CN106408155A - Reliability evaluating and preconceived fault set searching method based on related circuit set - Google Patents

Abstract

The invention discloses a reliability evaluating and preconceived fault set searching method based on a related circuit set. The method comprises the following steps that a quantitative reliability influence index QRI for evaluating construction of a single planned project of the power grid is provided, the single planed project comprises a reliability index set, and the quantitative reliability influence index QRI is the weighted sum of relative variances of reliability indexes; a related circuit searching algorithm is used to search the power grid for circuits which have trend change before and after construction of the planned project, and the searched circuits form the related circuit set; and the screened circuit set serves as a corresponding preconceived fault set, reliability indexes of corresponding preconceived faults are calculated, the reliability indexes are substituted to obtain the quantitative reliability influence index QRI, and the preconceived fault set is a fault set for calculating the reliability indexes correspondingly. According to the invention, global fault scanning can be avoided, the fault scanning range is narrowed, curse of dimensionality is avoided, and the reliability indexes can be calculated rapidly.

Description

Reliability assessment forecast failure collection searching method based on line related collection

Technical field

The present invention relates to the reliability assessment field of Power System Planning project is and in particular to a kind of be based on line related collection Search is thus reduce the implementation method of reliability assessment forecast failure collection scale.

Background technology

With the development of economic society, power system also becomes increasingly huger to meet growing workload demand Greatly, electric power networks are carried out with the difficulty of reliability assessment, and also with the growth of component number, exponentially increases.In existing Pang In big power system, the calculating time how reducing reliability assessment has become as the focus studied at present.

After needing each of power system element is carried out with exhaustive scanning due to traditional reliability estimation method Reliability index could be obtained, be more concerned with the scope of fault scanning, rather than the depth of local region scans, to single new The planned project assessment specific aim built is not strong, less efficient.Grid company is more desirable to know the construction of certain planned project simultaneously For electric network reliability lifting major embodiment where, quantized value is how many, to hold the significance level of each project, really Determine the optimization of sequence between projects, and determine the optimum programme of cost performance in limited fund.

Based on background above, the present invention proposes the reliability assessment towards single planned project, is more concerned with some The situation of change of electric network reliability level before and after project construction, and propose the reliability effect quantitative evaluation of single planned project Index QRI.Need the key technical problem solving be how intelligence identify after a planned project construction, which fault Impact to electrical network there occurs obvious change, and thus produces the multiple disturbance set being adapted with project, for reliability The assessment of impact.

Content of the invention

Present invention aims to above-mentioned problems of the prior art are it is proposed that a kind of be based on line related collection Forecast failure collection filtering algorithm, the method can by search for each project cluster line related collection, with line related collection group Become the scanning failure collection of reliability assessment, thus avoiding whole scan fault, reducing the scope of fault scanning, avoiding simultaneously The problem of dimension calamity, reaches the quick purpose calculating reliability index.

For reaching the purpose of foregoing invention, the present invention is realized by the technical scheme of following examples：

A kind of reliability assessment forecast failure collection searching method based on line related collection, comprises the steps：

Step S10, is proposed for the reliability effect quantizating index that the single planned project of assessment of electric power networks is built QRI, single planned project includes reliability index set, and it includes system and cuts down load-loss probability, expected load reduction With these three reliability indexs of the system order of severity, described reliability effect quantizating index QRI is the relative of each reliability index The value preset of variable quantity weighting；

Step S20, using line related searching algorithm, searches for and has trend to become in electric power networks before and after planned project construction The circuit composition line related collection changed, described line related collection refer in electrical network certain branch road hinder for some reason removed after, be subject to This branch road shifts influence on tidal flow and the branch road being directly or indirectly connected with this branch road two ends fulcrum；Search for a certain bar plotted line road Line related collection is exactly the front k bar shortest path seeking this circuit, that is, be converted into the front k bar power asked from source node to destination node It is worth minimum path, the weights in path refer to the resistance value in path here；

Step S30, using the line related filtering out collection as corresponding forecast failure collection, calculate corresponding forecast failure can By property index, each reliability index substitution step S10 is tried to achieve reliability effect quantizating index QRI, described forecast failure collection is Calculate the corresponding failure collection of each reliability index.

