CN105067948A - Small-current grounding line selection device and single-phase grounding detection method - Google Patents

Abstract

The invention discloses a small-current grounding line selection device and a single-phase grounding detection method. According to the single-phase grounding detection method, through simultaneously sampling a three-phase bus voltage and a three-phase line current and calculation of a least square filter, the magnitude and phase of each of the voltage and the current can be obtained, and furthermore a phase difference between the time before a fault and the time after the fault can be obtained. Furthermore through real-time calculation for the complex impedance of a line, an impedance change in the fault is detected. Through the two characteristics of the phase difference and the impedance change, the single-phase grounding fault characteristic of the line can be accurately determined and is not affected by direction of a loop power flow. The single-phase grounding detection method is a new solution for detecting a small-current grounding fault.

Description

A kind of Small Electric Current Earthing And Routing Device and single phase earth fault detection method

Technical field

The present invention relates to a kind of Small Electric Current Earthing And Routing Device and single phase earth fault detection method.

Background technology

As everyone knows, the transmission line of electricity of electrical network is different according to substation transformer neutral grounding mode, be divided into solidly grounded system and non-direct ground system, ultra-high-tension power transmission line (220kV and more than) belongs to solidly grounded system, is referred to as heavy current grounding system again.Low-voltage circuit (110kV and following) belongs to transformer neutral point non-direct ground system, is referred to as small current neutral grounding system again.Because small current neutral grounding system is responsible for by electric power distribution to final user, therefore, be referred to as power distribution network again, be called for short distribution.But because distribution is in densely populated areas, and circuit close building, residential block etc., therefore, also easily cause the accident by various interference.According to the national grid stroke analysis of 2015, single-phase earthing accounts for 80% of Distribution Network Failure.But because distribution Fault feature of single phase earthing is not obvious, the detection and treatment of singlephase earth fault is comparatively difficult.For this reason, one-phase earthing failure in electric distribution network Processing Technology Research is classified as emphasis problem in 2015 by national net company.This illustrates that distribution phase-to-ground fault detection does not fundamentally solve always.The disposal route generally adopted at present is the method for operating a switch, that is, when after generation singlephase earth fault, owing to can not judge it is which bar line fault, therefore, person on duty takes to operate a switch one by one test method(s), if after certain circuit is operated a switch, phenomenon of the failure disappears, then illustrate that fault is on this circuit.This method has had a strong impact on the power supply quality of resident, also have impact on the power supply reliability index of grid company.

From fault signature, after small current system generation single-phase earthing, fault phase ground current is little, and A, B, C three-phase phase is constant, and field apparatus can continuous service a period of time (regulatory requirements is within 2 hours), can increase the reliability of power supply like this.But in order to make fault eliminate rapidly, reducing fault face, just requiring find in time and excise faulty line.From the beginning of the sixties at the end of the fifties in last century China's First small current system earthing line selecting device succeed in developing, small current line selection equipment experienced by the trials and hardships of decades.But actual field operation result shows, the route selection effect of this Small Electric Current Earthing And Routing Device is unsatisfactory.Large quantities of work about electric power person is still devoted to the research improving route selection accuracy rate.

From Cleaning Principle, the method detecting small current neutral grounding at present mainly contains steady-state component method, harmonic component method, transient method, wavelet method.Wherein, steady-state component method, is divided into again zero-sequence current than execution, zero-sequence current phase comparing method, and colony is than volt phase comparing method.This method utilizes the zero-sequence current of faulty line to be numerically equal to non-fault line zero-sequence current sum, and namely the zero-sequence current of faulty line is maximum.So just find out faulty line by the amplitude of alternative route zero-sequence current.This method is a kind of method desirable in laboratory, length is had to have for on-the-spot central each bar circuit short, the laod unbalance of each bar outlet, institute's Current Transformer (TA) neither complete equipilibrium, and the circuit like this with regard to causing zero-sequence current maximum differs, and to establish a capital be faulty line.Based on what time everybody adds phase compare again except carrying out amplitude com parison above, because faulty line and non-fault line phase place are contrary, so just compensate for the unbalanced impact of outlet.Improve the accuracy of route selection.But everybody is generally circle collection to the sampling of fault amount from device inside, divide exactly and gather just complete for the calculating data acquisition of all outlets several times, there is a drawback is like this exactly the fault amount not gathering all outlets at synchronization completely, just easily occur erroneous judgement, this method is not suitable for the system of grounding through arc yet.

Harmonic component method, is divided into again 5 subharmonic size and Orientations, the methods such as each harmonic quadratic sum.But 5 subharmonic content are too little, can not effectively judge, this has just become the bottleneck of this method.

Transient state route selection method, half-wave method and wavelet analysis method headed by dividing again.First half-wave method occurs in phase voltage close to this hypothesis of maximum instantaneous based on ground connection, now fault phase capacitance charge is discharged to trouble spot by fault phase circuit, utilizes the amplitude of the first half-wave of faulty line transient zero-sequence current and the voltage feature all different from normal condition with direction to realize route selection.But the prerequisite that this method exists is fault must occur in phase voltage close to maximal value moment.

