-) is a phase angle calculation, and the calculated mode matrix and phase angle matrix respectively represent IMF₁The instantaneous amplitude and instantaneous phase of the component, denoted as a_{_Hi0m}And theta_{_Hi0m}；

Step 2.3: for IMF₁Instantaneous amplitude a of the component_{_Hi0m}The phase angles corresponding to the instantaneous amplitude values of the top 3 are called characteristic instantaneous phase positions in a descending order, and the instantaneous amplitude values are sequentially marked as a_{1_Hi0m}、 a_{2_Hi0m}、a_{3_Hi0m}The characteristic instantaneous phase is sequentially marked as theta_{1_Hi0m}、θ_{2_Hi0m}、θ_{3_Hi0m}。

And step 3: construction of fault line selection criterion 1-IMF₁Component feature instantaneous phase criterion: the method comprises the following specific steps:

step 3.1: using instantaneous amplitude a_{1_Hi0m}、a_{2_Hi0m}、a_{3_Hi0m}Characterizing instantaneous phase θ_{1_Hi0m}、θ_{2_Hi0m}、θ_{3_Hi0m}Accuracy, a modified confidence calculation is constructed, as shown in equations (12), (13) and (14):

wherein λ is_{1_Hi0m}，λ_{2_Hi0m}，λ_{3_Hi0m}The larger the value, the greater the confidence;

step 3.2: in order to construct a line selection criterion 1 with quantitative characterization confidence, 3 maximum instantaneous amplitude values a of a feeder line m are measured_{1_Hi0m}、a_{2_Hi0m}、a_{3_Hi0m}Corresponding 3 characteristic instantaneous phases theta_{1_Hi0m}、θ_{2_Hi0m}、θ_{3_Hi0m}Correcting to obtain corrected characteristic instantaneous phase

The specific correction formula is as follows (15), (16) and (17):

step 3.3: next, for convenience of explanation, IMF is used₁Taking the corrected characteristic instantaneous phase corresponding to the maximum value of the instantaneous amplitude of the component and the single-phase earth fault of the feeder 1 as examples, the following formula (18) is satisfied:

i.e. lambda_{1_Hi01}＝λ_{1_Hi02}＝…λ_{1_Hi0m}Before correction, the characteristic instantaneous phase theta_{1_Hi01}，θ_{1_Hi02}，θ_{1_Hi03}，……，θ_{1_Hi0m}Satisfies the following inequality (19) (for the reason: the faulty feeder 1 and other healthy feedersThe characteristic instantaneous phase inversion of the line, taking more than 90 degrees as the inversion in engineering application):

therefore, a corrected fault route selection criterion of the branch feeder 1 is constructed, and the following inequality (20) is:

when the inequality is satisfied, judging that the feeder line 1 has a single-phase earth fault; the line selection criterion of whether the other branch feeders have single-phase earth faults is similar to the criterion of the branch feeder 1;

step 3.4: finally, in the event of a bus fault, before correction, the characteristic instantaneous phase θ_{1_Hi01}，θ_{1_Hi02}，θ_{1_Hi03}，…，θ_{1_Hi0m}Inequality (21) is satisfied (reason: the characteristic instantaneous phase of each branch feeder is in phase, and 90 degrees or less is taken as in phase in engineering application):

and constructing a corrected bus fault line selection criterion, wherein the following inequality (22) is as follows:

and when the inequality is satisfied, judging that the bus has single-phase earth fault.

It should be noted that: the above is IMF₁The fault line selection criterion constructed by taking the maximum value of the instantaneous amplitude of the component as an example is also respectively based on IMF₁Second largest value of component instantaneous amplitude, IMF₁The 3 rd maximum value of the component instantaneous amplitude is taken as an example for constructing a fault line selection criterion, and the specific construction process is similar to that of the maximum value.

Step 4, constructing fault line selection criterion 2-IMF₁The method comprises the following specific steps of:

step 4.1: IMF for each feeder₁Maximum 3 instantaneous amplitudes a of the components_{1_Hi0m}、a_{2_Hi0m}、 a_{3_Hi0m}Calculating the energy values respectively according to the following formula (27):

E_{1_Hi0m}＝|a_{1_Hi0m}|²，E_{2_Hi0m}＝|a_{2_Hi0m}|²，E_{3_Hi0m}＝|a_{3_Hi0m}|² (27)，

further, 3 energy sums were constructed, respectively,

next, the instantaneous energy ratio is calculated as follows (28):

step 4.2: calculating the instantaneous energy relative entropy of each feeder relative to the other feeders, as in equation (29):

step 4.3: calculating the comprehensive relative energy entropy value of each feeder line, wherein IMF₁The comprehensive relative energy entropy corresponding to the maximum value of the instantaneous amplitude of the component is as follows:

the comprehensive relative energy entropy corresponding to the second largest value is:

the relative energy entropy corresponding to the 3 rd maximum value is:

step 4.4: and correcting the comprehensive relative energy entropy of each feeder line by adopting the following calculation formula, wherein the correction calculation formula of the maximum instantaneous amplitude, the second maximum value and the 3 rd maximum value is as the following formula (30):

step 4.5: calculating the average value of the corrected comprehensive relative energy entropy, wherein the calculation formula is as the following formula (31):

i.e. one for each feeder line

Therefore, a fault route selection criterion 2 is constructed, and specifically comprises the following steps: the largest 3 are selected