Described reliability effect quantitative evaluation index QRI：

In formula, QRI is reliability effect quantitative evaluation index；RI_i(before) refer to for reliability before planned project construction The value of mark i；RI_i(after) for the value of reliability index i after planned project construction；I is the reliability index set chosen；W_iFor The weight of each reliability index i in reliability index set I, and meet ∑ W_i=1.

Described electric power networks have rack data, and described rack data includes source node Vs, destination node Vt and branch road and compiles Number.

Further, the line related searching algorithm of described step S20 is specific as follows：

Step S21, imports the rack data of electric power networks, and rack data configuration is become bus admittance matrix γ, and The topological diagram of the non-negative weights being consequently formed；

Step S22, setting first tables of data pathstr stores the routing information in search procedure, described first tables of data Pathstr comprises four fields, respectively node name NA, to the predecessor node collection P of source node, path node sum L and Zong Lu Path length W；Assume that topological diagram G has n summit, vertex set is V={ v₁,v₂……v_n, the collection on m bar side is combined into E={ e₁, e₂……e_m, every a line e_i=(v_x,v_y) corresponding non-negative weight w (v_x,v_y), represent the weight on this side, dist (v_x,v_y) table Show from node v_xTo node v_yShortest path value, define R path power be the comprised side of path R weight sue for peace, be designated as

Step S23, repeatedly checks the physical planning rack of electrical network, determines minimum impedance multiple limit coefficient ε and path rank Number limit coefficient L_max；In power system, if the corresponding impedance of impedance multiple limit coefficient ε in this path is bigger, by tide The impact that circulation moves is less, and when impedance is more than certain value, this path may be considered impregnable；In the same manner, when this road Interstitial content L included in footpath_maxIt is believed that this path is long and invalid because of electrical distance after exceeding some, L_max Value and ε value determine after being required for repeatedly being calculated and checking according to the physical planning rack of electrical network；The embodiment of the present invention is taken Parameter is set to according to after calculating to example：L_max=12；ε=4；

Step S24, from source node v_sSet out, search for it and arrive destination node v_tMinimum nodes path, with the road found at first Footpath is shortest path, its path dist (v_s,v_t)_{min node}, wherein v_s,v_tRespectively cut-off source node and the purpose of branch road Node；

Step S25, from destination node v_tBacktracking searches its source node v_sPath, ask for front k bar shortest path, and will The paths record finally tried to achieve with the second tables of data kpathstr of the first tables of data pathstr same data structure in；

Step S26, reexamines whether all paths of the second tables of data kpathstr meetWhereinRepresent source node v_sTo destination node v_tTotal path length,Source node v for shortest path_sTo destination node v_t Total path length, and delete the path of the condition of being unsatisfactory for, all paths in the second last tables of data kpathstr are phase Close sets of lines.

Further, described step S24 is from source node v_sThe search set out includes：

Step S241, initialising first data table pathstr is sky, then by node v_sPut into first as a record In tables of data pathstr, its field information is [v_s,null,0,0]；

Step S242, in the n-th step circulation, finds out path node sum in the first tables of data pathstrInstitute for n-1 There is node v_n, to any one node v_n, find out bus admittance matrix line n or n-th row in all be not 0 node v_j, and v_jNot in v_nPredecessor node collectionIn, all these node v_jThe collection constituting is combined into A；Again by the information of each node in set A It is stored in the first tables of data pathstr, such as to v_j∈ A, its field information is Represent node v_nPredecessor node collection,Represent node v_nTo the path node sum of source node,Represent node v_nTo source The total path length of node；

Step S243, if judge v_jFor destination node v_t, then predecessor node collectionIt is shortest path, length isIf it is not, then judging path node sum whether more than path node sum limit coefficient L_maxIf exponent number is more than L_max, then stop to v_jDeep search；Conversely, then repeat step S242；If all sections in set A The path node sum of point is both greater than L_maxThen it is assumed that path no solves；If A is empty set, this node no adjacent node is described, equally Think that this path no solves.