Wavelet analysis method utilizes principle of wavelet analysis to carry out Accurate Analysis to signal, particularly to transient signal and feeble signal more responsive, reliably can extract fault signature.Utilize wavelet transformation a signal analysis can be become the small echo sum of different scale and position, after utilizing suitable small echo and wavelet basis to convert transient, easily tell faulty line and non-fault line.

Above four kinds of methods, steady-state component method and wavelet method need the transient characteristic of the whole circuit of Synchronization Analysis, and calculated amount is very large, and Harmonic Method is too little because of 5 these harmonic contents, can not effectively detect, and transient state route selection method is based upon on supposed premise, and actual use is little.

In sum, from Cleaning Principle, small current earthing wire-selecting still lacks reliable and effective method.

Summary of the invention

The present invention is in order to solve the problem, propose a kind of Small Electric Current Earthing And Routing Device and single phase earth fault detection method, this invention, by Simultaneous Monitoring line voltage distribution and electric current phasor, calculates its complex impedance in real time, when distribution line is normal, impedance is less, and voltage current phase relation is comparatively stable, when there is singlephase earth fault, fault phase is except normal current, also create vagabond current, impedance has obvious change, and electric current and voltage-phase also create larger difference.In addition, when there is single-phase earthing, have an erratic process instantaneously, during ground connection, feature is more obvious.Utilizing this method, when without the need to comparing with the feature of other circuit, accurately can detect the fault of ground path.The method not only has very strong applicability, but also greatly reduces amount of calculation, and make pick-up unit have very high cost performance, economic and social benefit is huge.

To achieve these goals, the present invention adopts following technical scheme:

A kind of small current neutral grounding single phase earth fault detection method, comprises the following steps:

(1) according to the form of expression and the testing goal of voltage, current signal, determine time window and sampling period, calculate least square filter;

(2) size of calculating voltage, electric current and both phase differential;

(3) by the phasor value of complex representation voltage and current, complex impedance is calculated;

(4) when the Voltage-current phase difference in former and later two sampling time windows and impedance variation all exceed respective threshold value, then judge low current grounding occurs.

In described step (1), the concrete grammar of the representation and testing goal of specifying voltage and current signal is:

The voltage of distribution outlet and phase current after voltage transformer (VT) summation current transformer converts, the voltage signal with DC component, characteristic component and harmonic components thereof that all can be expressed as:

$u\left(t\right)=U_0{e}^{-\mathrm{\λ}t}+\underset{k=1}{\overset{M}{\Σ}}{U}_{k}sin(k\mathrm{\ω}t+{\mathrm{\θ}}_k)---\left(1\right)$

In formula: U ₀-DC component, represents composition non-periodic in signal;

λ-DC component damping time constant;

U _k-kth order harmonic components peak value, k=1 ..., M;

θ _kthe phasing degree of-kth subharmonic;

ω-rotating speed angular frequency, ω=2 π f, f-50Hz, network operation frequency, is called for short power frequency;

U ₁and θ ₁correspond to the amplitude of power frequency and the phasing degree relative to sampling time window.

In described step (1), determine that the size of time window and the method in sampling period are:

Time window, refer to CPU single treatment from first sample to last sample time span, the size T of time window _wequal the function natural period, sampling period T is less than the function natural period;

Determine the size T of time window _w, and between sampling period T after, the sample number N defining method in a sample window is: N and T _wand the pass of T is:

$N=\frac{T_w}{T}---\left(2\right)$

In described step (1), the concrete grammar calculating least square filter is:

By the U of voltage signal ₀e ^{-λ t}launch by Taylor's radix and get first two, U _ksin (k ω t+ θ _k) launch by trigonometric function, then formula (1) can be expressed as form:

$u\left(t\right)=U_0-U_0\mathrm{\λ}t+\underset{k=1}{\overset{M}{\Σ}}sin\left(k\mathrm{\ω}t\right)\×U_{k}cos\left({\mathrm{\θ}}_k\right)+\underset{k=1}{\overset{M}{\Σ}}cos\left(k\mathrm{\ω}t\right)\×U_{k}sin\left({\mathrm{\θ}}_k\right)---\left(3\right)$

This function is at sampling instant t _i, with its sample u (t _i) between inevitable equal, equation of namely sampling is:

$U_0-U_0{\mathrm{\λt}}_i+\underset{k=1}{\overset{M}{\Σ}}sin\left({\mathrm{k\ωt}}_i\right)\×U_{k}cos\left({\mathrm{\θ}}_k\right)+\underset{k=1}{\overset{M}{\Σ}}cos\left({\mathrm{k\ωt}}_i\right)\×U_{k}sin\left({\mathrm{\θ}}_k\right)=u\left(t_i\right)---\left(4\right)$

T _i-the i-th sampling instant;

