CN104764981A - Distribution network line fault section locating method based on standardization drift rate - Google Patents

Abstract

The invention discloses a distribution network line fault section locating method based on a standardization drift rate. The fault feature quantity is extracted from a ground fault by analyzing the phase current features of fault phase current before a fault occurs and during the time period between fault occurring and the action of an arc suppression coil, and the full-process waveform standardization drift rate is adopted for locating. Only the fault phase current of a line needs to be measured, the scheme is simple, high adaptability is achieved, the current ubiquitous problems that when a small current grounding system has a single-phase earth fault, the fault current is weak, reliability is poor, and sensitivity is low can be solved well, and meanwhile no interference is introduced into the system.

Description

Based on the distribution line fault section location method of normalized excursion degree

Technical field

The present invention relates to distribution network line fault diagnostic method, specifically a kind of distribution line fault section location method based on normalized excursion degree.

Background technology

According to statistics, electric system is in operational process, and the power outage caused by Distribution Network Failure accounts for more than 95% of total power outage, and wherein the accident of 70% is caused by singlephase earth fault or bus-bar fault.And distribution neutral point extensively adopts non-effective earthing (small current neutral grounding) mode both at home and abroad, to avoid causing power failure when there is singlephase earth fault.For the singlephase earth fault of distribution, because fault characteristic value is faint, lack reliable fault line selection and fault locating method always.Along with people are to the raising of distribution automation level requirement, more in the urgent need to fundamentally solving the fault-location problem of distribution.

The Fault Locating Method that current Chinese scholars proposes is roughly divided into two classes: one is injecting signal, and two is locate based on the section of fault characteristic value.Injecting signal comprises " S " injection method, the comprehensive injection method of alternating current-direct current and parallel medium resistance method, these class methods increase the interference to system, and can not detect instantaneity and intermittent grounding fault.Section location based on fault characteristic value comprises zero mould electric current relative method, section zero sequence admittance method, zero sequence reactive power direction method, the location based on difference of phase currents, residual method of addition, traveling wave method etc., electrical power distribution automatization system mainly utilizes main website to realize the time synchronized of FTU, is at least several millisecond to time error.In the case, the method such as to compare no longer valid for the amplitude of transient signal, polarity, Waveform Correlation.

Data after current localization method all only utilizes fault to occur, and have ignored the utilization to information before fault.Meanwhile, most of localization method only considers zero-sequence current information (needing three phase information), high to transformer request, and acquisition of information is complicated, and these methods will lose efficacy when data phase shortage.

Summary of the invention

The object of the invention is to overcome above-mentioned the deficiencies in the prior art, propose a kind of distribution line fault section location method based on normalized excursion degree.

Principle of the present invention:

If fault phase is A phase, before fault, check point phase voltage is after fault, phase voltage is by Fig. 1, easily know that the two meets

${{\stackrel{\·}{U}}^{\′}}_A={\stackrel{\·}{U}}_A-{\stackrel{\·}{U}}_0$

1) check point 1,2 is positioned at trouble spot heteropleural

After fault, trouble spot upstream detection point electric current is

${{\stackrel{\·}{I}}^{\′}}_{1A}={\stackrel{\·}{I}}_{\mathrm{LA}}+{{\stackrel{\·}{I}}^{\′}}_{C1A}+{\stackrel{\·}{I}}_f$

In formula, be respectively A phase load electric current, capacitance current and fault current.Wherein

$\begin{array}{c}{{\stackrel{\·}{I}}^{\′}}_{\mathrm{CA}}=\mathrm{j\ω}{C}_{1A}{{\stackrel{\·}{U}}^{\′}}_A=\mathrm{j\ω}{C}_{1A}({\stackrel{\·}{U}}_A-{\stackrel{\·}{U}}_0)\\ ={\stackrel{\·}{I}}_{C1A}-\mathrm{j\ω}{C}_{1A}{\stackrel{\·}{U}}_0\end{array}$

In formula, for capacitance current before fault.Before and after assumed fault occurs, a cycle internal loading electric current is constant, namely

${\stackrel{\·}{I}}_{1A}={\stackrel{\·}{I}}_{\mathrm{LA}}+{\stackrel{\·}{I}}_{C1A}$

Simultaneous (3) ~ (5) formula obtains

${{\stackrel{\·}{I}}^{\′}}_{1A}-{\stackrel{\·}{I}}_{1A}=-\mathrm{j\ω}{C}_{1A}{\stackrel{\·}{U}}_0+{\stackrel{\·}{I}}_f$

The inductive current that when arc suppression coil utilizes fault, neutral point voltage skew produces carrys out bucking-out system capacitance current, and earth point residual flow (fault current) can be expressed as

${\stackrel{\·}{I}}_f={\stackrel{\·}{I}}_{\mathrm{C\Σ}}-{\stackrel{\·}{I}}_L=\mathrm{j\ω}{C}_{\mathrm{\Σ}}{\stackrel{\·}{U}}_0-\frac{{\stackrel{\·}{U}}_0}{Z_L}$