Further, described step S25 is from destination node v_tThe search set out includes：

Step S251, being located at the minimum nodes path tried to achieve in previous step is shortest path Initialising second data table kpathstr is empty set, empties the first tables of data pathstr；By destination node v_tRecord [v_t, Null, 0,0] it is stored in pathstr；

Step S252, in N step circulation, finds out path node sum in the first tables of data pathstrInstitute for N-1 There is node v_N；

Step S253, if v_NPath node sum more than maximum order limit L_max, then no longer calculate its adjacent segments Point；Conversely, setting all and v_NAdjacent and be not belonging to predecessor node collectionNode v_JConstitute set B；IfAlthough then not less than restriction, path has been above minimum nodes to explanation exponent number The certain multiple of path, continues search for nor becomes front k bar shortest path, therefore this path can terminate；

Step S254, if v_s∈ B, then this node and its relevant information are stored in the second tables of data kpathstr；Relatively The weighted value in this pathWithIfThen the weighted value of renewal shortest path isInstead It, then keep the weight of shortest pathConstant；IfAlthough then illustrating that this path is more than Shortest path length, but within the specific limits, can temporarily include front k bar shortest path, and the information of active path is deposited Enter in the second tables of data kpathstr；IfIt is then Invalid path；

Step S255, ifThen by all elements v in B_JInformation updating after be stored in one the 3rd tables of data Nextpathstr, to carry out next step to expand search；v_JThe information being stored in is

Step S256, judges whether the first tables of data pathstr interior joint has read, return to step S252 if not, in this way Then the 3rd tables of data nextpathstr is stored in the first tables of data pathstr, removes the 3rd tables of data nextpathstr, put N=N+1, return to step S251.

A kind of reliability assessment forecast failure collection searching method based on line related collection of the present invention, has beneficial as follows Effect：

Calculate the reliability index specific aim of the whole network for traditional system adequacy evaluation method using the fault set setting Not strong problem, first proposed reliability effect quantitative evaluation index QRI of single planned project, come quantitative description some The contribution rate to electric network reliability for the construction of planned project, proposes a kind of calculating line related collection to reduce electrical network anticipation simultaneously The scheme of fault set, producing in electrical network before and after being built by a certain planned project of front k bar Shortest Path Searching Algorithm searching can See the line set composition forecast failure collection of power flow changing, to reduce fault search scope.

Brief description

Fig. 1 is the block diagram of the reliability assessment forecast failure collection searching method based on line related collection of the present invention.

Fig. 2 is the FB(flow block) of the reliability assessment forecast failure collection searching method based on line related collection of the present invention.

Fig. 3 is IEEE39 node system wiring diagram.

Fig. 4 is position view in IEEE39 node system wiring diagram for the line related collection.

Specific embodiment

With reference to the accompanying drawings and examples the technical scheme in the embodiment of the present invention is clearly and completely described, shows So, described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.

Referring to Fig. 1 and Fig. 2, respectively the reliability assessment forecast failure collection based on line related collection of the embodiment of the present invention The block diagram of searching method and FB(flow block).