In the order of magnitude are

When it is satisfied with

When it is determined that the maximum value is

And the corresponding feeder line is a fault feeder line, otherwise, the fault feeder line is judged to be a bus fault.

Step 5, establishing a fault line selection criterion 3-a characteristic zero sequence current value polarity criterion, which comprises the following specific steps:

step 5.1: according to the selected 3 characteristic instantaneous amplitude values a_{1_Hi0m}、a_{2_Hi0m}、a_{3_Hi0m}Finding out 3 zero-sequence current amplitude points corresponding to the respective zero-sequence currents of each feeder line, sequentially defining the amplitude points as a maximum characteristic zero-sequence current value, a second maximum characteristic zero-sequence current value and a 3 rd maximum characteristic zero-sequence current value, and respectively representing the values as follows: d_{1_l}，d_{2_l}，d_{3_l}(ii) a And (3) constructing a polarity calculation formula of 3 characteristic zero sequence current values of each feeder line, wherein the formula (32) is as follows:

wherein, when k is 1,2, 3, d_{1_l}，d_{2_l}，d_{3_l}Respectively representing the maximum value, the second maximum value, the 3 rd maximum value and r in the zero sequence current waveform of the feed line l_ljDenotes d_{k_l}And d_{k_j}Polarity of r between_ljIs +1, represents d_{k_l}And d_{k_j}The phases are the same; r is_ljIs-1, represents d_{k_l}And d_{k_j}The phases are opposite, and therefore, a fault line selection matrix R is obtained, as shown in the following formula (33):

wherein, diagonal elements in the matrix R are all 1, and each non-diagonal element is a comparison result of the polarities of each two of the zero sequence current amplitudes of the feeder line characteristics;

step 5.2: constructing a fault line selection criterion 3, which specifically comprises the following steps: comparing the number of the elements of each row in the matrix R, which are '-1', to judge the fault feeder; if the number of '-1' in the l-th line is m-1, the feeder line l is judged to be a fault feeder line, and meanwhile, other feeder lines are judged to be sound feeder lines; and if each row element in the matrix R is '1', judging that the bus is a fault feeder line, and simultaneously judging that other feeder lines are sound feeder lines.

Step 6: and (3) comprehensive voting: the invention provides a multi-criterion comprehensive voting method by integrating the above 3 criteria, wherein the adopted principle is 'minority obeying majority', namely, when a single-phase earth fault occurs in a power distribution network, 3 fault line selection criteria are calculated and judged simultaneously, and further, when 2 or more criteria judge that a certain branch feeder line has a fault, the fault of the branch feeder line is voted; similarly, when 2 or more criteria determine that the bus fails, the voting bus fails; otherwise, if the judgment results of the 3 route selection criteria are inconsistent, the recalculation is returned, and then the voting is carried out.

The working principle of the multi-criterion comprehensive voting power distribution network fault line selection method is as follows:

firstly, complementary set empirical mode decomposition theory:

the complementary set Empirical Mode Decomposition (CEEMDAN) algorithm can not only reduce the calculation amount of the traditional Empirical Mode Decomposition (EMD) algorithm, but also effectively reduce the modal aliasing phenomenon. Here, E is defined_j(. The) is the operation factor of the jth Intrinsic Mode Function (IMF) obtained by adopting the EMD algorithm, and the specific CEEMDAN algorithm calculation steps are as follows:

step 1: definition of ε₀Is the amplitude, omega, of the added white Gaussian noiseⁱ(t) Gaussian white noise with unit variance, and decomposing the signal x (t) + epsilon by adopting EMD algorithm₀ωⁱ(t), further obtaining the 1 st IMF component as follows:

wherein the content of the first and second substances,

is omegaⁱ(t) the corresponding 1 st modal component.