In formula, C _Σfor the whole network line mutual-ground capacitor.Fault occur and in the arc suppression coil not operating period, | Z _l| very large, arc suppression coil compensation effect can be ignored.Then have

${{\stackrel{\·}{I}}^{\′}}_{1A}-{\stackrel{\·}{I}}_{1A}\≈\mathrm{j\ω}(C_{\mathrm{\Σ}}-C_{1A}){\stackrel{\·}{U}}_0$

In like manner, trouble spot detected downstream point phase current after fault

${{\stackrel{\·}{I}}^{\′}}_{2A}={\stackrel{\·}{I}}_{\mathrm{LA}}+{{\stackrel{\·}{I}}^{\′}}_{C2A}$

Variable quantity

${{\stackrel{\·}{I}}^{\′}}_{2A}-{\stackrel{\·}{I}}_{2A}=-\mathrm{j\ω}{C}_{2A}{\stackrel{\·}{U}}_0$

Visible, trouble spot heteropleural check point curent change feature is inconsistent, sized by concrete manifestation, phase place all not etc.

2) check point 1,2 is positioned at trouble spot homonymy

If check point is all positioned at upstream, trouble spot, according to 1) middle analysis, have

${{\stackrel{\·}{I}}^{\′}}_{1A}-{\stackrel{\·}{I}}_{1A}\≈\mathrm{j\ω}(C_{\mathrm{\Σ}}-C_{1A}){\stackrel{\·}{U}}_0$

If check point is all positioned at downstream, trouble spot, have

${{\stackrel{\·}{I}}^{\′}}_{1A}-{\stackrel{\·}{I}}_{1A}=-\mathrm{j\ω}{C}_{1A}{\stackrel{\·}{U}}_0$

${{\stackrel{\·}{I}}^{\′}}_{2A}-{\stackrel{\·}{I}}_{2A}=-\mathrm{j\ω}{C}_{2A}{\stackrel{\·}{U}}_0$

Difference due to adjacent check point electric capacity is section electric capacity, and its numerical value is very little, and therefore 2 curent change features are basically identical.

Step of the present invention:

If i _1A(n), i _2An phase current sampling sequence that () is adjacent check point.Be designated as zero under making fault that the data point in moment occurs, define two check point phase current variable quantities

Δi _1A(n)＝i _1A(n)-i _1A(n-N)

n∈[0,N-1]

Δi _2A(n)＝i _2A(n)-i _2A(n-N)

In formula, N=0.02f _s, be a periodic sampling point.For reducing fault resistance to the impact of localization method sensitivity, using the quadratic sum of small electric stream as braking amount, therefore define normalized excursion degree

$d_{p.u.}={\left[\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{[{\mathrm{\Δi}}_{1A}\left(n\right)-{\mathrm{\Δi}}_{2A}\left(n\right)]}^2}{\min \{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\Δi}}_{1A}{\left(n\right)}^2,\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\Δi}}_{2A}{\left(n\right)}^2}\right]}^{\frac{1}{2}}$

As the measurement index of phase current variable quantity otherness.Analyze according to upper joint, easily know the d of non-faulting section _p.u.close to 0, the d of fault section _p.u.for being greater than a positive number of 0.

Technical solution of the present invention is as follows:

A kind of distribution network line fault Section Location, its feature is, said method comprising the steps of:

Step S1, determines fault phase and fault moment t _f: when after generation earth fault being detected, according to phase voltage Changing Pattern, select fault phase, determine fault moment t according to phase voltage sudden change moment or arc-extinction device chugging moment _f;

Step S2, the operation time limit due to automatic tracking and compensating arc distinction device is 2 ~ 5 cycles, and the one-period internal fault information wherein after fault generation is the abundantest, therefore chooses a cycle, i.e. [t before and after pick-up unit fault _f-0.02s, t _f+ 0.02s] interval interior 2N the faulted phase current Wave data put altogether, if fault is designated as zero, normalized drift rate d under there is the data point in moment _p.u., formula is as follows:

$d_{p.u.}={\left[\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{[{\mathrm{\Δi}}_{1A}\left(n\right)-{\mathrm{\Δi}}_{2A}\left(n\right)]}^2}{\min \{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\Δi}}_{1A}{\left(n\right)}^2,\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{\mathrm{\Δi}}_{2A}{\left(n\right)}^2}\right]}^{\frac{1}{2}}$

In formula, Δ i _1A(n), Δ i _2An () is respectively two check point phase current variable quantities, formula is as follows:

Δi _1A(n)＝i _1A(n)-i _1A(n-N)

n∈[0,N-1]

Δi _2A(n)＝i _2A(n)-i _2A(n-N)

In formula, i _1A(n), i _2An phase current sampling sequence that () is adjacent check point, N is that a periodic sampling is counted;

According to normalized excursion degree size, step S3, judges whether each section is fault section, namely works as d _p.u.> d _p.u.set, be judged as fault section, otherwise be non-faulting section, wherein, d _p.u.setfor the action threshold value artificially arranged, get 1 ~ 2;

Step S4, according to network topology structure and check point distributing position on the line, travels through, successively until find out fault section.