Due to sampling instant t _iknown, therefore, sin (k ω t in formula (4) _i), cos (k ω t _i) for being known quantity, N number of sampling equation is listed one by one by formula (4), and obtain a matrix equation as follows, by t _irepresent with τ:

$\left(\begin{array}{ccccccc}1& \mathrm{\τ}& \sin \left(\mathrm{\ω}\mathrm{\τ}\right)& \cos \left(\mathrm{\ω}\mathrm{\τ}\right)& \mathrm{...}& \sin \left(M\mathrm{\ω}\mathrm{\τ}\right)& \cos \left(M\mathrm{\ω}\mathrm{\τ}\right)\\ 1& 2\mathrm{\τ}& \sin \left(\mathrm{\ω}2\mathrm{\τ}\right)& \cos \left(\mathrm{\ω}2\mathrm{\τ}\right)& \mathrm{...}& \sin \left(M\mathrm{\ω}2\mathrm{\τ}\right)& \cos \left(M\mathrm{\ω}2\mathrm{\τ}\right)\\ & .& .& .& \mathrm{...}& .& .\\ 1& N\mathrm{\τ}& \sin \left(\mathrm{\ω}N\mathrm{\τ}\right)& \cos \left(\mathrm{\ω}N\mathrm{\τ}\right)& \mathrm{...}& \sin \left(M\mathrm{\ω}N\mathrm{\τ}\right)& \cos \left(M\mathrm{\ω}N\mathrm{\τ}\right)\end{array}\right)\left(\begin{array}{c}U0\\ -\mathrm{\λ}U0\\ U_1\cos \left({\mathrm{\θ}}_1\right)\\ U_1\sin \left({\mathrm{\θ}}_1\right)\\ .\\ U_M\cos \left({\mathrm{\θ}}_M\right)\\ U_M\sin \left({\mathrm{\θ}}_M\right)\end{array}\right)=\left(\begin{array}{c}u\left(1\right)\\ u\left(2\right)\\ .\\ u\left(N\right)\end{array}\right)---\left(5\right)$

By matrix of coefficients A, known variables matrix X, sampling matrix U represents, then matrix equation (5) can be expressed as follows again:

A·X＝U(6)

Therefore, A is that a N is capable, what 2 (M+1) arranged determines parameter matrix, X is the matrix of variables of one 2 (M+1) row 1 row, U be capable 1 row of N determine parameter matrix, make 2 (M+1)=N, A with regard to a square formation, and the highest identifiable design harmonic multiples M=N/2-1, meet Shen agriculture sampling thheorem;

According to matrix theory, the condition that coefficient matrices A exists inverse matrix A-1 is: | A| ≠ 0, and A*A-1=I, I are unit matrix, namely diagonal entry is 1;

Therefore, matrix of unknown parameters to be measured is:

X＝A ^-1·U(7)

A ^-1be least square filter.

In described step (2), concrete grammar is, and: matrix of variables X is,

$X=\left(\begin{array}{c}U0\\ -\mathrm{\λ}U0\\ U_1\cos \left({\mathrm{\θ}}_1\right)\\ U_1\sin \left({\mathrm{\θ}}_1\right)\\ .\\ U_M\cos \left({\mathrm{\θ}}_M\right)\\ U_M\sin \left({\mathrm{\θ}}_M\right)\end{array}\right),$

Use x _ithe element of i-th row of representing matrix X, then: x ₁=U ₀, i.e. DC component, and

x ₃＝U ₁cosθ ₁,x ₄＝U ₁sinθ ₁(8)

X ₃be exactly the X-axial projection of voltage or current phasor, x ₄for Y-axial projection of shaking, therefore, the peak-to-peak value of voltage, electric current is:

$U_1=\sqrt{{x_3}^2+{x_4}^2}---\left(9\right)$

The effective value of voltage or electric current is:

Relative to the function phase angle of time window starting point be:

${\mathrm{\θ}}_1={\tan }^{-1}\frac{U_1{\mathrm{sin\θ}}_1}{U_1{\mathrm{cos\θ}}_1}={\tan }^{-1}\frac{x_4}{x_3}---\left(10\right)$

When distribution line normally works, voltage-phase is ahead of electric current, and phase differential is therebetween:

Δθ ₀＝θ _0u-θ _0i(11)

Wherein, θ _0uand θ _0ibe the voltage-phase and electric current that obtain according to formula (10) respectively, when there is single-phase earthing, phase differential is therebetween:

Δθ ₁＝θ _1u-θ _1i(12)

Therefore, the phase place produced during fault is changed to:

Δθ＝Δθ ₁-Δθ ₀(13)

In described step (3), the process calculating complex impedance size is specially:

The complex representation form of voltage and current phasor is respectively:

U＝U _x+i·U _y(14)

I＝I _x+i·I _y(15)

Wherein, U _xand I _xbe the x in formula (8) ₃, wherein, U _yand I _ybe the x in formula (8) ₄numerical value.