A kind of reliability assessment forecast failure collection searching method based on line related collection, comprises the steps：

Step S10, is proposed for the reliability effect quantizating index that the single planned project of assessment of electric power networks is built QRI, single planned project includes reliability index set, and it includes, and system cuts down load-loss probability LOLP, expected load is cut Decrement EENS and these three reliability indexs of system order of severity SI, described reliability effect quantizating index QRI is each reliability The value preset of the relative variation weighting of index；

Step S20, using line related searching algorithm, searches for and has trend to become in electric power networks before and after planned project construction The circuit composition line related collection changed, described line related collection refer in electrical network certain branch road hinder for some reason removed after, be subject to This branch road shifts influence on tidal flow and the branch road being directly or indirectly connected with this branch road two ends fulcrum；Search for a certain bar plotted line road Line related collection is exactly the front k bar shortest path seeking this circuit, that is, be converted into the front k bar power asked from source node to destination node It is worth minimum path, the weights in path refer to the resistance value in path here；

Step S30, using the line related filtering out collection as corresponding forecast failure collection, calculates corresponding reliability index, Each reliability index substitution step S10 is tried to achieve reliability effect quantizating index QRI, described forecast failure collection is to calculate each reliability The property corresponding failure collection of index.

Described reliability effect quantitative evaluation index QRI：

In formula, QRI is reliability effect quantitative evaluation index；RI_i(before) refer to for reliability before planned project construction The value of mark i；RI_i(after) for the value of reliability index i after planned project construction；I is the reliability index set chosen；W_iFor The weight of each reliability index i in reliability index set I, and meet ∑ W_i=1.

System described in step S10 cuts down load-loss probability, expected load reduction and the system order of severity, and these three can By property index, each index definition is as follows：

1) system cuts down load-loss probability LOLP, represents and leads to lose the probability of load due to system element off-capacity Size, its expression formula：

In formula,P (x) expression system is in the probability of state x, can be by following side Formula calculates, for element k, if its fault rate is λ_k(times/year), repair rate is μ_k(times/year), mean repair time is r_k(little When/time), then have：

This is the state probability of an element, if containing N number of independent component in power system, system mode probability is then etc. Long-pending in wherein each element state probability；

2) expected load reduction EENS, the annual average electricity of expression system lack confession number, unit be MWh/, its table Reaching formula is：

In formula, t (x) represents the persistent period that x cutting load is powered from failure system state to recovery load；L_vX () represents Under failure system state x, for restoring the system to minimum load reduction necessary to a static security operating point；

3) system order of severity SI, its expression formula is：

L in formula_maxFor system peak load, the unit of SI is " system is divided ", and a SI is equivalent to total system in peak load When have a power failure 1 minute, be to the system failure order of severity one kind tolerance.

Described electric power networks have rack data, and described rack data includes source node Vs, destination node Vt and branch road and compiles Number.

As a specific embodiment, in conjunction with Fig. 1 and Fig. 2, the line related searching algorithm of described step S20 is concrete such as Under：

Step S21, imports the rack data of electric power networks, and rack data configuration is become bus admittance matrix γ, and The topological diagram of the non-negative weights being consequently formed；The described topology information importing electric power networks, compiles to each node and branch road Number and set up corresponding relation, importing the admittance of each bar branch road, forming bus admittance matrix, thus forming the topology of non-negative weights Figure, as shown in formula (6), considers the simplicity calculating, only takes the imaginary part of each element in bus admittance matrix to node admittance, That is ignoring the impact of resistance in electrical network；

Step S22, setting first tables of data pathstr stores the routing information in search procedure, described first tables of data Pathstr comprises four fields, respectively node name NA, to the predecessor node collection P of source node, path node sum L and Zong Lu Path length W；Assume that topological diagram G has n summit, vertex set is V={ v₁,v₂……v_n, the collection on m bar side is combined into E={ e₁, e₂……e_m, every a line e_i=(v_x,v_y) corresponding non-negative weight w (v_x,v_y), represent the weight on this side, dist (v_x,v_y) table Show from node v_xTo node v_yShortest path value, define R path power be the comprised side of path R weight sue for peace, be designated as