Step 2: calculating a decomposition remainder r₁(t), as follows:

r₁(t)＝x(t)-c₁(t) (2)，

and step 3: decomposition of r₁(t)+ε₁E₁[ωⁱ(t)]Obtaining the 1 st order IMF component, and further defining the 2 nd order IMF component as:

and 4, step 4: when K is 2, …, K, the kth order remainder is calculated. Define the (k + 1) th order mode as:

and 5: repeat step 4 until the resulting remainder can no longer be decomposed (the remainder contains no more than 2 poles), which is expressed as:

thus, the signal f (t) can be expressed as:

II, Hilbert transformation theory;

for a time signal x (t), its Hilbert transform y (t) can be derived:

wherein, P is Cauchy main value, and P is 1 in use. According to this definition, x (t) and y (t) can constitute an analytic signal z (t):

Z(t)＝x(t)+jy(t)＝a(t)e^jθ(t) (8)，

wherein a (t) represents the amplitude of the signal,

theta (t) is the instantaneous phase,

thirdly, a multi-criterion comprehensive voting fault line selection method based on Hilbert transformation;

starting from the aspect of characteristic IMF component phase angle information, overall waveform change trend information and zero-sequence current characteristic point polarity information 3, the comprehensive fault route selection criterion combining phase discrimination, relative entropy calculation and characteristic point polarity discrimination is respectively constructed, and the comprehensive fault route selection criterion is as follows:

1. preprocessing data;

on the basis of the analysis, the reliability of the phase angle information can be described by using the amplitude information when the Hilbert is used for extracting the phase angle information to select lines; in addition, if the phase angle information of a plurality of sampling points can be fully utilized, the influence of errors of individual sampling points on the algorithm can be avoided, and the reliability of the line selection method is improved.

Preprocessing original zero-sequence current data: subtracting the period before the fault from the period after the fault, and expressing the following expression:

i_0m(t)＝i_0m[(t_f-T/2):(t_f+T/2)]-i_0m[(t_f-3T/2):(t_f-T/2)] (9)

in formula (9): m is a feeder line number; t is t_fIs the time to failure; t is a power frequency period; i.e. i_0m[(t_f-T/2):(t_f+T/2)]Is i_0mValues in the first and second half power frequency periods before and after the fault time.

To i_0m(k) Performing CEEMDAN decomposition to obtain IMF component, which is time-amplitude matrix IMF (i)_0m) The rows of the matrix correspond to the amplitude values and the columns correspond to the sampling points. Considering the compensation function of the arc suppression coil, in order to avoid the influence of the arc suppression coil on the fault line selection method in the dynamic adjustment process, the invention adopts IMF₁Component, to IMF₁The components are Hilbert-transform, and the specific expression is as follows:

to Hi_0mCalculating the modulus and phase angle for each element in the equation (10), (11):

θ[Hi_0m(l,k)]＝arctan[y_m(l,k)/x_m(l,k)] (11)，

in the above formula, Hi_0m(l,k)＝x_m(l,k)+jy_m(l, k), j is an imaginary unit; 1,2, …, m; k is 1,2, …, n; a (-) modulo operation; theta (-) is a phase angle calculation, and the calculated mode matrix and phase angle matrix respectively represent IMF₁The instantaneous amplitude and instantaneous phase of the component, denoted as a_{_Hi0m}And theta_{_Hi0m}。

For IMF₁Instantaneous amplitude a of the component_{_Hi0m}The phase angles corresponding to the instantaneous amplitude values of the top 3 are called characteristic instantaneous phase positions in a descending order, and the instantaneous amplitude values are sequentially marked as a_{1_Hi0m}、a_{2_Hi0m}And a_{3_Hi0m}The characteristic instantaneous phase is sequentially marked as theta_{1_Hi0m}、θ_{2_Hi0m}And theta_{3_Hi0m}。

2. Criterion 1: IMF1 component feature instantaneous phase criterion;

in the field of fault line selection research, when a branch feeder line has a single-phase earth fault, the polarities of zero-sequence currents of a fault feeder line and a non-fault feeder line are opposite in the initial first half wave; when the bus has a single-phase earth fault, the initial first half-wave polarities of the zero-sequence currents of all the branch feeder lines are the same. Therefore, the method can be further expanded, and when the branch feeder line is in fault, the characteristic instantaneous phase of the fault feeder line is opposite to the characteristic instantaneous phase of the non-fault feeder line; when a bus fails, the characteristic instantaneous phases of all branch feeders are in phase. However, according to the rule, a line selection criterion constructed based on the characteristic often has a misjudgment phenomenon along with a high-resistance grounding fault at the tail end of a feeder line, a fault under strong noise interference or a fault at a voltage zero crossing point, and based on the misjudgment, the characteristic instantaneous phase criterion with the characteristic confidence is constructed, and the specific analysis is as follows:

from the foregoing analysis, the magnitude of the instantaneous amplitude can reflect the accuracy of the phase angle, and therefore, the correction confidence calculation formula is defined, as shown in formulas (12) to (14):

in the above formula, λ_{1_Hi0m}，λ_{2_Hi0m}，λ_{3_Hi0m}The larger the value, the greater the confidence.

