CN105606931A - Quantum-genetic-algorithm-based fault diagnosis method for medium-voltage distribution network - Google Patents

Abstract

The invention relates to a quantum-genetic-algorithm-based fault diagnosis method for a medium-voltage distribution network. The method is characterized in that the method comprises the following steps that: step one, an improved medium-voltage distribution network fault diagnosis model is established by employing actual and expected values of element movement and combining the circuit breaker failure protection and the protection state of automatic opening and closing of the circuit breaker; and step two, a quantum-genetic-algorithm-based improved medium-voltage distribution network fault diagnosis model is solved to carryout fault diagnosis on the medium-voltage distribution network. According to the invention, the distribution network fault diagnosis model is established by analyzing the element type and the protection configuration situation of the distribution system; protection and the movement situation of and circuit breaker after a fault scene are simulated; and then the fault model is solved by using the quantum genetic algorithm. Therefore, functions that a fault element can be localized accurately after fault occurrence at a distribution network and the fault of the distribution network can be diagnosed rapidly and comprehensively can be realized.

Description

A kind of medium voltage distribution network method for diagnosing faults based on quantum genetic algorithm

Technical field

The present invention relates to urban power distribution network planning and fault diagnosis technology field, particularly a kind of based on amountThe medium voltage distribution network method for diagnosing faults of sub-genetic algorithm.

Background technology

Distribution system, as the important step of electrical energy production, transmission and use, is the user's request that intergrates with practiceSide with send out, the crucial tie of transmission system. Therefore, how after power distribution network breaks down, fault to be carried outIt is the key component of current urban distribution network assessment development that location effectively and reasonably is also diagnosed. ConsiderIn power distribution network, the residing adverse circumstances of communication information device and different regions power distribution automation development journeyThe reason such as inconsistent of degree has caused in the fault message of power distribution network and has existed a large amount of uncertain factors; AndOnce there is multiple complex fault in power distribution network, will have fault element and non-event in dead electricity regionBarrier element, considers the huge of distribution system scale, and number of elements and of a great variety, was difficult in the short timeWhether inside determine element in malfunction, in addition the line protective devices in power distribution network or breaker meetingThere is the situation of tripping or malfunction, therefore, the model that above-mentioned these factors all can cause distribution network failure to be analyzedEnclose expansion, fault message in the process of uploading relevant departments, occur distortion etc. and cause accurately determiningFault element, has brought harm to the safe and stable operation of distribution system. And along with distribution system wiring shapeIncreasingly sophisticated, the equipment component of formula is on the increase, equipment scale constantly increases, and user's request side pairRequirement in power supply progressively improves. Therefore, effectively and reasonably to distribution system carry out fault diagnosis forThe equal tool of reliable electricity consumption of the comprehensive development of power system and Demand-side is of great significance.

At present, more for the method for diagnosing faults of power system, main thought is all by power systemThe action message of middle switch element is carried out fault judgement and is analyzed. At present relevant power system failure diagnosticMethod mainly contains: (1) Fault Diagnosis of Distribution Network algorithm based on rough set and decision tree is mainly to utilizeRough set there is the ability of better processing uncertain information, realized the fault to fault sample decision tableRule is asked for; (2) the Fault Diagnosis for HV Circuit Breakers method based on probabilistic neural network (PNN), effectivelyThe fault characteristic of analyzing primary cut-out, carries out fault location; (3) particle cluster algorithm and neutral net phaseIn conjunction with analog-circuit fault diagnosis method, fault-signal is effectively decomposed, then by normalization placeReason is extracted fault characteristic information and input learning sample as neutral net with this; (4) based on sequential mouldThe method for diagnosing faults of sticking with paste Petri net, by setting up fault diagnosis model, completes relay protection is movedEvaluation. The fault diagnosis that above method is power system provides good Research Thinking, but still existsBelow limitation: the first, fail to consider protection or switch failure, malfunction and letter in distribution network failure situationDiagnostic accuracy when breath distortion; The second, the timing of power system element action under failure condition does not haveHave and taken into full account.