Complex impedance is:

$Z=\frac{U}{I}=\frac{U_x+i\·U_y}{I_x+i\·I_y}=x+i\·y---\left(16\right)$

The size of complex impedance is:

$\left|Z\right|=\sqrt{x^2+y^2}---\left(17\right).$

In described step (4), the determination methods that low current grounding is reported to the police is, when the Voltage-current phase difference in former and later two sampling time windows and impedance variation all exceed respective threshold value, then sends earth fault and reports to the police, that is:

If | Z| > Z ₀, and | Δ θ | > θ _s, then report to the police;

Wherein, Z ₀and θ _sbe respectively impedance settings and phasing degree setting value.

A kind of Small Electric Current Earthing And Routing Device based on said method, comprise route protection unit and controller, wherein, route protection unit comprises current potential conversion module and MCU, wherein, the input end of route protection unit connects voltage current transformer, AC signal is changed into MCU by current potential conversion module can sampled signal, controller comprises analog-to-digital conversion module, sampled signal can convert digital signal to, MCU calculates least square filter and voltage, the phase differential of electric current, be transferred to controller, controller calculates complex impedance, judge whether low current grounding occurs.

Based on a Little Current Ground Connection System for said method, comprise and be multiplely distributed in above-mentioned Small Electric Current Earthing And Routing Device on distribution network transformer substation circuit and main frame, main frame connects each Small Electric Current Earthing And Routing Device, and whether monitoring distribution line breaks down in real time.

Beneficial effect of the present invention is:

(1) small current earthing wire-selecting method of the present invention's proposition is more accurate and sensitive than current known method; Why say that the method is sensitive and accurate, reason has two, the first, other method is not directly measured and is compared the phase relation of electric current and voltage, only adopting traditional phase-sequence component method to obtain positive sequence, negative phase-sequence, zero sequence, is a kind of more general three-phase fault determination methods, and due to the effect of neutral point ground capacitance after singlephase earth fault, the phase place of phase current and phase voltage there occurs very large change, can directly and accurately judge which creates SPA sudden phase anomalies mutually.Its two, after single-phase earthing, the impedance of circuit is changed suddenly due to adding of capacitive reactances by emotional resistance, changes in amplitude is obvious, therefore easily select, and other method has no employing impedance method, two kinds of judgements adopt simultaneously, add accuracy and the reliability of detection.

(2) small current earthing wire-selecting method that the present invention proposes no longer needs fault phase and whole healthy phases to screen one by one, improves reaction velocity;

(3) the present invention is by real-time computational scheme complex impedance, and impedance variation when its fault being detected, by these two features, can accurately judge line single phase grounding failure feature, and not affect by direction of tide during looped network;

(4) adopt the present invention propose detection method, be expected to the small current neutral grounding difficult problem thoroughly solving distribution, can further improve the power supply reliability index of electrical network, to power grid enterprises and society all significant.

Accompanying drawing explanation

Fig. 1 VIPS-100 visual protection and automated system structural drawing;

Fig. 2 small current earthing wire-selecting unit principle schematic.

Embodiment:

Below in conjunction with accompanying drawing and embodiment, the invention will be further described.

For detecting distribution line singlephase earth fault, the three-phase current of device synchro measure distribution network transformer substation low-pressure side bus three-phase voltage and each outlet, the voltage and current phase differential of the every phase of real-time calculating and complex impedance thereof, when there is impedance and SPA sudden phase anomalies, namely thinking and there occurs singlephase earth fault.Below for realizing four steps and the method for the inventive method.

Step 1: calculate least square filter

Step 2: the size of calculating voltage, electric current and phase differential

Step 3: the size calculating complex impedance

Step 4: compare the Voltage-current phase difference in former and later two sampling time windows and impedance variation, when exceeding threshold value, then sends low current grounding and reports to the police.

Described step 1) to calculate the process of least square filter as follows:

First, representation and the testing goal of voltage and current signal is specified.

The voltage of distribution outlet and phase current after voltage transformer (VT) summation current transformer converts, the voltage signal with DC component, characteristic component and harmonic components thereof that all can be expressed as:

$u\left(t\right)=U_0{e}^{-\mathrm{\λ}t}+\underset{k=1}{\overset{M}{\Σ}}{U}_{k}sin(k\mathrm{\ω}t+{\mathrm{\θ}}_k)---\left(1\right)$

In formula: U ₀-DC component, represents composition non-periodic in signal;

λ-DC component damping time constant;

U _k-kth order harmonic components peak value, k=1 ..., M;

θ _kthe phasing degree of-kth subharmonic;

ω-rotating speed angular frequency, ω=2 π f, f-50Hz, network operation frequency, is called for short power frequency.

U ₁and θ ₁being correspond to the amplitude of power frequency and the phasing degree relative to sampling time window, is the major parameter needing this method to detect.

Secondly, need the size determining time window, and the sampling period.