In order to construct a line selection criterion with quantitative characterization confidence, 3 characteristic instantaneous phases theta corresponding to 3 maximum instantaneous amplitudes of a feeder line m are subjected to_{1_Hi0m}，θ_{2_Hi0m}，θ_{3_Hi0m}Correcting to obtain corrected characteristic instantaneous phase

And

the specific modified formula is as shown in formulas (15) to (17):

for analysing the law of existence of difference between corrected characteristic instantaneous phases of each feeder line to

Analysis was performed for the examples, others

Similarly, further analysis

It can be seen that there are 3 cases:

case 1: by IMF₁The maximum value of the instantaneous amplitude of the component is taken as an example, and the IMF of each feeder line is assumed under the working condition of all single-phase earth faults in the power distribution network₁Instantaneous amplitude maximum a of the component_{1_Hi01}，a_{1_Hi02}，…， a_{1_Hi0m}Satisfies the calculation formula (18):

i.e. lambda_{1_Hi01}＝λ_{1_Hi02}＝…λ_{1_Hi0m}。

Taking the single-phase earth fault of the feeder 1 as an example, the characteristic instantaneous phase theta is not corrected_{1_Hi01}，θ_{1_Hi02}，θ_{1_Hi03}，……，θ_{1_Hi0m}Satisfies inequality (19):

therefore, the corrected branch feeder fault criterion of the invention can be derived, such as an inequality (20):

thereby, when the above inequality (20) is satisfied, it is determined that the single-phase ground fault has occurred on the feeder line 1.

Similarly, when the bus fails, the characteristic instantaneous phase theta is not corrected_{1_Hi01}，θ_{1_Hi02}，θ_{1_Hi03}，…，θ_{1_Hi0m}Satisfies inequality (21):

therefore, the corrected bus fault criterion of the invention can be derived, such as an inequality (22):

therefore, when the above inequality (22) is satisfied, it is determined that the bus bar has a single-phase ground fault.

The above is IMF₁The selection criterion constructed by taking the maximum value of the instantaneous amplitude of the component as an example also needs to take IMF as an example₁Second largest value of component instantaneous amplitude, IMF₁The 3 rd maximum value of the component instantaneous amplitude is taken as an example for constructing a line selection criterion, the specific construction process is completely consistent with that of the maximum value, and the difference is that the following criterion construction methods are completely the same by adopting the formulas (13), (14), (16) and (17) respectively.

Further analyzing the 2 nd and 3 rd cases, it can be found that the 2 nd and 3 rd cases have no practical physical significance in the fault line selection of the power distribution network, and therefore, cannot be used as a fault criterion, and the specific derivation is as follows:

case 2: when each feeder IMF₁Instantaneous amplitude maximum a of the component_{1_Hi01}，a_{1_Hi02}，…， a_{1_Hi0m}When formula (23) is satisfied: (by IMF)₁Component instantaneous amplitude maximum value as an example)

Then the derivation is:

as can be seen from the analysis, the formula (24) has no physical significance and is therefore not suitable.

Case 3: when each feeder IMF₁Component transientsMaximum value of amplitude a_{1_Hi01}，a_{1_Hi02}，…， a_{1_Hi0m}When formula (25) is satisfied: (by IMF)₁Component instantaneous amplitude maximum value as an example)

Then the derivation is:

as can be seen from the analysis, the formula (26) has no physical significance and is therefore not suitable.

By the same token, IMF₁The similar situation exists for the second largest, 3 rd largest instantaneous magnitude of a component!

3. Criterion 2: IMF1 component instantaneous energy relative entropy criterion;

relative entropy can be used to measure the difference of two waveforms. The smaller the relative entropy, the smaller the difference between the two waveforms; the larger the relative entropy, the larger the difference between the two waveforms. Using IMF₁The fault line selection criterion of the component instantaneous energy relative entropy theory comprises the following specific steps:

step 1: IMF for each feeder₁Maximum 3 instantaneous amplitudes a of the components_{1_Hi0m}、a_{2_Hi0m}、 a_{3_Hi0m}Respectively calculating the energy values, wherein the calculation formula is as shown in formula (27):

E_{1_Hi0m}＝|a_{1_Hi0m}|²，E_{2_Hi0m}＝|a_{2_Hi0m}|²，E_{3_Hi0m}＝|a_{3_Hi0m}|² (27)，

further, 3 energy sums were constructed, respectively,

finally, the instantaneous energy ratio is calculated, see equation (28):

step 2: calculating the instantaneous energy relative entropy of each feeder relative to the other feeders, as in equation (29):

and step 3: calculating the comprehensive relative energy entropy value of each feeder line, wherein IMF₁The comprehensive relative energy entropy corresponding to the maximum value of the instantaneous amplitude of the component is as follows:

the comprehensive relative energy entropy corresponding to the second largest value is:

the relative energy entropy corresponding to the minimum is:

and 4, step 4: and (5) correcting the comprehensive relative energy entropy of each feeder line by adopting a correction calculation formula of the calculation formula (30). Wherein, the correction calculation formula of the maximum instantaneous amplitude, the second maximum value and the 3 rd maximum value is as formula (30):

and 5: calculating the average value of the corrected comprehensive relative energy entropy, wherein the calculation formula is shown as a formula (31),

i.e. one for each feeder line

Therefore, a line selection criterion is formed, and specifically comprises the following steps: the largest 3 are selected

Are respectively in the order of magnitude

When it is satisfied with

When it is determined that the maximum value is

And the corresponding feeder line is a fault feeder line, otherwise, the fault feeder line is judged to be a bus fault.