Step S23, repeatedly checks the physical planning rack of electrical network, determines minimum impedance multiple limit coefficient ε and path rank Number limit coefficient L_max；In power system, if the corresponding impedance of impedance multiple limit coefficient ε in this path is bigger, by tide The impact that circulation moves is less, and when impedance is more than certain value, this path may be considered impregnable；In the same manner, when this road Interstitial content L included in footpath_maxIt is believed that this path is long and invalid because of electrical distance after exceeding some, L_max Value and ε value determine after being required for repeatedly being calculated and checking according to the physical planning rack of electrical network；The embodiment of the present invention is taken Parameter is set to according to after calculating to example：L_max=12；ε=4；

Step S24, from source node v_sSet out, search for it and arrive destination node v_tMinimum nodes path, with the road found at first Footpath is shortest path, its path dist (v_s,v_t)_{min node}, wherein v_s,v_tRespectively cut-off source node and the purpose of branch road Node；

Step S25, from destination node v_tBacktracking searches its source node v_sPath, ask for front k bar shortest path, and will The paths record finally tried to achieve with the second tables of data kpathstr of the first tables of data pathstr same data structure in；

Step S26, reexamines whether all paths of the second tables of data kpathstr meetWhereinRepresent source node v_sTo destination node v_tTotal path length,Source node v for shortest path_sTo destination node v_t Total path length, and delete the path of the condition of being unsatisfactory for, all paths in the second last tables of data kpathstr are phase Close sets of lines.

As a specific embodiment, in conjunction with shown in Fig. 2, described step S24 is from source node v_sThe search set out includes：

Step S241, initialising first data table pathstr is sky, then by node v_sPut into first as a record In tables of data pathstr, its field information is [v_s,null,0,0]；

Step S242, in the n-th step circulation, finds out path node sum in pathstrAll node v for n-1_n, To any one node v_n, find out bus admittance matrix line n or n-th row in all be not 0 node v_j, and v_jNot in v_n Predecessor node collectionIn, all these node v_jThe collection constituting is combined into A；Again the information of each node in set A is stored in In pathstr, such as to v_j∈ A, its field information is Represent node v_n's Predecessor node collection,Represent node v_nTo the path node sum of source node,Represent node v_nLong to the total path of source node Degree；

Step S243, if judge v_jFor destination node v_t, then predecessor node collectionIt is shortest path, length isIf it is not, then judging path node sum whether more than path node sum limit coefficient L_maxIf exponent number is more than L_max, then stop to v_jDeep search；Conversely, then repeat step S242；If all sections in set A The path node sum of point is both greater than L_maxThen it is assumed that path no solves；If A is empty set, this node no adjacent node is described, equally Think that this path no solves.

As a specific embodiment, in conjunction with shown in Fig. 2, described step S25 is from destination node v_tThe search set out includes：

Step S251, being located at the minimum nodes path tried to achieve in previous step is shortest path Initialising second data table kpathstr is empty set, empties the first tables of data pathstr；By destination node v_tRecord [v_t, Null, 0,0] it is stored in pathstr；

Step S252, in N step circulation, finds out path node sum in the first tables of data pathstrInstitute for N-1 There is node v_N；

Step S253, if v_NPath node sum more than maximum order limit L_max, then no longer calculate its adjacent segments Point；Conversely, setting all and v_NAdjacent and be not belonging to predecessor node collectionNode v_JConstitute set B；IfAlthough then not less than restriction, path has been above minimum nodes to explanation exponent number The certain multiple of path, continues search for nor becomes front k bar shortest path, therefore this path can terminate；

Step S254, if v_s∈ B, then this node and its relevant information are stored in the second tables of data kpathstr；Relatively The weighted value in this pathWithIfThen the weighted value of renewal shortest path isInstead It, then keep the weight of shortest pathConstant；IfAlthough then illustrating that this path is big In shortest path length, but within the specific limits, can temporarily include front k bar shortest path, and the information by active path It is stored in the second tables of data kpathstr；IfIt is then Invalid path；

Step S255, ifThen by all elements v in B_JInformation updating after be stored in one the 3rd tables of data Nextpathstr, to carry out next step to expand search；v_JThe information being stored in is

Step S256, judges whether the first tables of data pathstr interior joint has read, return to step S252 if not, in this way Then the 3rd tables of data nextpathstr is stored in the first tables of data pathstr, removes the 3rd tables of data nextpathstr, put N=N+1, return to step S251.