Compared to above-mentioned several method, quantum genetic algorithm (QuantumGeneticAlgorithm, QGA)Quantum theory is dissolved in the middle of classical genetic algorithm effectively, is there is hunting zone than traditional genetic algorithmWider, the advantage such as the search efficiency of global optimizing is higher, and adaptability is stronger, and can ensure algorithmConvergence.

Summary of the invention

The object of the invention is to overcome the deficiencies in the prior art, a kind of reasonable in design, fault diagnosis is providedComprehensively, analyze accurately and locate the medium voltage distribution network method for diagnosing faults based on quantum genetic algorithm fast.

The present invention solves its technical problem and takes following technical scheme to realize:

A medium voltage distribution network method for diagnosing faults based on quantum genetic algorithm, comprises the following steps:

Reality and the desired value of step 1, the action of employing element also incorporates breaker fail protection and breakerThe guard mode of automatic reclosing is set up modified medium voltage distribution network fault diagnosis model; The target of this modelFunction is:

$\begin{array}{c}E\left(x\right)=\underset{k=1}{\overset{n}{\Σ}}|r_{k,m}-r_{k,m}^{*}||1-r_{k,s}{r}_{k,s}^{*}-\stackrel{\⊕}{\Σ}{r}_{k,l}{r}_{k,l}^{*}|+\underset{k=1}{\overset{n}{\Σ}}|r_{k,s}-r_{k,s}^{*}||1-\stackrel{\⊕}{\Σ}{r}_{k,l}{r}_{k,l}^{*}|\\ +\underset{k=1}{\overset{n}{\Σ}}|r_{k,l}-r_{k,l}^{*}|+\underset{i=1}{\overset{q}{\Σ}}|r_{i,cb}-r_{i,cb}^{*}|+\underset{i=1}{\overset{q}{\Σ}}|r_{i,auto}-r_{i,auto}^{*}|\\ +\underset{i=1}{\overset{q}{\Σ}}|C_i-C_i^{*}||1-r_{i,cb}{r}_{i,cb}^{*}-r_{i,auto}{r}_{i,auto}^{*}|\end{array}---\left(2\right)$

In above-mentioned expression formula, r_k,mAnd r_k,m ^*Represent respectively the virtual condition of each element main protection and expect shapeState; r_k,sAnd r_k,s ^*Represent respectively virtual condition and the expectation state of the nearly back-up protection of discrete component; r_k,lAnd r_k,l ^*Represent respectively virtual condition and the expectation state of discrete component back-up protection far away; C_iAnd C_i ^*Represent respectively to open circuitThe virtual condition of device and expectation state;Represent to connect exclusive disjunction;_ri,cbWith_ri,cb ^*Represent respectively breaker failureVirtual condition and the expectation state of protection; r_i,autoAnd r_i,auto ^*Represent respectively the actual shape of breaker automatic reclosingState and expectation state.

Step 2, solve the modified medium voltage distribution network fault diagnosis model based on quantum genetic algorithm, centeringBe press-fitted electrical network and carry out fault diagnosis.

And the concrete steps of described step 2 comprise:

(1) adopt traditional genetic algorithm to after element is encoded in intermediate distribution system, according to quantum ratioSpecial coded system is revised, thereby formulates the encoding scheme that is adapted to quantum genetic algorithm in order to representMedium voltage distribution network troubleshooting issue;

(2) solve described modified medium voltage distribution network fault diagnosis model basis according to quantum genetic algorithmFault element in the medium voltage distribution network of result of calculation location is also differentiated protection and the correctness of breaker action,Carry out accident analysis.

And the specific coding method of (1) step of described step 2 is:

Suppose that power supply interrupted district entirety is for individual chromosome q, in distribution system, component population is described chromosome qIn gene number n; Adopt the coded system of the quantum bit in quantum genetic algorithm, use a pair of plural numberDefine a quantum bit position, individual chromosome q adopts quantum bit coding to solve fault diagnosis to askThe concrete form of topic is:

$q^t=\left[\begin{array}{cccc}{\mathrm{\α}}_1^t& {\mathrm{\α}}_2^t& ...& {\mathrm{\α}}_n^t\\ {\mathrm{\β}}_1^t& {\mathrm{\β}}_2^t& ...& {\mathrm{\β}}_n^t\end{array}\right]---\left(4\right)$

In above-mentioned expression formula,WithFor plural form represents the probability amplitude of quantum bit corresponding state; T is for dyingThe algebraically of colour solid.