4. Criterion 3: judging the polarity of the characteristic zero sequence current value;

according to the analysis, after the ground fault occurs, the fault judgment can be carried out by mistake only by using the line selection criterion of the polarity of the first half wave under the working conditions of voltage zero crossing fault, high resistance ground fault and the like, and the first half wave is not easy to define and obtain. Therefore, on this basis, according to the selected 3 characteristic instantaneous amplitudes, 3 zero-sequence current amplitude points corresponding to respective zero-sequence currents of each feeder line are found out, which are sequentially defined as a maximum, a second maximum and a 3 rd maximum characteristic zero-sequence current value, and are respectively expressed as: d_{1_l}，d_{2_l}，d_{3_l}. The fault feeder line is further judged by constructing a polarity calculation formula of 3 characteristic zero-sequence current values of each feeder line, and the method specifically comprises the following steps:

the polarity calculation expression is as follows (32):

when k is 1,2, 3, d_{1_l}，d_{2_l}，d_{3_l}Respectively representing the maximum value, the second maximum value, the 3 rd maximum value and r in the zero sequence current waveform of the feed line l_ljDenotes d_{k_l}And d_{k_j}Polarity of r between_ljIs +1, represents d_{k_l}And d_{k_j}The phases are the same; r is_ljIs-1, represents d_{k_l}And d_{k_j}The phases are opposite. The fault routing matrix R is obtained according to the formula (33), i.e.

Diagonal elements in the matrix R are all 1, and each non-diagonal element is a comparison result of the polarities of each two of the characteristic zero-sequence current amplitudes of each feeder line.

Thereby establishing criterion 3: the characteristic zero-sequence current value polarity criterion is specifically as follows: comparing the number of the elements of each row in the matrix R, which are '-1', to judge a fault feeder line; if the number of "-1" in the l-th row is m-1, the feeder line l has a fault, and other feeder lines are non-fault feeder lines; and if each row element in the matrix R is '1', the bus fault is judged.

5. Comprehensively voting;

the invention provides a comprehensive voting method by combining the 3 criteria, wherein the adopted principle is 'minority obeying majority', namely, when the 2 or more criteria judge that a branch feeder line has a fault, the branch feeder line has the fault; similarly, when 2 or more criteria determine that the bus fails, the voting bus fails; otherwise, if the output judgment of the 3 route selection criteria is inconsistent, the recalculation is returned, and then the voting is carried out.

The multi-criterion comprehensive voting power distribution network fault line selection method of the invention is further explained in detail by the specific embodiment as follows:

examples

The method comprises the following steps of establishing a 10kV radial power distribution network simulation model by using ATP-EMTP software, as shown in figure 1, wherein parameters of an overhead line and a cable line are shown in a table 1:

TABLE 1 line parameters

The method comprises the steps that constant load Z is adopted to be equivalent to each feeder line load, wherein Z is 100+ j20 omega, the compensation degree of an arc suppression coil is calculated according to 8% of overcompensation, and the inductance L of the arc suppression coil is 0.61633H through calculation; in addition, the active loss of the arc suppression coil generally accounts for2.5% -5.0% of inductive loss, 3% of the simulation example, and the arc suppression coil resistance r is calculated_L＝0.03ωL＝5.8058Ω。

In order to verify the correctness of the line selection method, the verification is respectively carried out from the aspects of a fault initial phase angle, a grounding resistance value, the compensation degree and the noise intensity, and the specific tests are as follows:

different fault initial phase angles: in order to test the accuracy of the method of the invention when the single-phase earth fault occurs in different voltage initial phase angles, respectively giving out the I on the overhead line₁Cable line l₂Cable-wire hybrid wire₃And the calculation results of 3 line selection criteria when the bus has a ground fault are shown in table 2. It can be seen that the method of the invention can accurately judge the fault feeder line at 0 degree, 30 degrees, 45 degrees, 60 degrees, 70 degrees and 90 degrees.

Different grounding resistance values: for the feeder l₁，l₂，l₃The buses respectively simulate grounding resistances of 10 omega, 100 omega, 200 omega, 500 omega and 1000 omega, and specific judgment results are shown in table 2.

Different compensation degrees: the fault current is further reduced due to the fact that the arc suppression coil is overcompensated left and right, therefore, the simulation overcompensation degrees are 5%, 8% and 10%, the judgment results are shown in the table 2, and it can be seen that the method is suitable for judging the fault feeder line under different overcompensation degrees, and the judgment results are accurate.

