CN101533061B - Large power transmission network fault locating method based on sparse PMU configuration - Google Patents

Abstract

The invention discloses a large power transmission network fault locating method based on sparse PMU configuration. Compared with the existing power transmission line fault location method, the locating method collects fault voltage information and establishes a fault matching index relevant to fault location by only using a little PMU configurated by the power system; so that, the locating method initially judges suspected fault line of the large power system and further realizes precise fault locating by the traversal search method. The large power transmission network fault locating method based on sparse PMU configuration provided by the invention can greatly lower the configuration requirements of the locating algorithm for PMU on the premise of not reducing locating precision, prevents algorithm from being affected by saturation, transition resistance, fault type and the like of a current transformer (CT), improves practicability and commonality of the fault locating algorithm, effectively and quickly supports secure state assessment and maintenance of the power system, and satisfies the requirement of analysis, operation and dispatching of the modern power system.

Description

Large-scale power transmission network Fault Locating Method based on sparse PMU configuration

Technical field

The invention belongs to operation, analysis and the dispatching technique field of electric system, particularly a kind of large-scale power transmission network Fault Locating Method based on sparse PMU configuration.

Background technology

In recent years, along with the development of China's ultra-high-tension power transmission line and the big interconnected implementation of electrical network, quicker, more practical requirement has been proposed for the Fault Locating Method of transmission line of electricity.Up to now, scholar both domestic and external has proposed a large amount of principles and method in the localization of fault field.Present Fault Locating Method can be divided into single-ended [1-3] and both-end [4-6] (or multiterminal [7-8]) localization method according to the quantity of information that adopts.The one-end fault location is exactly that actual measurement voltage, the current information that only utilizes circuit one end determined fault distance.One-end fault ranging do not need channel transfer to side data, be not subjected to the restriction of mechanics of communication condition, the offside system impedance changes and the influence of excessive resistance but be difficult to overcome on principle.And can realize accurate localization of fault on the electric parameters principle of employing line double-end (multiterminal), just need the information transmission of both-end (or multiterminal).Along with the development of automatic technology, the continuous progress of the information transmission technology; Especially in recent years, the development and the exploitation of synchronous phasor measuring device (PMU) make both-end (or multiterminal) fault localization become possibility.Therefore just more and more obtain people's attention based on both-end (or multiterminal) range finding of PMU.

Though at present can realize that based on the research [4-8] of PMU accurate transmission line malfunction locatees, will be faced with following problem in actual applications: (1) existing fault localization method fault location position exactly depends on the configuration quantity of PMU.Have only when the two ends of every transmission line of electricity all dispose PMU or the bus bar configuration in interval, can realize localization of fault [9] effectively.But in the The Study on Fault of actual electric network, this configuration requirement many times can't be met [10].Be subjected to expense restriction, electrical network still can not dispose PMU so to high-density.Only could guarantee the existence of PMU at present in the transformer station of big power station or key.(2) the existing fault localization method mostly needs electric current phasor to participate in calculating.Because when fault took place, current transformer (TA) very easily reached capacity, and will cause sample waveform to distort, and then can not correctly reflect the true fault electric current, cause very big error [11] therefore can for localization of fault based on electric current phasor.

At the requirement of present Fault Locating Method practicality, therefore need a kind of method of only utilizing a small amount of PMU to realize large-scale power transmission network localization of fault, and can effectively avoid the influence of current transformer (TA).

Pertinent literature:

[1] Chen Zheng, Dong Xinzhou, Luo Chengmu, the robustness of single-ended power frequency electric parameters fault localization algorithm, Tsing-Hua University's journal (natural science edition), 2003,43 (3:310-3131

[2]Tamer?Kawady，Jurgen?Stenzel.A?practical?fault?location?approach?for?doublecircuit?transmission?lines?using?single?end?data，IEEE?Transactions?on?Power?Delivery，2003，18(4)：1166-1173.

[3] Suo Nanjiale, Wang Shugang, Zhang Chao utilizes the accurate localization of fault research of double-circuit line on same pole of one-terminal current, Proceedings of the CSEE, 2005,25 (23): 25-30.