And, in described step 2 (2) step, solve medium voltage distribution network fault according to quantum genetic algorithm and examineThe computational methods of disconnected model, comprise the steps:

1. centering is press-fitted Power grid structure analysis, specifies component kind and quantity in power distribution network,Determine and in accident analysis, need the element analyzed;

2. dwindle fault diagnosis scope according to the operating state of switch in power distribution network and protection, determine power distribution networkPower supply interrupted district after fault and the element that need to analyze;

3. after occurring according to distribution network failure, the state of each element, switch, protection and breaker, sets upElement state matrix also arranges order according to the modified medium voltage distribution network fault diagnosis model described in step 1Scalar functions;

4. according to step 2 (1) step for representing the amount of being adapted to of medium voltage distribution network troubleshooting issueThe probability amplitude of the encoding scheme setting member of sub-genetic algorithm, and to step 2. determined element carry outAssignment, the random number that produces between numerical value interval [0,1], the element state that 3. itself and step setCompare, if random number is more than or equal to probability amplitude, the measurement result of element gets 1, otherwise gets0；

5. the element measured value that 4. the element state value of 3. step being determined and step are determined is brought step 1 intoIn the object function of modified medium voltage distribution network fault diagnosis model, carry out objective function evaluates, determine targetFunction initial value;

6. set the quantum genetic of population scale, chromosome length, corner step-length and maximum iteration timeThe optimization principles of algorithm and utilize quantum genetic algorithm to step 3. object function be optimized calculating; And willThe initial value of result of calculation and step definite object function in 5. compares, if this result of calculation is less thanOr equal initial value, retain currency as target function value; If be greater than initial value, fresh target moreFunctional value; Carry out algorithm iteration, until optimum results meets precision or reaches iterations simultaneously;

7. analysis result, determines the fault element in power distribution network, carries out fault and studies and judges analysis.

Advantage of the present invention and good effect are:

1, the present invention, by analyzing component kind and protection configuring condition in distribution system, sets up power distribution networkFault diagnosis model, the action situation of protection and breaker after simulated failure sight, utilizes quantum genetic to calculateMethod solves fault model, thereby accurately locates fault element, and differentiates protection and breaker actionCorrectness. Realize that distribution network failure carries out fast, the function of comprehensive diagnostic.

2, the present invention can be from the angle of distribution system grid structure, has taken into full account in power distribution networkComponent information, protection information and breaker information. For component information, the quantity of Main Analysis element,The configuring condition of kind and protection; For protection information, Main Analysis main protection, nearly back-up protection withAnd the configuring condition of back-up protection far away; For breaker information, Main Analysis breaker fail protection andThe configuring condition of automatic reclosing, and then comprehensively, fast power distribution network is carried out to fault diagnosis, make distributionNet can be fixed a breakdown rapidly after fault, ensures the overhaul efficiency of power distribution network. Thereby be existing power distribution networkProviding support of method for diagnosing faults, is conducive to promote urban power distribution network planning and fault diagnosis level,Promote the Rational Development of Construction of Intercity Network structure and trouble hunting means.

Brief description of the drawings

Fig. 1 is fault diagnosis process chart of the present invention;

Fig. 2 is the power distribution network contact structure chart in embodiments of the invention.

Detailed description of the invention

Below in conjunction with accompanying drawing, the embodiment of the present invention is described in further detail:

A medium voltage distribution network method for diagnosing faults based on quantum genetic algorithm, as shown in Figure 1, comprisesFollowing steps:

Reality and the desired value of step 1, the action of employing element also incorporates breaker fail protection and breakerThe guard mode of automatic reclosing is set up modified medium voltage distribution network fault diagnosis model.