Compared with prior art, the invention has the beneficial effects as follows: to earth fault, by analysis of failure phase current before fault occurs and the phase current feature of fault generation to arc suppression coil action, therefrom extract fault characteristic value, and adopt the waveform Euclidean distance of overall process to position.Therefore, only need the faulted phase current of measuring circuit, only consider the tradition of zero sequence (needing three phase information) before having broken, data acquisition is simple, applicability is strong; Synchronous from the angle of signal, total system adopts through the synchronous Monitoring Data of GPS, makes the difference between different check point sensitiveer.The problem that fault current faint, poor reliability when can solve current ubiquitous single-phase grounded malfunction in grounded system of low current very well, sensitivity are low, can not introduce interference to system simultaneously.

Accompanying drawing explanation

Fig. 1 is single-phase earthing of small current earthing system schematic diagram

Fig. 2 is distributed fault section positioning system structure figure

Fig. 3 is 10kV system emulation figure

Fig. 4 is earth fault current change quantity waveform along the line

Embodiment

Fault waveform required for the present invention comes from distributed fault section positioning system, and system architecture as shown in Figure 2.Distribution network failure section positioning system is by master station, transformer station's (bus) measurement mechanism and be distributed in distribution line node failure locating device everywhere and form.Circuit is divided into some sections by localization of fault node in topology, each node installation three groups of measurement mechanisms, and real-time synchronization gathers circuit three-phase current and voltage.

According to the Fault Locating Method of invention, in 10kV power distribution network analogue system, dissimilar fault is set.System construction drawing as shown in Figure 3, be 1., 2., 3. sector number, 2. go up at section by fault verification.Sample frequency is 20kHz (each cycle data point N=400), and the moment occurs fault is 0.7s, and during earth fault, arc-extinction device is set to 0.04s actuation time, and the threshold value of normalized excursion degree is set to 1.

Embodiment citing for fault section judges:

Step S1, system is according to after residual voltage start detection to the generation of earth fault, and record A phase in bus three-phase voltage and reduce, B, C phase raises, and is defined as A phase fault; Determine that fault moment is 0.71s according to the phase voltage sudden change moment;

Step S2, chooses the faulted phase current Wave data (being designated as zero under establishing fault that the data point in moment occurs) of totally 800 points in each pick-up unit [0,69,0.73] interval, according to the definition of normalized excursion degree, calculates d _p.u., result is as shown in table 1, and wherein, during 500 Ω stake resistance, fault phase current change quantity waveform along the line as shown in Figure 4.

Table 1 singlephase earth fault simulation result

Step S3, section normalized excursion degree 2. meets d _p.u.>1, is judged as its fault section; Otherwise 1., 3. section is non-faulting section.

Claims (3)

1., based on a distribution line fault section location method for normalized excursion degree, it is characterized in that, said method comprising the steps of:

Step S1, determines fault phase and fault moment t _f;

Step S2, chooses pick-up unit [t _f-0.02s, t _f+ 0.02s] interval interior 2N the faulted phase current Wave data put altogether, if fault is designated as zero, normalized drift rate d under there is the data point in moment _p.u., formula is as follows:

$d_{p.u.}={\left[\frac{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}{[\mathrm{\Δ}{i}_{1A}\left(n\right)-\mathrm{\Δ}{i}_{2A}\left(n\right)]}^2}{\min \{\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{\Δ}{i}_{1A}{\left(n\right)}^2,\underset{n=1}{\overset{N}{\mathrm{\Σ}}}\mathrm{\Δ}{i}_{2A}{\left(n\right)}^2\}}\right]}^{\frac{1}{2}}$

In formula, Δ i _1A(n), Δ i _2An () is respectively two check point phase current variable quantities, formula is as follows:

Δi _1A(n)＝i _1A(n)-i _1A(n-N)

n∈[0,N-1]

Δi _2A(n)＝i _2A(n)-i _2A(n-N)

In formula, i _1A(n), i _2An phase current sampling sequence that () is adjacent check point, N is that a periodic sampling is counted;

According to normalized excursion degree size, step S3, judges whether each section is fault section, namely works as d _p.u.> d _p.u.set, be judged as fault section, otherwise be non-faulting section;

Step S4, according to network topology structure and check point distributing position on the line, travels through, successively until find out fault section.

2. distribution network line fault localization method according to claim 1, is characterized in that, described step S1 comprises:

When after generation earth fault being detected, according to phase voltage Changing Pattern, select fault phase, determine fault moment t according to phase voltage sudden change moment or arc-extinction device chugging moment _f.

3. distribution network line fault localization method according to claim 1, is characterized in that d in described step S3 _p.u.setfor action threshold value, get 1 ~ 2.