So-called time window, refer to CPU single treatment from first sample to last sample time span.This time span is generally greater than or equals the natural period of function, and concerning 50Hz signal, the function natural period is 20ms, and the size Tw of time window will equal the function natural period.Sampling period T is less than the function natural period.

Determine the size Tw of time window, and between sampling period T after, the sample number N in a sample window also just determines.The pass of N and Tw and T is:

$N=\frac{T_w}{T}---\left(2\right)$

In order to obtain least square filter, first formula (1) is done following process:

Launched by Taylor's radix by U0e-λ t in formula (1) and get first two, Uksin (k ω t+ θ k) launches by trigonometric function, then formula (1) can be expressed as form:

$u\left(t\right)=U_0-U_0\mathrm{\λ}t+\underset{k=1}{\overset{M}{\Σ}}sin\left(k\mathrm{\ω}t\right)\×U_{k}cos\left({\mathrm{\θ}}_k\right)+\underset{k=1}{\overset{M}{\Σ}}cos\left(k\mathrm{\ω}t\right)\×U_k\sin \left({\mathrm{\θ}}_k\right)---\left(3\right)$

This function is at sampling instant ti, inevitable equal with between its sample u (ti), and equation of namely sampling is:

$U_0-U_0{\mathrm{\λt}}_i+\underset{k=1}{\overset{M}{\mathrm{\Σ}}}\sin \left({\mathrm{k\ωt}}_i\right)\×U_k\cos \left({\mathrm{\θ}}_k\right)+\underset{k=1}{\overset{M}{\mathrm{\Σ}}}\cos \left({\mathrm{k\ωt}}_i\right)\×U_k\sin \left({\mathrm{\θ}}_k\right)=u\left(t_i\right)---\left(4\right)$

T _i-the i-th sampling instant.

Because sampling instant ti is known, therefore, in formula (4), sin (k ω ti), cos (k ω ti) also can think known quantity, only have U ₀, λ, Ukcos (θ k), Uksin (θ k) be unknown quantity.If N number of sampling equation is listed one by one by formula (4), and obtain a matrix equation as follows (for equation for purpose of brevity, ti τ is represented):

$\left(\begin{array}{ccccccc}1& \mathrm{\τ}& \sin \left(\mathrm{\ω}\mathrm{\τ}\right)& \cos \left(\mathrm{\ω}\mathrm{\τ}\right)& \mathrm{...}& \sin \left(M\mathrm{\ω}\mathrm{\τ}\right)& \cos \left(M\mathrm{\ω}\mathrm{\τ}\right)\\ 1& 2\mathrm{\τ}& \sin \left(\mathrm{\ω}2\mathrm{\τ}\right)& \cos \left(\mathrm{\ω}2\mathrm{\τ}\right)& \mathrm{...}& \sin \left(M\mathrm{\ω}2\mathrm{\τ}\right)& \cos \left(M\mathrm{\ω}2\mathrm{\τ}\right)\\ & .& .& .& \mathrm{...}& .& .\\ 1& N\mathrm{\τ}& \sin \left(\mathrm{\ω}N\mathrm{\τ}\right)& \cos \left(\mathrm{\ω}N\mathrm{\τ}\right)& \mathrm{...}& \sin \left(M\mathrm{\ω}N\mathrm{\τ}\right)& \cos \left(M\mathrm{\ω}N\mathrm{\τ}\right)\end{array}\right)\left(\begin{array}{c}U0\\ -\mathrm{\λ}U0\\ U_1\cos \left({\mathrm{\θ}}_1\right)\\ U_1\sin \left({\mathrm{\θ}}_1\right)\\ .\\ U_M\cos \left({\mathrm{\θ}}_M\right)\\ U_M\sin \left({\mathrm{\θ}}_M\right)\end{array}\right)=\left(\begin{array}{c}u\left(1\right)\\ u\left(2\right)\\ .\\ u\left(N\right)\end{array}\right)---\left(5\right)$

If by matrix of coefficients A, known variables matrix X, sampling matrix U represents, then matrix equation (5) can be expressed as follows again:

A·X＝U(6)

Therefore, A is that a N is capable, and what 2 (M+1) arranged determines parameter matrix, and X is the matrix of variables of one 2 (M+1) row 1 row, U be capable 1 row of N determine parameter matrix.If make 2 (M+1)=N, then A is with regard to a square formation, and the highest identifiable design harmonic multiples M=N/2-1, meet Shen agriculture sampling thheorem.

According to matrix theory, the condition that coefficient matrices A exists inverse matrix A-1 is: | A| ≠ 0, and A*A-1=I.I is unit matrix, and namely diagonal entry is 1.

Therefore, matrix of unknown parameters to be measured is:

X＝A ^-1·U(7)

A-1 is least square filter.