According to conventional method, adopt reality and the desired value of element action can set up following medium voltage distribution networkFault diagnosis model, the object function of this model is:

$E\left(x\right)=\underset{k=1}{\overset{n}{\Σ}}|r_{k,m}-r_{k,m}^{*}|+\underset{k=1}{\overset{n}{\Σ}}|r_{k,s}-r_{k,s}^{*}|+\underset{k=1}{\overset{n}{\Σ}}|r_{k,l}-r_{k,l}^{*}|+\underset{i=1}{\overset{q}{\Σ}}|C_i-C_i^{*}|---\left(1\right)$

In above-mentioned expression formula, r_k,mAnd r_k,m ^*Represent respectively the virtual condition of each element main protection and expect shapeState; r_k,sAnd r_k,s ^*Represent respectively virtual condition and the expectation state of the nearly back-up protection of discrete component; r_k,lAnd r_k,l ^*Represent respectively virtual condition and the expectation state of discrete component back-up protection far away; C_iAnd C_i ^*Represent respectively to open circuitThe virtual condition of device and expectation state.

Find by further analysis, in above-mentioned basic model, exist certain problem. First, rightCan not only consider to control the guard mode of its action in the expectation of breaker operating state, should be by opening circuitThe virtual condition of the equipment in device protection domain and relevant protection thereof determines jointly; The second, above-mentioned basisMay there is many solutions problem in model, in actual application, be unfavorable for fault diagnosis fast, accurately enterOK.

In this step 1, consider the problems referred to above, at the order of this medium voltage distribution network fault diagnosis modelOn the basis of scalar functions, further incorporate the protection such as breaker fail protection and breaker automatic reclosingState, sets up modified medium voltage distribution network fault diagnosis model, and the object function of this model is:

$\begin{array}{c}E\left(x\right)=\underset{k=1}{\overset{n}{\Σ}}|r_{k,m}-r_{k,m}^{*}||1-r_{k,s}{r}_{k,s}^{*}-\stackrel{\⊕}{\Σ}{r}_{k,l}{r}_{k,l}^{*}|+\underset{k=1}{\overset{n}{\Σ}}|r_{k,s}-r_{k,s}^{*}||1-\stackrel{\⊕}{\Σ}{r}_{k,l}{r}_{k,l}^{*}|\\ +\underset{k=1}{\overset{n}{\Σ}}|r_{k,l}-r_{k,l}^{*}|+\underset{i=1}{\overset{q}{\Σ}}|r_{i,cb}-r_{i,cb}^{*}|+\underset{i=1}{\overset{q}{\Σ}}|r_{i,auto}-r_{i,auto}^{*}|\\ +\underset{i=1}{\overset{q}{\Σ}}|C_i-C_i^{*}||1-r_{i,cb}{r}_{i,cb}^{*}-r_{i,auto}{r}_{i,auto}^{*}|\end{array}---\left(2\right)$

In above-mentioned expression formula, r_k,mAnd r_k,m ^*Represent respectively the virtual condition of each element main protection and expect shapeState; r_k,sAnd r_k,s ^*Represent respectively virtual condition and the expectation state of the nearly back-up protection of discrete component; r_k,lAnd r_k,l ^*Represent respectively virtual condition and the expectation state of discrete component back-up protection far away; C_iAnd C_i ^*Represent respectively to open circuitThe virtual condition of device and expectation state;Represent to connect exclusive disjunction; r_i,cbAnd r_i,cb ^*Represent respectively breaker failureVirtual condition and the expectation state of protection; r_i,autoAnd r_i,auto ^*Represent respectively the reality of breaker automatic reclosingState and expectation state. For the desired value of related elements protection, in the time of protection action, the value of r* gets 1, noBe 0; For breaker, in the time that breaker should trip, the value of C* gets 1, otherwise is 0; For automaticallyReclosing device, in the time that automatic reclosing should close a floodgate, the value of r* gets 1, otherwise is 0. Fault after improvementDiagnostic model considers that protecting component is more comprehensive, and has considered the protection of relative breaker, makes model moreAdd complete, effective.

Step 2, solve the modified medium voltage distribution network fault diagnosis model based on quantum genetic algorithm, centeringBe press-fitted electrical network and carry out fault diagnosis.