Different noise intensities: in practical application, the influence of external working condition noise needs to be considered, so that the signal-to-noise ratios are respectively tested to be 10dB, 7dB, 5dB and 3dB, and the test results are shown in Table 2. The method has strong capability of resisting external working condition noise interference, and can prepare to judge the fault feeder line under the noise interference.

Wherein, the given table 2(a) is the result of judging the maximum value of the characteristic instantaneous amplitude by adopting the criterion 1, the phase threshold value is set to be 40 degrees, and the criterion 1 can be seen to accurately judge a fault feeder line; similarly, it can be seen that the criterion 2 in table 2(b) and the criterion 3 in table 2(c) can accurately determine a faulty feeder line; furthermore, by combining a fault line selection comprehensive voting mechanism, the comprehensive voting method of the multi-criterion comprehensive voting power distribution network fault line selection method is accurate in judgment result.

TABLE 2 Fault determination results under different conditions

(a) Maximum value adopts criterion 1 to judge result

(b) Criterion 2 decision result

(c) Criterion 3 to judge the result

Therefore, comprehensively, The multi-criterion comprehensive voting power distribution network fault line selection method provided by The invention adopts a Complementary Ensemble Empirical Mode Decomposition (CEEMDAN) algorithm to perform feature extraction on The zero-sequence current of each feeder line, and obtains The 1 st Intrinsic Mode component (IMF) of each feeder line₁) Computing IMF using Hilbert transform₁And (3) respectively constructing 3 line selection criteria from different aspects by selecting the maximum 3 instantaneous amplitudes and the maximum instantaneous phases of the components, wherein the criteria are criteria 1,2 and 3 respectively, and finally forming a comprehensive voting method so as to realize fault line selection of the power distribution network. The criterion 1 of the invention is specifically as follows: finding out 3 instantaneous phases corresponding to 3 maximum instantaneous amplitudes of each feeder line, and correcting the 3 instantaneous phases by adopting the instantaneous amplitudes to obtain characteristic transientsA phase; if the absolute value of the difference between the characteristic instantaneous phase of a feed line and the characteristic instantaneous phase of the other feed lines is greater than 90 DEG lambda_{1_Hi0m}Judging that the feeder line is a fault feeder line; on the other hand, if it is less than 90 DEG.λ_{1_Hi0m}And judging that the bus is a fault feeder. Criterion 2 of the present invention is specifically: IMF with each feeder₁Calculating the energy relative entropy, the comprehensive relative entropy, the corrected comprehensive relative entropy and the average value of the corrected comprehensive relative entropy of each feeder line relative to other feeder lines on the basis of the 3 instantaneous amplitude values with the largest components

I.e. one for each feeder line

Thus, a line selection criterion is formed: the largest 3 are selected

Are respectively in the order of magnitude

When it is satisfied with

When it is determined that the maximum value is

And the corresponding feeder line is a fault feeder line, otherwise, the fault feeder line is judged to be a bus fault. Criterion 3 of the present invention is specifically: IMF with each feeder₁On the basis of 3 instantaneous amplitude values with the largest components, the largest 3 zero-sequence current values in the zero-sequence current waveforms of the feeder lines are correspondingly found out, so that a polarity calculation expression and a fault line selection matrix R are formed, and the fault feeder line can be judged by comparing the number of elements in each row of the matrix R, which is '-1'; if the number of "-1" in the l-th line is m-1, the feeder line l has a fault, and other feeder lines are healthy feeder lines; and if each row element in the matrix R is '1', the bus fault is judged.

Based on the above conclusions, the invention constructs a comprehensive voting method for the power distribution network so as to improve the accuracy of fault line selection, and specifically comprises the following steps: the invention provides a comprehensive voting method by combining the 3 route selection criteria, wherein the adopted principle is 'minority obeying majority', namely, when 2 or more criteria judge that a branch feeder line has a fault, the branch feeder line has the fault; similarly, when 2 or more criteria determine that the bus fails, the voting bus fails; otherwise, if the output judgment of the 3 route selection criteria is inconsistent, the recalculation is returned, and then the voting is carried out. A large number of simulations and field experiments show that the multi-criterion comprehensive voting power distribution network fault line selection method is not influenced by a primary phase angle, a grounding resistance value, a compensation degree, noise intensity and the like, and has high fault line selection accuracy.