Step 2) process of calculating voltage, the size of electric current and phase differential is as follows:

Matrix of variables X in formula (6), (7) is the unknown column matrix in formula (5) left side:

$X=\left(\begin{array}{c}U0\\ -\mathrm{\λ}U0\\ U_1\cos \left({\mathrm{\θ}}_1\right)\\ U_1\sin \left({\mathrm{\θ}}_1\right)\\ .\\ U_M\cos \left({\mathrm{\θ}}_M\right)\\ U_M\sin \left({\mathrm{\θ}}_M\right)\end{array}\right)$

If use x _ithe element of i-th row of representing matrix X, then: x ₁=U ₀, i.e. DC component.

x ₃＝U ₁cosθ ₁,x ₄＝U ₁sinθ ₁(8)

X ₃be exactly the X-axial projection of voltage or current phasor, x ₄for Y-axial projection of shaking.Therefore, the peak-to-peak value of voltage, electric current is:

$U_1=\sqrt{{x_3}^2+{x_4}^2}---\left(9\right)$

The effective value of voltage or electric current is:

Relative to the function phase angle of time window starting point be:

${\mathrm{\θ}}_1={\tan }^{-1}\frac{U_1{\mathrm{sin\θ}}_1}{U_1{\mathrm{cos\θ}}_1}={\tan }^{-1}\frac{x_4}{x_3}---\left(10\right)$

When distribution line normally works, voltage-phase is ahead of electric current, and phase differential is therebetween:

Δθ ₀＝θ _0u-θ _0i(11)

Wherein, θ _0uand θ _0ithe voltage-phase and electric current that obtain according to formula (10) respectively.When there is single-phase earthing, phase differential is therebetween:

Δθ ₁＝θ _1u-θ _1i(12)

Therefore, the phase place produced during fault is changed to:

Δθ＝Δθ ₁-Δθ ₀(13)

Described step 3) calculate complex impedance size process as follows:

The complex representation form of voltage and current phasor is respectively:

U＝U _x+i·U _y(14)

I＝I _x+i·I _y(15)

Wherein, U _xand I _xbe the x in formula (8) ₃, wherein, U _yand I _ybe the x in formula (8) ₄numerical value.

Complex impedance is:

$Z=\frac{U}{I}=\frac{U_x+i\·U_y}{I_x+i\·I_y}=x+i\·y---\left(16\right)$

The size of complex impedance is:

$\left|Z\right|=\sqrt{x^2+y^2}---\left(17\right)$

Described step 4) low current grounding report to the police determination methods be, when in former and later two sampling time windows Voltage-current phase difference and impedance variation all exceed respective threshold value time, then send earth fault report to the police.That is:

If | Z| > Z ₀, and | Δ θ | > θ _s, then report to the police.

Wherein, Z ₀and θ _sbe respectively impedance settings and phasing degree setting value.

The system architecture of device is a general-purpose platform, can be adapted to each grade transformer substation.But route protection unit is wherein specialized designs small current earthing wire-selecting function then, designs specially for 110kV and following distribution network transformer substation route protection.System protection unit is divided into main-transformer (abbreviation main transformer) protection, route protection, capacitor protection etc. by device type, has how many equipment just to have how many protected locations.All protected locations form a LAN (Local Area Network) eventually through 100Mbps network.Be provided with the controller of two redundancies each other in each protected location, intelligent object is connected as a single entity by TBUS network by each controller.Active and standby controller lays respectively at the two ends of TBUS net, and during one of them controller failure, another controller can adapter TBUS.System architecture can see Fig. 1.

Small current earthing wire-selecting function of the present invention is positioned at route protection unit.Wherein, each route protection unit is sampled busbar voltage (three-phase) and line current (three-phase) all simultaneously.Route protection unit adopts digital Direct Sampling and digital processing technology, and signal sample circuit is see Fig. 2.In order to play respective performance advantage, the MCU of route protection unit is responsible for the calculating of formula of the present invention (7), (8), (9), and the calculating of formula (10) ~ (17) is responsible for by active and standby controller.The network refresh time of TBUS is 20ms, can complete a Refresh Data in a power frequency period, ensure that to judge whether to break down in a power frequency period.

In Fig. 2, the AC signal (u1, u2) from voltage current transformer converts through current potential, becomes the signal (Vo1, Vo2) that MCU can sample.MCU inside is integrated with A/D converter, and CPU directly carries out computing to digital signal.Result of calculation to be communicated to active and standby controller in the mode of TBUS by inner UART mouth again.

By reference to the accompanying drawings the specific embodiment of the present invention is described although above-mentioned; but not limiting the scope of the invention; one of ordinary skill in the art should be understood that; on the basis of technical scheme of the present invention, those skilled in the art do not need to pay various amendment or distortion that creative work can make still within protection scope of the present invention.

Claims (10)

1. a small current neutral grounding single phase earth fault detection method, is characterized in that: comprise the following steps:

(1) according to the form of expression and the testing goal of voltage, current signal, determine time window and sampling period, calculate least square filter;

(2) size of calculating voltage, electric current and both phase differential;

(3) by the phasor value of complex representation voltage and current, complex impedance is calculated;

(4) when the Voltage-current phase difference in former and later two sampling time windows and impedance variation all exceed respective threshold value, then judge low current grounding occurs.