The concrete steps of described step 2 comprise:

(1) adopt traditional genetic algorithm to after element is encoded in intermediate distribution system, pass through introduction volumeThe concept of quantum bit in sub-theory (qubit), revises according to the coded system of quantum bit, fromAnd formulate the encoding scheme that is adapted to quantum genetic algorithm in order to represent medium voltage distribution network troubleshooting issue.

What pay close attention to due to the troubleshooting issue of distribution system is the power supply interrupted district that is subject to fault effects, because ofTotal element number in this supposition power supply interrupted district is n. For traditional genetic algorithm, power supply interrupted district entiretyBe equivalent to chromosome; And element is wherein equivalent to gene, adopt binary system to encode, " 1 " representsThere is fault in this element, " 0 " represents that this element does not break down, and is therefore equivalent to the base in chromosomeBecause number is n.

And in quantum genetic algorithm, used a kind of coded system based on quantum bit, use a pair ofA quantum bit position of plural number definition, a system with m quantum bit position can be described as:

$\left|\begin{array}{c}{\mathrm{\α}}_1\\ {\mathrm{\β}}_1\end{array}\right|\left.\begin{array}{c}{\mathrm{\α}}_2\\ {\mathrm{\β}}_2\end{array}\right|\begin{array}{c}\mathrm{...}\\ \mathrm{...}\end{array}\left|\begin{array}{c}{\mathrm{\α}}_m\\ {\mathrm{\β}}_m\end{array}\right|---\left(3\right)$

Wherein, | α_i|²+|β_i|²＝1(i＝1,2,......,m)。

Quantum bit is a bifurcation quantized system of serving as information memory cell, is to be defined in a two dimensionA unit vector in complex vector space, this space by a pair of specific orthonormal basis | 0 >, | 1 > form.

Therefore,, based on above-mentioned analysis, the specific coding method of (1) step of described step 2 is:

Suppose that power supply interrupted district entirety is for individual chromosome q, in distribution system, component population is described chromosome qIn gene number n; Adopt the coded system of the quantum bit in quantum genetic algorithm, use a pair of plural numberDefine a quantum bit position, individual chromosome q adopts quantum bit coding to solve fault diagnosis to askThe concrete form of topic is:

$q^t=\left[\begin{array}{cccc}{\mathrm{\α}}_1^t& {\mathrm{\α}}_2^t& ...& {\mathrm{\α}}_n^t\\ {\mathrm{\β}}_1^t& {\mathrm{\β}}_2^t& ...& {\mathrm{\β}}_n^t\end{array}\right]---\left(4\right)$

In above-mentioned expression formula,WithFor plural form represents the probability amplitude of quantum bit corresponding state; T is for dyingThe algebraically of colour solid.

Final the obtained fitness of quantum genetic algorithm is target in medium voltage distribution network fault diagnosis modelThe value of function E (x).

(2) solve described modified medium voltage distribution network fault diagnosis model basis according to quantum genetic algorithmFault element in the medium voltage distribution network of result of calculation location is also differentiated protection and the correctness of breaker action,Carry out accident analysis.

In described step 2 (2) step, solve medium voltage distribution network fault diagnosis model according to quantum genetic algorithmComputational methods, comprise the steps:

1. centering is press-fitted Power grid structure analysis, comprising the component kind in clear and definite power distribution networkFor example, and quantity and determining needs the element of analyzing in accident analysis: main transformer, bus, feedbackThe protection configuring condition of line circuit etc. and element.

2. power supply interrupted district after definite distribution network failure;

After power distribution network breaks down, need to dwindle event according to the operating state of switch in power distribution network and protectionBarrier diagnostic area, determines the power supply interrupted district after distribution network failure and the element that need to analyze, therebyMake accident analysis have more specific aim.

3. after occurring according to distribution network failure, the state of each element, switch, protection and breaker, sets upElement state matrix also arranges object function according to the modified medium voltage distribution network fault diagnosis model of step 1,For follow-up analysis of optimization calculating is prepared.