[4]Joe-Air?Jiang，Jun-Zhe?Yang，Ying-Hong?Lin，An?Adaptive?PMU?Based?FaultDetection/Location?Technique?for?Transmission?Lines?Part?I：Theory?and?Algorithms，IEEE?Transactions?on?Power?Delivery.2000，15(2)：486-493

[5]FAN?Chunju?DU?Xiuhua?LI?Shengfang.An?Adaptive?Fault?Location?TechniqueBased?on?PMU?for?Transmission?Line，IEEE?Transactions?on?Power?Delivery.2007，15(4)：1-6

[6] Suo Nanjiale, Zhang Yining, Qi Jun is based on the time domain method Two-terminal Fault Location principle of parameter recognition, electric power network technique, 2006,30 (8): 65-70.

[7]J.V.H.Sanderson，R.G.R.Santana，and?B.Al-Fakri，“Improved?directionalcomparison?based?algorithm?for?protection?of?multi-terminal?transmission?lines，”inProc.5th?Int.Conf.Developments?Power?Syst.Protection，1993，vol.368，pp.153-156.

[8]S.M.Brahma，“Fault?location?scheme?for?a?multi-terminal?transmission?lineusing?synchronized?voltage?measurements，”IEEE?Trans.Power?Del.，vol.20，no.2，pp.1325-1331，Apr.2005.

[9]LIEN?K.P.，LIU?C.W.，YU?C.S..Transmission?Network?Fault?LocationObservability?With?Minimal?PMU?Placement[J].IEEE?Transaction?on?PowerDelivery.2006.21(3)：1128-1137.

[10]Yuan?Liao.Fault?Location?for?Single-Circuit?Line?Based?on?Bus-ImpedanceMatrix?Utilizing?Voltage?Measurements[J].IEEE?transactions?on?power?delivery，2008，23(2)：609-617.

[11] Chen Zheng, Dong Xinzhou, Luo Chengmu. the saturated a kind of solution that influences distance accuracy of current transformer. Automation of Electric Systems [J], 2002,26 (1): 39-4

Summary of the invention

The objective of the invention is in order to overcome the shortcoming that traditional Fault Locating Method relies on to the strict demand of electrical network PMU configuration, to electric current phasor, be different from existing power transmission line fault locating method, a kind of large-scale power transmission network Fault Locating Method based on sparse PMU configuration is provided.

Large-scale power transmission network Fault Locating Method based on sparse PMU configuration comprises the steps:

1) presses the PMU configuration requirement at power transmission network respective nodes configuration PMU;

2) after large-scale power grid is short-circuited fault, the partial electric grid of action is taken place as the fault zone in relay protection device;

3) at this electric network fault zone, utilize its nodal impedance matrix,, generate abort situation coupling index in conjunction with the voltage phasor of fault collection; And then be objective function with this fault coupling index, independent variable is all circuits in electric network fault zone, constitutes minimum value and finds the solution problem;

4) if this fault zone is the node number greater than 10 electric network composition, obtain the fault matching value of these all nodes of zone, choose the node of calculated value minimum, the circuit that order links to each other with these nodes is suspicious faulty line collection; If this fault zone is the node number less than 10 electric network composition, all circuits in this zone can be directly as the suspicious circuit of fault;

5) use the traversal search method to the suspicious circuit of all faults in the step 4), travel through, calculate the fault matching value of all traversal points, thereby find the line fault position at minimum value place with the step-length of 0.01% line length.

Described PMU configuration requirement is:

1) guarantees that any circuit both sides bus all can be communicated to certain PMU after this circuit is removed;

2) PMU is evenly distributed on each zone of electrical network.