2. a kind of small current neutral grounding single phase earth fault detection method as claimed in claim 1, is characterized in that: in described step (1), and the concrete grammar of the representation and testing goal of specifying voltage and current signal is:

The voltage of distribution outlet and phase current after voltage transformer (VT) summation current transformer converts, the voltage signal with DC component, characteristic component and harmonic components thereof that all can be expressed as:

$u\left(t\right)=U_0{e}^{-\mathrm{\λ}t}+\underset{k=1}{\overset{M}{\Σ}}{U}_{k}sin(k\mathrm{\ω}t+{\mathrm{\θ}}_k)---\left(1\right)$

In formula: U ₀-DC component, represents composition non-periodic in signal;

λ-DC component damping time constant;

U _k-kth order harmonic components peak value, k=1 ..., M;

θ _kthe phasing degree of-kth subharmonic;

ω-rotating speed angular frequency, ω=2 π f, f-50Hz, network operation frequency, is called for short power frequency;

U ₁and θ ₁correspond to the amplitude of power frequency and the phasing degree relative to sampling time window.

3. a kind of small current neutral grounding single phase earth fault detection method as claimed in claim 1, is characterized in that: in described step (1), determines that the size of time window and the method in sampling period are:

Time window, refer to CPU single treatment from first sample to last sample time span, the size T of time window _wequal the function natural period, sampling period T is less than the function natural period;

Determine the size T of time window _w, and between sampling period T after, the sample number N defining method in a sample window is: N and T _wand the pass of T is:

$N=\frac{T_w}{T}---\left(2\right).$

4. a kind of small current neutral grounding single phase earth fault detection method as claimed in claim 1, is characterized in that: in described step (1), and the concrete grammar calculating least square filter is:

By the U of voltage signal ₀e ^{-λ t}launch by Taylor's radix and get first two, U _ksin (k ω t+ θ _k) launch by trigonometric function, then formula (1) can be expressed as form:

$u\left(t\right)=U_0-U_0\mathrm{\λ}t+\underset{k=1}{\overset{M}{\Σ}}sin\left(k\mathrm{\ω}t\right)\×U_{k}cos\left({\mathrm{\θ}}_k\right)+\underset{k=1}{\overset{M}{\Σ}}cos\left(k\mathrm{\ω}t\right)\×U_{k}sin\left({\mathrm{\θ}}_k\right)---\left(3\right)$

This function is at sampling instant t _i, with its sample u(t _i) between inevitable equal, equation of namely sampling is:

$U_0-U_0{\mathrm{\λt}}_i+\underset{k=1}{\overset{M}{\Σ}}sin\left({\mathrm{k\ωt}}_i\right)\×U_{k}cos\left({\mathrm{\θ}}_k\right)+\underset{k=1}{\overset{M}{\Σ}}cos\left({\mathrm{k\ωt}}_i\right)\×U_{k}sin\left({\mathrm{\θ}}_k\right)=u\left(t_i\right)---\left(4\right)$

T _i-the i-th sampling instant;

Due to sampling instant t _iknown, therefore, sin (k ω t in formula (4) _i), cos (k ω t _i) for being known quantity, N number of sampling equation is listed one by one by formula (4), and obtain a matrix equation as follows, by t _irepresent with τ:

$\left(\begin{array}{ccccccc}1& \mathrm{\τ}& \sin \left(\mathrm{\ω}\mathrm{\τ}\right)& \cos \left(\mathrm{\ω}\mathrm{\τ}\right)& \mathrm{...}& \sin \left(M\mathrm{\ω}\mathrm{\τ}\right)& \cos \left(M\mathrm{\ω}\mathrm{\τ}\right)\\ 1& 2\mathrm{\τ}& \sin \left(\mathrm{\ω}2\mathrm{\τ}\right)& \cos \left(\mathrm{\ω}2\mathrm{\τ}\right)& \mathrm{...}& \sin \left(M\mathrm{\ω}2\mathrm{\τ}\right)& \cos \left(M\mathrm{\ω}2\mathrm{\τ}\right)\\ & .& .& .& \mathrm{...}& .& .\\ 1& N\mathrm{\τ}& \sin \left(\mathrm{\ω}N\mathrm{\τ}\right)& \cos \left(\mathrm{\ω}N\mathrm{\τ}\right)& \mathrm{...}& \sin \left(M\mathrm{\ω}N\mathrm{\τ}\right)& \cos \left(M\mathrm{\ω}N\mathrm{\τ}\right)\end{array}\right)\left(\begin{array}{c}U0\\ -\mathrm{\λ}U0\\ U_1\cos \left({\mathrm{\θ}}_1\right)\\ U_1\sin \left({\mathrm{\θ}}_1\right)\\ .\\ U_M\cos \left({\mathrm{\θ}}_M\right)\\ U_M\sin \left({\mathrm{\θ}}_M\right)\end{array}\right)=\left(\begin{array}{c}u\left(1\right)\\ u\left(2\right)\\ .\\ u\left(N\right)\end{array}\right)---\left(5\right)$