Claims (5)

1. A multi-criterion comprehensive voting power distribution network fault line selection method is characterized by comprising the following steps:

step 1: preprocessing zero-sequence current data: the specific expression is as follows formula (9):

i_0m(t)＝i_0m[(t_f-T/2):(t_f+T/2)]-i_0m[(t_f-3T/2):(t_f-T/2)] (9)，

wherein i_0m(t) is zero sequence current after data preprocessing, m is a feeder line number, and t is 1,2, …, k; t is t_fIs the time of failure; t is a power frequency period; i.e. i_0m[(t_f-T/2):(t_f+T/2)]Is i_0mFrom the moment of failure t_fBefore T/2 to fault time T_fValues between the last T/2; i.e. i_0m[(t_f-3T/2):(t_f-T/2)]Is i_0mFrom the moment of failure t_fFirst 3T/2 to fault time T_fValues between the first T/2;

step 2: performing complementary set empirical mode decomposition;

and step 3: construction of fault line selection criterion 1-IMF₁Component feature instantaneous phase criterion;

and 4, step 4: construction of fault line selection criterion 2-IMF₁Component instantaneous energy relative entropy criterion;

and 5: constructing a fault line selection criterion 3-a characteristic zero-sequence current value polarity criterion;

step 6: according to the three fault line selection criteria in the steps 3, 4 and 5, when a single-phase earth fault occurs in the power distribution network, the three fault line selection criteria are calculated and judged at the same time, the principle that minority obeys majority is adopted for comprehensive voting, and when 2 or more criteria judge that a branch feeder line has a fault, the branch feeder line is voted for the fault; when the bus is judged to have a fault by 2 or more criteria, the voting bus has a fault; if the judgment results of the 3 route selection criteria are inconsistent, the calculation is returned again, and then the voting is carried out.

2. The method for fault line selection of a multi-criterion comprehensive voting power distribution network according to claim 1, wherein in the step 2, the complementary set empirical mode decomposition specifically comprises:

step 2.1: to i_0m(t) performing CEEMDAN decomposition to obtain each intrinsic mode component, and recording the intrinsic mode component as IMF, wherein the IMF component is a time-amplitude matrix IMF (i)_0m) The rows of the matrix correspond to the amplitudes and the columns correspond to the sampling points, the 1 st IMF component is adopted and is recorded as IMF₁To IMF₁The components are subjected to Hilbert transform to obtain Hi_0mThe specific expression is as follows:

step 2.2: to Hi_0mThe modulus and phase angle are calculated for each element, the formula (10) and (11) are calculated as follows:

die:

phase angle: theta [ Hi ]_0m(l,k)]＝arctan[y_m(l,k)/x_m(l,k)] (11)，

Wherein, Hi_0m(l,k)＝x_m(l,k)+jy_m(l, k), j is an imaginary unit; l is the number of the feeder line, and l is 1,2, …, m; k is 1,2, …, n; a (-) modulo operation;theta (-) is a phase angle calculation, and the calculated mode matrix and phase angle matrix respectively represent IMF₁The instantaneous amplitude and instantaneous phase of the component, denoted as a_{_Hi0m}And theta_{_Hi0m}；

Step 2.3: for IMF₁Instantaneous amplitude a of the component_{_Hi0m}The phase angles corresponding to the instantaneous amplitude values of the top 3 are called characteristic instantaneous phase positions in a descending order, and the instantaneous amplitude values are sequentially marked as a_{1_Hi0m}、a_{2_Hi0m}、a_{3_Hi0m}The characteristic instantaneous phase is sequentially marked as theta_{1_Hi0m}、θ_{2_Hi0m}、θ_{3_Hi0m}。

3. The method as claimed in claim 1, wherein in step 3, the fault line selection criterion 1-IMF is constructed₁The component characteristic instantaneous phase criterion specifically comprises the following steps:

step 3.1: using instantaneous amplitude a_{1_Hi0m}、a_{2_Hi0m}、a_{3_Hi0m}Characterizing instantaneous phase θ_{1_Hi0m}、θ_{2_Hi0m}、θ_{3_Hi0m}Accuracy, a modified confidence calculation is constructed, as shown in equations (12), (13) and (14):

wherein λ is_{1_Hi0m}，λ_{2_Hi0m}，λ_{3_Hi0m}The larger the value, the greater the confidence;

step 3.2: for constructing a line selection criterion 1 with quantitative characterization confidence, 3 maximum instantaneous moments of a feeder line m are calculatedAmplitude a_{1_Hi0m}、a_{2_Hi0m}、a_{3_Hi0m}Corresponding 3 characteristic instantaneous phases theta_{1_Hi0m}、θ_{2_Hi0m}、θ_{3_Hi0m}Correcting to obtain corrected characteristic instantaneous phase