4. according to step 2 (1) step for representing the amount of being adapted to of medium voltage distribution network troubleshooting issueThe probability amplitude of the encoding scheme setting member of sub-genetic algorithm, and to step 2. determined element carry outAssignment, the random number that produces between numerical value interval [0,1], the element state that 3. itself and step setCompare, if random number is more than or equal to probability amplitude, the measurement result of element gets 1, otherwise gets0；

5. the element measured value that 4. the element state value of 3. step being determined and step are determined is brought step 1 intoIn the object function of modified medium voltage distribution network fault diagnosis model, carry out objective function evaluates, determine targetFunction initial value;

6. set the quantum genetic of population scale, chromosome length, corner step-length and maximum iteration timeThe optimization principles of algorithm and utilize quantum genetic algorithm to step 3. object function be optimized calculating; And willThe initial value of result of calculation and step definite object function in 5. compares, if this result of calculation is less thanOr equal initial value, retain currency as target function value; If be greater than initial value, fresh target moreFunctional value; Carry out algorithm iteration, until optimum results meets precision or reaches iterations simultaneously;

7. analysis result, determines the fault element in power distribution network, carries out fault and studies and judges analysis.

In the present embodiment, taking the actual distribution net work structure in somewhere as shown in Figure 2 as example, describe:

1, distribution system grid structure is analyzed

(1) component kind in clear and definite power distribution network and quantity and determining needs point in accident analysisThe element of analysing.

As shown in Figure 2, in the present embodiment the contact structure aspects of distribution system taking " handing in hand " formula as main,In it, contain altogether 6 main transformers, 6 10kV buses, 24 feeder lines, 45 breakers (often cut-off15 of road devices, 30 of normally closed breakers), 108 protections.

(2) based on power distribution network general structure, the element in network and switch are numbered.

Be e by 36 element numbers₁～e₃₆, wherein, bus is numbered B₁～B₆, main transformer is numbered T₁～T₆，Feeder line is numbered L₁～L₂₄. 45 breaker number consecutivelies are CB₁～CB₄₅. In 108 protections, mainEach 36 of protection, nearly back-up protection and back-up protection far away, main protection r_1m～r_36mNumbering situation be:B_1m～B_6m，T_1m～T_6m，L_1m～L_24m. Nearly back-up protection r_1s～r_36sNumbering situation be: B_1s～B_6s，T_1s～T_6s，L_1s～L_24s. Back-up protection r far away_1l～r_36lNumbering situation be: B_1l～B_6l，T_1l～T_6l，L_1l～L_24l. Wherein,M, s, l represent respectively main protection, nearly back-up protection and back-up protection far away. Because feeder line in example power distribution networkBetween be " handing in hand " connection form, therefore back-up protection far away each other between feeder line.

2, distribution system Analysis on Fault Diagnosis

Simulated example test distribution system breaks down, and detects that alarm signal comes from protection T_3s、B_3l、T_4m、L_9sAll there is action, breaker aspect CB₁₆、CB₁₇、CB₂₀、CB₁₃、CB₁₄、CB₂₈、CB₂₉、CB₃₀、CB₃₁Tripping operation.

Therefore, obtain carrying out the event of fault diagnosis according to above example distribution system network topology structureBarrier region and fault element are: B₃、B₄、T₃、T₄、L₈～L₁₅, the state vector that said elements is correspondingE＝[e₁,e₂,……,e₁₂]; The breaker that need to carry out fault diagnosis is: CB₁₆、CB₁₇、CB₁₉～CB₂₆，The virtual condition vector C=[c that above-mentioned breaker is corresponding₁、c₂、……、c₁₀]＝[1,1,0,1,0,0,0,0,0,0]；The protection that need to carry out fault diagnosis is B_3m、B_3s、B_3l、B_4m、B_4s、B_4l、T_3m、T_3s、T_3l、T_4m、T_4s、T_4l、L_8m～L_15m、L_8s～L_15s、L_8l～L_15l, the virtual condition vector of above-mentioned protection correspondence is:R＝[r₁r₂、……r₃₆]＝[0,0,1,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,1,0,0,0,0,0,0,0,0,0,0,0,0,0,0]。