By matrix of coefficients A, known variables matrix X, sampling matrix U represents, then matrix equation (5) can be expressed as follows again:

A·X＝U(6)

Therefore, A is that a N is capable, 2(M+1) arrange determine parameter matrix, X is a 2(M+1) matrix of variables that arranges of row 1, U be capable 1 row of N determine parameter matrix, make 2 (M+1)=N, A is with regard to a square formation, and the highest identifiable design harmonic multiples M=N/2-1, meet Shen agriculture sampling thheorem;

According to matrix theory, there is inverse matrix A in coefficient matrices A ^-1condition be: | A| ≠ 0, and A*A ^-1=I, I are unit matrix, and namely diagonal entry is 1;

Therefore, matrix of unknown parameters to be measured is:

X＝A ^-1·U(7)

A ^-1be least square filter.

5. a kind of small current neutral grounding single phase earth fault detection method as claimed in claim 1, is characterized in that: in described step (2), concrete grammar is, and: matrix of variables X is,

$X=\left(\begin{array}{c}U0\\ -\mathrm{\λ}U0\\ U_1\cos \left({\mathrm{\θ}}_1\right)\\ U_1\sin \left({\mathrm{\θ}}_1\right)\\ .\\ U_M\cos \left({\mathrm{\θ}}_M\right)\\ U_M\sin \left({\mathrm{\θ}}_M\right)\end{array}\right),$

Use x _ithe element of i-th row of representing matrix X, then: x ₁=U ₀, i.e. DC component, and

x ₃＝U ₁cosθ ₁,x ₄＝U ₁sinθ ₁（8）

X ₃be exactly the X-axial projection of voltage or current phasor, x ₄for Y-axial projection of shaking, therefore, the peak-to-peak value of voltage, electric current is:

$U_1=\sqrt{{x_3}^2+{x_4}^2}---\left(9\right)$

The effective value of voltage or electric current is:

6. a kind of small current neutral grounding single phase earth fault detection method as claimed in claim 1, is characterized in that: in described step (2), relative to the function phase angle of time window starting point is:

${\mathrm{\θ}}_1={\tan }^{-1}\frac{U_1{\mathrm{sin\θ}}_1}{U_1{\mathrm{cos\θ}}_1}={\tan }^{-1}\frac{x_4}{x_3}---\left(10\right)$

When distribution line normally works, voltage-phase is ahead of electric current, and phase differential is therebetween:

Δθ ₀＝θ _0u-θ _0i（11）

Wherein, θ _0uand θ _0ibe the voltage-phase and electric current that obtain according to formula (10) respectively, when there is single-phase earthing, phase differential is therebetween:

Δθ ₁＝θ _1u-θ _1i（12）

Therefore, the phase place produced during fault is changed to:

Δθ＝Δθ ₁-Δθ ₀（13）。

7. a kind of small current neutral grounding single phase earth fault detection method as claimed in claim 1, is characterized in that: in described step (3), and the process calculating complex impedance size is specially:

The complex representation form of voltage and current phasor is respectively:

U＝U _x+i·U _y（14）

I＝I _x+i·I _y（15）

Wherein, U _xand I _xbe the x in formula (8) ₃, wherein, U _yand I _ybe the x in formula (8) ₄numerical value;

Complex impedance is:

$Z=\frac{U}{I}=\frac{U_x+i\·U_y}{I_x+i\·I_y}=x+i\·y---\left(16\right)$

The size of complex impedance is:

$\left|Z\right|=\sqrt{x^2+y^2}---\left(17\right).$

8. a kind of small current neutral grounding single phase earth fault detection method as claimed in claim 1, it is characterized in that: in described step (4), the determination methods that low current grounding is reported to the police is, when the Voltage-current phase difference in former and later two sampling time windows and impedance variation all exceed respective threshold value, then send earth fault to report to the police, that is:

If | Z| > Z ₀, and | Δ θ | > θ _s, then report to the police;

Wherein, Z ₀and θ _sbe respectively impedance settings and phasing degree setting value.

9. the Small Electric Current Earthing And Routing Device based on the method such as according to any one of claim 1-8, comprise route protection unit and controller, wherein, route protection unit comprises current potential conversion module and MCU, wherein, the input end of route protection unit connects voltage current transformer, AC signal is changed into MCU by current potential conversion module can sampled signal, controller comprises analog-to-digital conversion module, sampled signal can convert digital signal to, MCU calculates least square filter and voltage, the phase differential of electric current, be transferred to controller, controller calculates complex impedance, judge whether low current grounding occurs.

10. the Little Current Ground Connection System based on the method such as according to any one of claim 1-8, comprise and be multiplely distributed in Small Electric Current Earthing And Routing Device as claimed in claim 9 on distribution network transformer substation circuit and main frame, main frame connects each Small Electric Current Earthing And Routing Device, and whether monitoring distribution line breaks down in real time.