Beneficial effect below by citing checking computations explanation foregoing invention embodiment：

In the present embodiment, method is used IEEE39 node system as simulation example, and emulation tool uses by China The PSD-BPA simulation software of Electric Power Research Institute's exploitation.IEEE39 node system includes 10 electromotors, 46 branch road (bags Include transformer branch and circuit branch road), 39 nodes, electric pressure 100kV, system wiring figure is as shown in Figure 3.

Restrictive condition is arranged：

Assume to plan a newly-built circuit between BUS_39 and BUS_3, the restrictive condition of setting is that path exponent number is 10, that is, path contain up to 10 nodes；Minimum impedance limits multiple and is 4 times, that is, when the path searching is more than currently It is believed that this path is invalid when 4 times of shortest path length.

Line related collection Search Results：

After screening, the front k bar shortest path between BUS_39 and BUS_3 is：

Front k bar shortest path between table 1 BUS_39 and BUS_3

According to the definition of line related collection, the circuit included in front k bar shortest path forms line related collection. Therefore under current restrictive condition, line related collection is as shown in the table.

Table 2 BUS_39-BUS_3 plans the line related collection of circuit

Line related collection verifies：

Whether it is these branch roads found out for the larger branch road of verification power flow changing, this method passes through BPA simulation software, right Before and after planning line construction, IEEE39 node system carries out Load flow calculation, and each bar Branch Power Flow change is as shown in the table.

BPA simulation result before and after the construction of table 3 planned project, unit/MW

By the analysis of upper table, in IEEE39 node system, new when planning between node BUS_3 and BUS_39 When building a circuit, Line Flow is only had 28 branch roads by this planning line influence, accounts for the 60.1% of total circuitry number, and it Be near newly-built circuit, not there is significant change in the active power of remaining 18 branch road, this turned out when planning During a newly-built circuit, its impact to trend is in certain limit around this circuit really.

Synopsis 2, by the branch road found out by the line related collection searching algorithm of this method, overstriking shows in table 3.When Setting path exponent number is 10, and when minimum impedance limits multiple for 4 times, the circuit being found only has 17, occupies power flow changing The 60% of the number of lines.Notice the circuit which includes by before the descending sequence of change of power flow 50% simultaneously, say Bright method can correctly identify the position of power flow transfer.

Restrictive condition is changed：

In order to lift the precision of line related collection search, the limit coefficient in program is modified, specially：Path rank Number is changed to 12 ranks, minimum impedance limit multiple be maintained 4 times constant.

Line related collection Search Results：

In addition to path obtained before, can also search for other 5 paths as shown in the table.

Front k bar shortest path (L between table 4 BUS_39 and BUS_3_max=12；ε=4)

The circuit that now line related concentration will increase is as shown in the table.

Table 5 BUS_39-BUS_3 plans the line related collection (L of circuit_max=12；ε=4)

From above-mentioned two tables, after path node limits and increases, the number of path being found increases, and wherein comprise props up Road is also more.When node limits number and increases to 12 by 10, introductory path increases to 9 by original 4, line related collection Closing increases to 27 by original 17.Circuit in table 5 is found out successively in table 3 corresponding circuit, now line related Contain 28 27 having in power flow changing circuit in set, account for 96.5%.And do not include in line related set Circuit be BUS_6 to BUS_31, its active power variable quantity also only has 0.1MW, and variable quantity is relatively small.Now it is considered that Two limit coefficient (node limits number and minimum impedance limits multiple) set by this method, have reached receptible essence Degree scope.

Now position in wiring diagram for the line related collection is as shown in Figure 4.