The specific correction formula is as follows (15), (16) and (17):

step 3.3: by IMF₁Taking the corrected characteristic instantaneous phase corresponding to the maximum value of the instantaneous amplitude of the component and the single-phase earth fault of the feeder 1 as examples, the following formula (18) is satisfied:

i.e. lambda_{1_Hi01}＝λ_{1_Hi02}＝…λ_{1_Hi0m}Before correction, the characteristic instantaneous phase theta_{1_Hi01}，θ_{1_Hi02}，θ_{1_Hi03}，……，θ_{1_Hi0m}Satisfies the following inequality (19):

and constructing a corrected fault line selection criterion of the branch feeder line 1, wherein the following inequality (20) is as follows:

when the inequality is satisfied, judging that the feeder line 1 has a single-phase earth fault; the line selection criterion of whether the other branch feeders have single-phase earth faults is similar to the criterion of the branch feeder 1;

step 3.4: when the bus is in fault and before correction, the characteristic instantaneous phase theta_{1_Hi01}，θ_{1_Hi02}，θ_{1_Hi03}，…，θ_{1_Hi0m}Satisfies inequality (21):

and constructing a corrected bus fault line selection criterion, wherein the following inequality (22) is as follows:

and when the inequality is satisfied, judging that the bus has single-phase earth fault.

4. The method as claimed in claim 1, wherein in step 4, the fault line selection criteria 2-IMF are constructed₁The criterion of the relative entropy of the component instantaneous energy is specifically as follows:

step 4.1: IMF for each feeder₁Maximum 3 instantaneous amplitudes a of the components_{1_Hi0m}、a_{2_Hi0m}、a_{3_Hi0m}Calculating the energy values respectively according to the following formula (27):

E_{1_Hi0m}＝|a_{1_Hi0m}|²，E_{2_Hi0m}＝|a_{2_Hi0m}|²，E_{3_Hi0m}＝|a_{3_Hi0m}|² (27)，

further, 3 energy sums were constructed, respectively,

next, the instantaneous energy ratio is calculated as follows (28):

step 4.2: calculating the instantaneous energy relative entropy of each feeder relative to the other feeders, as shown in equation (29):

step 4.3: calculating the comprehensive relative energy entropy value of each feeder line, wherein IMF₁The comprehensive relative energy entropy corresponding to the maximum value of the instantaneous amplitude of the component is as follows:

the comprehensive relative energy entropy corresponding to the second largest value is:

the relative energy entropy corresponding to the 3 rd maximum value is:

step 4.4: and correcting the comprehensive relative energy entropy of each feeder line by adopting the following calculation formula, wherein the correction calculation formula of the maximum instantaneous amplitude, the second maximum value and the 3 rd maximum value is as the following formula (30):

step 4.5: calculating the average value of the corrected comprehensive relative energy entropy, wherein the calculation formula is as the following formula (31):

i.e. one W per feeder_l ^sTherefore, a fault route selection criterion 2 is constructed, and specifically comprises the following steps: the largest 3W are selected_l ^sIn the order of magnitude are

W_s ^s，W_t ^sWhen it is satisfied

When it is determined that the maximum value is

And the corresponding feeder line is a fault feeder line, otherwise, the fault feeder line is judged to be a bus fault.

5. The method for fault line selection of the multi-criterion comprehensive voting power distribution network according to claim 1, wherein in step 5, the step of constructing a fault line selection criterion 3-a characteristic zero-sequence current value polarity criterion specifically comprises:

step 5.1: according to the selected 3 characteristic instantaneous amplitude values a_{1_Hi0m}、a_{2_Hi0m}、a_{3_Hi0m}Finding out 3 zero-sequence current amplitude points corresponding to the respective zero-sequence currents of each feeder line, sequentially defining the amplitude points as a maximum characteristic zero-sequence current value, a second maximum characteristic zero-sequence current value and a 3 rd maximum characteristic zero-sequence current value, and respectively representing the values as follows: d_{1_l}，d_{2_l}，d_{3_l}(ii) a And (3) constructing a polarity calculation formula of 3 characteristic zero sequence current values of each feeder line, wherein the formula (32) is as follows:

wherein, when k is 1,2, 3, d_{1_l}，d_{2_l}，d_{3_l}Respectively representing the maximum value, the second maximum value, the 3 rd maximum value and r in the zero sequence current waveform of the feed line l_ljDenotes d_{k_l}And d_{k_j}Polarity of r between_ljIs +1, represents d_{k_l}And d_{k_j}The phases are the same; r is_ljIs-1, represents d_{k_l}And d_{k_j}The phases are opposite, and therefore, a fault line selection matrix R is obtained, as shown in the following formula (33):

wherein, diagonal elements in the matrix R are all 1, and each non-diagonal element is a comparison result of the polarities of each two of the zero sequence current amplitudes of the feeder line characteristics;

step 5.2: constructing a fault line selection criterion 3, which specifically comprises the following steps: comparing the number of the elements of each row in the matrix R, which are '-1', to judge the fault feeder; if the number of '-1' in the l-th line is m-1, the feeder line l is judged to be a fault feeder line, and meanwhile, other feeder lines are judged to be sound feeder lines; and if each row element in the matrix R is '1', judging that the bus is a fault feeder line, and simultaneously judging that other feeder lines are sound feeder lines.

Images