Described generation abort situation coupling index step is:

1), known at node D at the electric network fault zone of n node _i(i=1.2...m) locate to dispose PMU, the back takes place fault gathers each node D _iVoltage phasor, and then obtain additional positive sequence voltage phasor

2) assumed fault occurs in any circuit i-j, and is xL apart from node i _IjKm place, wherein L _IjBe the length of any circuit i-j, and 0≤x≤1; Make this abort situation be new node and the numbering (n+1), obtain nodal impedance matrix Z about abort situation x;

3) by step 1) and step 2) obtain fault as follows coupling index:

$\mathrm{\δ}\left(x\right)=\sqrt{\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{(K_{D_i}-\stackrel{\‾}{K_{D_i}})}^2/m},$ $K_{D_i}=\mathrm{\Δ}{V}_{D_i}/Z_{D_i(n+1)},$ $\stackrel{\‾}{K_{D_i}}=\underset{i=1}{\overset{m}{\mathrm{\Σ}}}{K}_{D_i}/m$

Wherein

Be row D among the impedance matrix Z _i, row n+1 element.

The present invention compares the beneficial effect that has with existing existing technology:

1) be applicable to the large-scale power transmission network of arbitrary structures or node number, algorithm has versatility.And this method is practical simple, need not to find the solution complicated nonlinear equation, need not to know faulty line, fault type, transition resistance etc. in advance to have good application prospects;

2) only need a little P MU configuration, utilize the accurate localization of fault of fault coupling index realization the fault zone electrical network.Break through the strict demand of traditional algorithm, have engineering significance PMU;

3) only utilize voltage phasor to carry out localization of fault, avoided having improved the practicality of fault location algorithm because of the saturated positioning error that causes of current transformer.

Description of drawings

Fig. 1 is based on the large-scale power transmission network Fault Locating Method of sparse PMU configuration;

Fig. 2 is the additional positive sequence network synoptic diagram of any fault zone electrical network;

Fig. 3 is the fault zone electrical network synoptic diagram of embodiment 1;

Fig. 4 is the fault zone power network line of embodiment 1 and the relation curve of matching value;

Fig. 5 is the fault zone electrical network synoptic diagram of embodiment 2

Fig. 6 is the fault zone grid nodes of embodiment 2 and the relation curve of matching value;

Fig. 7 is the fault zone power network line of embodiment 2 and the relation curve of matching value.

Embodiment

Below in conjunction with accompanying drawing, embodiments of the invention are elaborated, the process flow diagram of this invention is as shown in Figure 1.

Embodiment 1:

Consider any large-scale power transmission network, adopt one embodiment of the present invention that its short trouble is positioned, each step division is as follows:

Step 1: considerable for guaranteeing all line faults of power transmission network, dispose PMU by following PMU configuration requirement in the network respective nodes:

(1) guarantees that any circuit both sides bus all can be communicated to certain PMU after this circuit is removed.

(2) make PMU be evenly distributed on each zone of electrical network.

Step 2: after large-scale power grid is short-circuited fault,, relay protection device the partial electric grid of action can be taken place as preliminary fault zone thus because the action of relay protection will concentrate on the faulty line near zone substantially.

Step 3: at this electric network fault zone, utilize its nodal impedance matrix,, generate abort situation coupling index in conjunction with the voltage phasor of fault collection; And then be objective function with this fault coupling index, independent variable is all circuits in electric network fault zone, constitutes minimum value and finds the solution problem.

It is xL apart from bus i that assumed fault betides transmission line i-j _IjKm place, wherein xL _IjBe the length of circuit i-j arbitrarily, and 0＜x＜1, can get additional positive sequence network figure as shown in Figure 2, among the figure only the trouble spot injection current source is arranged

, wherein transmission line adopts the π model.

There is n rank positive sequence impedance matrix Z during this fault zone electrical network normal condition ₀, the trouble spot is (n+1) node if make, and then can set up (n+1) rank positive sequence bus admittance matrix Z (process slightly) to this network.By derivation as can be known, matrix Z is the function about fault unknown number x.