Reliability index and the calculating of QRI：

The present invention chooses load-loss probability (LOLP), expected load reduction (EENS) and the system order of severity (SI) 3 Reliability index forms reliability index set, makees forecast failure collection with line related collection respectively and the method for whole scan is counted Calculate the change of system reliability before and after planned project is built.

In order to verify the correctness of the present invention, need to carry out 4 Calculation of Reliability：

(1) index is obtained using the reliability that whole scan assesses original rack

(2) index is obtained using the reliability that the forecast failure collection original rack of assessment made by line related collection

(3) index is obtained using the reliability that whole scan assesses new rack

(4) index is obtained using the reliability that the forecast failure collection new rack of assessment made by line related collection

Result of calculation is as shown in table 6：

Table 6 reliability index result of calculation

For the simplicity calculating, the present invention carrys out parameter QRI using the method for equal rights, is respectively adopted line related collection Make forecast failure collection and method calculated reliability quantitative evaluation index QRI of whole scan is as follows：

Use two methods calculated QRI index result closely respectively, demonstrate the correctness of the present invention and have Effect property.

Above-described embodiment only not limits technical scheme described in the invention in order to the present invention to be described；Therefore, although This specification with reference to each above-mentioned embodiment to present invention has been detailed description, but, the ordinary skill of this area Personnel should be appreciated that and still can the present invention be modified or equivalent；And all without departing from the present invention spirit and The technical scheme of scope and its improvement, it all should be covered in the middle of scope of the presently claimed invention.

Claims (6)

1. a kind of reliability assessment forecast failure collection searching method based on line related collection is it is characterised in that include following walking Suddenly：

Step S10, is proposed for the reliability effect quantizating index QRI that the single planned project of assessment of electric power networks is built, single Individual planned project includes reliability index set, and it includes system and cuts down load-loss probability, expected load reduction and system These three reliability indexs of the order of severity, described reliability effect quantizating index QRI is the relative variation of each reliability index The value preset of weighting；

Step S20, using line related searching algorithm, searching in electric power networks has power flow changing before and after planned project construction Circuit forms line related collection.

Step S30, using the line related filtering out collection as corresponding forecast failure collection, calculates the reliability of corresponding forecast failure Index, each reliability index substitution step S10 is tried to achieve reliability effect quantizating index QRI, and described forecast failure collection is to calculate The corresponding failure collection of each reliability index.

2. the reliability assessment forecast failure collection searching method based on line related collection according to claim 1, its feature It is, described reliability effect quantitative evaluation index QRI：

$QRI=\left(\underset{i\∈I}{\Σ}{W}_i\frac{|{\mathrm{RI}}_i\left(before\right)-{\mathrm{RI}}_i\left(after\right)|}{{\mathrm{RI}}_i\left(before\right)}\right)*100\%$

In formula, QRI is reliability effect quantitative evaluation index；RI_i(before) for reliability index i before planned project construction Value；RI_i(after) for the value of reliability index i after planned project construction；I is the reliability index set chosen；W_iFor reliability The weight of each reliability index i in property index set I, and meet ∑ W_i=1.,

3. the reliability assessment forecast failure collection searching method based on line related collection according to claim 1, its feature It is, described electric power networks have rack data, described rack data includes source node Vs, destination node Vt and branch number.

4. the reliability assessment forecast failure collection searching method based on line related collection according to claim 3, its feature It is, the line related searching algorithm of described step S20 is specific as follows：

Step S21, imports the rack data of electric power networks, and rack data configuration is become bus admittance matrix γ, and thus The topological diagram of the non-negative weights being formed；

Step S22, setting first tables of data pathstr stores the routing information in search procedure, described first tables of data Pathstr comprises four fields, respectively node name NA, to the predecessor node collection P of source node, path node sum L and Zong Lu Path length W；

Step S23, checks the physical planning rack of electrical network repeatedly, determines minimum impedance multiple limit coefficient ε and path exponent number limit Coefficient L processed_max；