Thus, arrange the table of E (x) according to the object function shown in modified medium voltage distribution network fault diagnosis modelReach form, adopt quantum genetic algorithm to solve object function. The optimization principles setting of optimized algorithmAs follows: population scale is taken as 50, chromosome length is got and is done 10, and corner step-length is set as 0.001, maximumIterations is 800 times. The minimum of a value that finds E (x) after 236 iteration is 9, now makes E (x)Minimum element state vector is E=[e₁,e₂,……,e₁₂]=[0,0,1,1,0,1,0,0,0,0,0,0], corresponding faultElement is: T₃、T₄And L₉。

3, result of calculation analysis

By the result after action situation and the fault diagnosis of each element in power distribution network after fault, Ke YifenAnalyse and obtain: as main transformer T₃After breaking down, main protection T_3mTripping, nearly back-up protection T_3sAction, opens circuitDevice CB₁₆Action; As main transformer T₄After breaking down, main protection T_3mAction, breaker CB₁₇Action; WhenFeeder line L₉After breaking down, main protection L_9mTripping, nearly back-up protection L_9sAction, breaker CB₂₀Action. And normal open switch action is to incite somebody to action because needs utilize contact between feeder line in the distribution system of the present embodimentLoad on dead electricity circuit shifts, and does not do too much analysis at this.

In sum, the fault in the distribution system of the present embodiment is one and contains element main protection trippingMulticomponent fault, utilize the medium voltage distribution network method for diagnosing faults based on quantum genetic algorithm of the present inventionCan carry out exactly fault location and find out fault element. Therefore the present invention carries out event to power distribution networkBarrier diagnosis can obtain unique solution accurately, and modified medium voltage distribution network fault diagnosis mould of the present inventionType can bear the problem such as breaker, protection tripping, for the fault diagnosis of distribution system, ensures distributionThe good operation of system has very high reference value.

It is emphasized that embodiment of the present invention is illustrative, instead of determinate, because ofThis present invention includes and is not limited to the embodiment described in detailed description of the invention, every by art technology peopleOther embodiments that member's technical scheme according to the present invention draws, belong to the scope of protection of the invention equally.

Claims (4)

1. the medium voltage distribution network method for diagnosing faults based on quantum genetic algorithm, is characterized in that: bagDraw together following steps:

Reality and the desired value of step 1, the action of employing element also incorporates breaker fail protection and breakerThe guard mode of automatic reclosing is set up modified medium voltage distribution network fault diagnosis model; The target of this modelFunction is:

$\begin{array}{c}E\left(x\right)=\underset{k=1}{\overset{n}{\mathrm{\Σ}}}|r_{k,m}-r_{k,m}^{*}||1-r_{k,s}{r}_{k,s}^{*}-\stackrel{\⊕}{\mathrm{\Σ}}{r}_{k,l}-r_{k,l}^{*}|+\underset{k=1}{\overset{n}{\mathrm{\Σ}}}|r_{k,s}-r_{k,s}^{*}||1-\stackrel{\⊕}{\mathrm{\Σ}}{r}_{k,l}-r_{k,l}^{*}|\\ +\underset{k=1}{\overset{n}{\mathrm{\Σ}}}|r_{k,l}-r_{k,l}^{*}|+\underset{i=1}{\overset{q}{\mathrm{\Σ}}}|r_{i,cb}-r_{i,cb}^{*}|+\underset{i=1}{\overset{q}{\mathrm{\Σ}}}|r_{i,auto}-r_{i,auto}^{*}|\\ +\underset{i=1}{\overset{q}{\mathrm{\Σ}}}|C_i-C_i^{*}||1-r_{i,cb}{r}_{i,cb}^{*}-r_{i,auto}-r_{i,auto}^{*}|\end{array}$

In above-mentioned expression formula, r_k,mAnd r_k,m ^*Represent respectively the virtual condition of each element main protection and expect shapeState; r_k,sAnd r_k,s ^*Represent respectively virtual condition and the expectation state of the nearly back-up protection of discrete component; r_k,lAnd r_k,l ^*Represent respectively virtual condition and the expectation state of discrete component back-up protection far away; C_iAnd C_i ^*Represent respectively to open circuitThe virtual condition of device and expectation state;Represent to connect exclusive disjunction; r_i,cbAnd r_i,cb ^*Represent respectively breaker failure guarantorThe virtual condition of protecting and expectation state; r_i,autoAnd r_i,auto ^*Represent respectively the virtual condition of breaker automatic reclosingAnd expectation state.