And then additional positive sequence voltage phasor of node and current in the fault point source can set up fault the time

Relation:

${\left[\begin{array}{cccccccc}\mathrm{\&Delta;}{\stackrel{.}{V}}_1& ...& \mathrm{\&Delta;}{\stackrel{.}{V}}_i& ...& \mathrm{\&Delta;}{\stackrel{.}{V}}_j& ...& \mathrm{\&Delta;}{\stackrel{.}{V}}_n& \mathrm{\&Delta;}{\stackrel{.}{V}}_{n+1}\end{array}\right]}^T=\mathrm{<figure-callout\; id="0"\; label="Z"\; filenames="H2009100973344E00022.png,H2009100973344E00031.png"\; state="\{\{state\}\}">Z</figure-callout>}{\left[\begin{array}{cccccccc}0& 0& 0& 0& 0& 0& 0& {\stackrel{.}{I}}_f\end{array}\right]}^T---\left(1\right)$

Wherein

(p=1,2...n n+1) are the additional positive sequence voltage phasor of node p, Can decompose by the voltage phasor change amount before and after the fault and obtain:

$\mathrm{\&Delta;}{\stackrel{.}{V}}_p=\mathrm{\&Delta;}{\stackrel{.}{V}}_p^{\mathrm{Phasor}}\&times;M---\left(2\right)$

Wherein

Be the voltage phasor change amount before and after the fault, M is a positive sequence Symmetric Decomposition matrix.

By formula (1) as can be known in this network the voltage of arbitrary node p can be expressed as:

$\mathrm{\&Delta;}{\stackrel{.}{V}}_p=Z_{p(n+1)}*{\stackrel{.}{I}}_f---\left(3\right)$

Z wherein _{P (n+1)}Element for row p, row n+1 among the impedance matrix Z.

And then as can be known, the fault injection current

Can calculate by following formula:

${\stackrel{.}{I}}_f=\mathrm{\&Delta;}{\stackrel{.}{V}}_p/Z_{p(n+1)}---\left(4\right)$

Found by formula (4): the trouble spot injection current can be obtained by additional positive sequence voltage phasor and its associated transitions impedance computation of arbitrary node in the network.If there is PMU in (may away from faulty line) hypothesis node k and l, thereby can obtain:

$\mathrm{\&Delta;}{\stackrel{.}{V}}_k/Z_{k(n+1)}=\mathrm{\&Delta;}{\stackrel{.}{V}}_l/Z_{l(n+1)}---\left(5\right)$

In theory, formula (5) is the equation about unknown number x, can solve position of failure point thus.Also promptly just can realize localization of fault at the two node configuration PMU of place for minimum needs of certain regional power grid.But for certain bar circuit, may have formula (5) is permanent situation about setting up, and promptly can't realize the effective location to this circuit.Therefore, for realizing this method considerable fully to the whole network, need the PMU of configuration some, thereby guarantee that all there is effective equation in every circuit.PMU configuration requirement in the step 1 has been proposed for this reason.

And then, if the node configuration PMU of m place in the electrical network zone, then for node D _i(i=1.2...m) exist

$K_{D_I}=\mathrm{\&Delta;}{V}_{D_i}/Z_{D_i(n+1)}---\left(6\right)$

Can push away according to formula (5), only have a following relation in the trouble spot:

${\mathrm{<figure-callout\; id="1"\; label="K\; D"\; filenames="H2009100973344E00011.png,H2009100973344E00021.png"\; state="\{\{state\}\}">K</figure-callout>}}_{D_{}}={\mathrm{<figure-callout\; id="2"\; label="K\; D"\; filenames="H2009100973344E00021.png,H2009100973344E00023.png"\; state="\{\{state\}\}">K</figure-callout>}}_{D_{}}=...=K_{D_m}---\left(7\right)$

Formula (7) is the Nonlinear System of Equations about abort situation x, directly finds the solution very difficulty and at first failure judgement place circuit.For this reason, definition is mated index based on formula (7) failure definition:

$\mathrm{\&delta;}\left(x\right)=\sqrt{\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}{(K_{D_i}-\stackrel{\&OverBar;}{K_{D_i}})}^2/m}$ Wherein, $\stackrel{\&OverBar;}{K_{D_i}}=\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}{K}_{D_i}/m---\left(8\right)$

When and during only in the trouble spot, formula (7) could be mated fully, i.e. δ (x)=0, so fault-location problem is converted into and makes coupling index (8) obtain the optimization problem of minimum value, independent variable is a transmission line of electricity all in the electrical network zone, and for every circuit 0≤x≤1.