Step S24, from source node v_sSet out, search for it and arrive destination node v_tMinimum nodes path, with the path found at first be Shortest path, its path dist (v_s,v_t)_minnode, wherein v_s,v_tRespectively cut-off source node and the destination node of branch road；

Step S25, from destination node v_tBacktracking searches its source node v_sPath, ask for front k bar shortest path, and will be final The paths record tried to achieve with the second tables of data kpathstr of the first tables of data pathstr same data structure in；

Step S26, reexamines whether all paths of the second tables of data kpathstr meetWhereinTable Show source node v_sTo destination node v_tTotal path length,Source node v for shortest path_sTo destination node v_tTotal road Electrical path length, and delete the path of the condition of being unsatisfactory for, all paths in the second last tables of data kpathstr are line related Collection.

5. the reliability assessment forecast failure collection searching method based on line related collection according to claim 4, its feature It is, described step S24 is from source node v_sThe search set out includes：

Step S241, initialising first data table pathstr is sky, then by node v_sPut into the first data as a record In table pathstr, its field information is [v_s,null,0,0]；

Step S242, in the n-th step circulation, finds out path node sum in the first tables of data pathstrAll sections for n-1 Point v_n, to any one node v_n, find out bus admittance matrix line n or n-th row in all be not 0 node v_j, and v_jNo In v_nPredecessor node collectionIn, all these node v_jThe collection constituting is combined into A；Again the information of each node in set A is deposited Enter in the first tables of data pathstr, such as to v_j∈ A, its field information is Represent node v_nPredecessor node collection,Represent node v_nTo the path node sum of source node,Represent node v_nSave to source The total path length of point；

Step S243, if judge v_jFor destination node v_t, then predecessor node collectionIt is shortest path, length isIf it is not, then judging path node sum whether more than path node sum limit coefficient L_maxIf exponent number is more than L_max, then stop to v_jDeep search；Conversely, then repeat step S242；If all sections in set A The path node sum of point is both greater than L_maxThen it is assumed that path no solves；If A is empty set, this node no adjacent node is described, equally Think that this path no solves.

6. the reliability assessment forecast failure collection searching method based on line related collection according to claim 4, its feature It is, described step S25 is from destination node v_tThe search set out includes：

Step S251, being located at the minimum nodes path tried to achieve in previous step is shortest pathJust Beginningization second tables of data kpathstr is empty set, empties the first tables of data pathstr；By destination node v_tRecord [v_t,null, 0,0] it is stored in pathstr；

Step S252, in N step circulation, finds out path node sum in the first tables of data pathstrAll sections for N-1 Point v_N；

Step S253, if v_NPath node sum more than maximum order limit L_max, then no longer calculate its adjacent node；Instead It, if all and v_NAdjacent and be not belonging to predecessor node collectionNode v_JConstitute set B；IfAlthough then not less than restriction, path has been above minimum nodes to explanation exponent number The certain multiple of path, continues search for nor becomes front k bar shortest path, therefore this path can terminate；

Step S254, if v_s∈ B, then this node and its relevant information are stored in the second tables of data kpathstr；Relatively this path Weighted valueWithIfThen the weighted value of renewal shortest path isConversely, then Keep the weight of shortest pathConstant；IfAlthough it is the shortest then to illustrate that this path is more than Path, but within the specific limits, can temporarily include front k bar shortest path, and the information of active path is stored in In two tables of data kpathstr；IfIt is then Invalid path；

Step S255, ifThen by all elements v in B_JInformation updating after be stored in one the 3rd tables of data Nextpathstr, to carry out next step to expand search；v_JThe information being stored in is

Step S256, judges whether the first tables of data pathstr interior joint has read, return to step S252 if not, then will in this way 3rd tables of data nextpathstr is stored in the first tables of data pathstr, removes the 3rd tables of data nextpathstr, puts N=N + 1, return to step S251.