Step 2, solve the modified medium voltage distribution network fault diagnosis model based on quantum genetic algorithm, centeringBe press-fitted electrical network and carry out fault diagnosis.

2. a kind of medium voltage distribution network fault diagnosis based on quantum genetic algorithm according to claim 1Method, is characterized in that: the concrete steps of described step 2 comprise:

(1) adopt traditional genetic algorithm to after element is encoded in intermediate distribution system, according to quantum ratioSpecial coded system is revised, thereby formulates the encoding scheme that is adapted to quantum genetic algorithm in order to representMedium voltage distribution network troubleshooting issue;

(2) solve described modified medium voltage distribution network fault diagnosis model basis according to quantum genetic algorithmFault element in the medium voltage distribution network of result of calculation location is also differentiated protection and the correctness of breaker action,Carry out accident analysis.

3. a kind of medium voltage distribution network fault diagnosis based on quantum genetic algorithm according to claim 2Method, is characterized in that: the specific coding method of described step (1) is:

Suppose that power supply interrupted district entirety is for individual chromosome q, in distribution system, component population is described chromosome qIn gene number n; Adopt the coded system of the quantum bit in quantum genetic algorithm, use a pair of plural numberDefine a quantum bit position, individual chromosome q adopts quantum bit coding to solve fault diagnosis to askThe concrete form of topic is:

$q^t=\left[\begin{array}{cccc}{\mathrm{\α}}_1^t& {\mathrm{\α}}_2^t& ...& {\mathrm{\α}}_n^t\\ {\mathrm{\β}}_1^t& {\mathrm{\β}}_2^t& ...& {\mathrm{\β}}_n^t\end{array}\right]$

In above-mentioned expression formula,WithFor plural form represents the probability amplitude of quantum bit corresponding state; T is for dyingThe algebraically of colour solid.

4. a kind of medium voltage distribution network fault diagnosis based on quantum genetic algorithm according to claim 2Method, is characterized in that: described step solves medium voltage distribution network fault according to quantum genetic algorithm in (2)The computational methods of diagnostic model, comprise the steps:

1. centering is press-fitted Power grid structure analysis, specifies component kind and quantity in power distribution network,Determine and in accident analysis, need the element analyzed;

2. dwindle fault diagnosis scope according to the operating state of switch in power distribution network and protection, determine power distribution networkPower supply interrupted district after fault and the element that need to analyze;

3. after occurring according to distribution network failure, the state of each element, switch, protection and breaker, sets upElement state matrix also arranges object function according to the modified medium voltage distribution network fault diagnosis model of step 1;

4. according to step 2 (1) step for representing the amount of being adapted to of medium voltage distribution network troubleshooting issueThe probability amplitude of the encoding scheme setting member of sub-genetic algorithm, and to step 2. determined element carry outAssignment, the random number that produces between numerical value interval [0,1], the element state that 3. itself and step setCompare, if random number is more than or equal to probability amplitude, the measurement result of element gets 1, otherwise gets0；

5. the element measured value that 4. the element state value of 3. step being determined and step are determined is brought step 1 intoIn the object function of modified medium voltage distribution network fault diagnosis model, carry out objective function evaluates, determine targetFunction initial value;

6. set the quantum genetic of population scale, chromosome length, corner step-length and maximum iteration timeThe optimization principles of algorithm also utilizes quantum genetic algorithm to be optimized calculating to step object function 3.; AndThe initial value of result of calculation and step definite object function in is 5. compared, if this result of calculation is littleIn or equal initial value, retain currency as target function value; If be greater than initial value, upgrade orderOffer of tender numerical value; Carry out algorithm iteration simultaneously, until optimum results meets precision or reaches iterations beOnly;

7. analysis result, determines the fault element in power distribution network, carries out fault and studies and judges analysis.