Step 4: if this fault zone is the node number greater than 10 electric network composition, obtain the fault matching value of these all nodes of zone, choose the node of calculated value minimum (or near minimum value), the circuit that order links to each other with these nodes is suspicious faulty line collection, further dwindles the fault zone thus; If this fault zone is the node number less than 10 electric network composition, can directly skip this step, all circuits in this zone can be directly as the suspicious circuit of fault;

Step 5: use the traversal search method to the suspicious circuit of all faults in the step 4), travel through, calculate the fault matching value of all traversal points, thereby find the line fault position at minimum value place with the step-length of 0.01% line length.

After supposing that a certain large-scale power transmission network breaks down, through the definite fault zone electrical network as shown in Figure 1 of relay protection action message, only bus 1,2,3 places dispose PMU, and only gather the false voltage phasor.

For the specific implementation process of this method is described, suppose to take place AB phase earth fault (ABG), transition resistance is 5 Ω, fault is positioned at circuit 2-7 apart from node 2 the line is busy road length overall 53% place.Because this fault zone (the node number is less than 10) simple in structure, all circuits all can be used as the suspicious circuit of fault, adopt the traversal search method can obtain abort situation and the relation curve that mates index, as shown in Figure 4.As shown in Figure 4, apart from bus 2 about 53.04% places, matching value is obtained minimum value at circuit 2-7, and positioning error is 0.04%.

For verification algorithm is applicable to different failure conditions, be example on the circuit various short troubles to take place, carried out a large amount of emulation for different transition resistances (not waiting), fault type and fault distance situation from 1～1K, test result sees Table 1.The result shows that this algorithm all can obtain high orientation precision under different condition.This algorithm is subjected to the influence of transition resistance, fault type and fault distance less; When the transition resistance variations, bearing accuracy changes also very little; Even when transition resistance was big, algorithm still can obtain high orientation precision.

Simulation result under table 1 different faults type, the transition resistance

Faulty line	Abort situation	Fault type	Transition resistance (Ω)	The % positioning error
					Circuit 2-7	35% from bus 2	ABG	5	0.04％
Circuit 7-8	43% from bus 7	ABCG	1	0.09％
					Circuit 8-9	55% from bus 8	BCG	10	0.02％
Circuit 9-3	27% from bus 9	ABS	50	0.19％
					Circuit 7-5	50% from bus 7	AG	100	0.29％
Circuit 5-4	63% from bus 5	CG	1K	0.40％
					Circuit 4-1	70% from bus 4	BG	50	0.50％

Circuit 4-6	50% from bus 4	BCG	50	0.05％
					Circuit 7-9	50% from bus 7	AB	100	0.22％

By embodiment 1 as can be known, the inventive method is only utilized a little P MU configuration, can realize the accurate location of any short trouble to the fault zone electrical network of this simple structure, and is not subjected to that current transformer is saturated to be influenced, and has practicality and future in engineering applications.

Embodiment 2:

Present embodiment uses the method identical with embodiment 1, suppose that a certain large-scale power transmission network breaks down after, through relay protection action message preliminary judgement fault zone electrical network as shown in Figure 5, and only at node 3,8; 11,14,16,19; 22,27,29,39 places configuration PMU.On transmission line of electricity 21-16 apart from bus 21 the line is busy road length overall 30% place common singlephase earth fault takes place.

Because this fault zone electrical network is complicated (the node number is greater than 10), at first can determine suspicious faulty line collection.The method according to this invention, assumed fault occur in the situation of all associated bus nodes on the transmission line, can obtain the bus position and mate the graph of a relation of index with electricity, as shown in Figure 6.As shown in Figure 6, the coupling desired value at faulty line place bus 16,21 places obtains minimum value, and much smaller than other nodes (deviate all is expressed as 5 greater than 5 node among the figure).Last figure shows the bus near more from the trouble spot, and the coupling index is also more little, can carry out preliminary selecting to possible faulty line according to this principle.According to the fault Precision Orientation Algorithm,, can obtain position of failure point and the relation curve that mates index, as shown in Figure 7 then in the suspicious sets of lines of roughly possible fault (circuit 16-15,17-16,21-16,24-16,22-21) traversal.As shown in Figure 7, apart from bus 21 about 29.90% places, the coupling index obtains minimum value at circuit 16-21, and positioning error is 0.1%.

In like manner embodiment 1, and table 2 has been verified the accuracy of this method under various different faults situations.

Simulation result under table 2 different faults type, the transition resistance

Faulty line	Abort situation	The matching value minimum node	The suspicious sets of lines of fault	The % positioning error
					Circuit 14-4	43% from node 14	14，4	3-4，4-5，14-15，14-21，14-4	0.24％
Circuit 5-4	67% from node 5	4，5，6	4-3，4-5，4-14，5-6，5-8，6-11，6-31	0.39％
					Circuit 9-39	23% from node 9	8，9，39	8-5，8-7，8-9，9-39，39-1	0.12％
Circuit 26-28	50% from node 26	26.28	26-25，26-27，26-29，28-26，28-29	0.53％
					Circuit 22-23	75% from node 22	22	22-23，21-22	0.29％
Circuit 25-26	50% from node 25	2，25，26	2-1，2-3，2-25，25-26，26-27，26-28，26-29	0.40％

By embodiment 2 as can be known, the inventive method can only be utilized a little P MU configuration equally, and the accurate location to this baroque fault zone electrical network is realized any short trouble proves its versatility.

Claims (1)

1. large-scale power transmission network Fault Locating Method based on sparse PMU configuration, it comprises the steps:

1) presses the PMU configuration requirement at power transmission network respective nodes configuration PMU;

2) after large-scale power grid is short-circuited fault, the partial electric grid of action is taken place as the fault zone in relay protection device;

3) at this electric network fault zone, utilize its nodal impedance matrix,, generate abort situation coupling index in conjunction with the voltage phasor of fault collection; And then be objective function with this fault coupling index, independent variable is all circuits in electric network fault zone, constitutes minimum value and finds the solution problem;

4) if this fault zone is the node number greater than 10 electric network composition, obtain the fault matching value of these all nodes of zone, choose the node of calculated value minimum, the circuit that order links to each other with these nodes is suspicious faulty line collection; If this fault zone is the node number less than 10 electric network composition, all circuits in this zone can be directly as the suspicious circuit of fault;

5) use the traversal search method to the suspicious circuit of all faults in the step 4), travel through, calculate the fault matching value of all traversal points, thereby find the line fault position at minimum value place with the step-length of 0.01% line length;

Described PMU configuration requirement is:

1) guarantees that any circuit both sides bus all can be communicated to certain PMU after this circuit is removed;

2) PMU is evenly distributed on each zone of electrical network;

Described generation abort situation coupling index step is:

1), known at node D at the electric network fault zone of n node _i(i=1.2...m) locate to dispose PMU, the back takes place fault gathers each node D _iVoltage phasor, and then obtain additional positive sequence voltage phasor

N represents the number of nodes in the electrical network, and m represents the quantity of the node of configuration PMU in the electrical network;

2) assumed fault occurs in any circuit i-j, and is xL apart from node i _IjKm place, wherein L _IjBe the length of any circuit i-j, and 0≤x≤1; Make this abort situation be new node and the numbering (n+1), obtain nodal impedance matrix Z about abort situation x;

3) by step 1) and step 2) obtain fault as follows coupling index:

$\mathrm{\&delta;}\left(x\right)=\sqrt{\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}{(K_{D_i}-\stackrel{\&OverBar;}{K_{D_i}})}^2/m},$ $K_{D_i}=\mathrm{\&Delta;}{V}_{D_i}/Z_{D_i(n+1)},$ $\stackrel{\&OverBar;}{K_{D_i}}=\underset{i=1}{\overset{m}{\mathrm{\&Sigma;}}}{K}_{D_i}/m$

Wherein Be row D among the impedance matrix Z _i, row n+1 element.

Images