CN108242806B

CN108242806B - Feeder automation switch distribution method and system

Info

Publication number: CN108242806B
Application number: CN201611217411.1A
Authority: CN
Inventors: 李志铿; 孙浩; 张磊; 王海华; 杨海森; 张学昶; 赵青宇; 刘金生
Original assignee: China Energy Engineering Group Guangdong Electric Power Design Institute Co Ltd
Current assignee: China Energy Engineering Group Guangdong Electric Power Design Institute Co Ltd
Priority date: 2016-12-26
Filing date: 2016-12-26
Publication date: 2020-02-18
Anticipated expiration: 2036-12-26
Also published as: CN108242806A

Abstract

The invention relates to a feeder automation switch distribution method and a feeder automation switch distribution system. The method comprises the following steps: searching a ring main unit, a plurality of areas, an overhead line, a reconfigurable switch of the ring main unit, the number of users in each area and the number of switches of each ring main unit according to the obtained network topological graph; acquiring the equivalent failure rate of each area and the power failure duration of each area in the repair stage; establishing a dual-target distribution point planning model according to preset time, power failure time of a repair stage of each area, preset cost, a modifiable switch, a ring main unit, areas, equivalent failure rate of each area, the number of users of each area and the number of switches of each ring main unit; solving a binocular point distribution planning model to obtain a pareto frontier formed by a plurality of solutions, and selecting an optimized solution from the pareto frontier; and selecting a stationing switch from the reconfigurable switch according to the optimization solution, and generating and outputting an automatic stationing scheme according to the stationing switch. The switch is suitable for switch transformation of cables and overhead lines, and has the advantages of high cost, long average power failure time and high practicability.

Description

Feeder automation switch distribution method and system

Technical Field

The invention relates to the technical field of power systems, in particular to a feeder automation switch distribution method and a feeder automation switch distribution system.

Background

With the rapid and continuous development of society and economy, power companies urgently need to improve the quality and reliability of power supply so as to meet the continuous requirement of users on power supply. The feeder automation technology is an effective means for rapidly improving the power supply reliability of the power system.

The working principle of feeder automation is as follows: the on-site power distribution terminal transmits fault information to the control center through a channel, the control center performs fault analysis according to network topology, switch states, fault information and the like, judges a fault section and issues a remote control command, three remote switches on two sides of a fault area are tripped to realize fault isolation, the outgoing line switch of the reclosing transformer substation supplies power to a fault upstream area, the interconnection switch is closed to supply power to a fault downstream area, and power supply of a non-fault area of a line is completed. The feeder automation switch placement is not only directly related to the investment cost of engineering construction, but also closely linked to the reliability level after the feeder automation is implemented.

The switch on the feeder line comprises a switch on the overhead feeder line and a switch in the cable ring main unit. The traditional feeder automation transformation object generally only changes a pole-mounted switch on an overhead feeder into a three-remote switch, only considers the problem of cost or reliability unilaterally, and has low practicability.

Disclosure of Invention

In view of the above, it is necessary to provide a feeder automation switch placement method and system with high practicability.

Acquiring a network topological graph of a feeder line to be distributed, searching a ring main unit, an overhead line, a plurality of regions of a ring main unit which can be modified, and acquiring the number of users in each region and the number of switches in each ring main unit according to the network topological graph;

acquiring the equivalent fault rate of each area according to the preset cable fault rate and the preset overhead line fault rate, and acquiring the power failure time of the repair stage of each area according to the preset cable repair time and the preset overhead line repair time;

establishing a dual-target distribution point planning model aiming at the minimum reconstruction cost and the minimum average power failure time according to the preset time, the power failure time of the repair stage of each area, the preset cost, the reconstructable switch, the ring main unit, the area, the equivalent failure rate of each area, the number of users of each area and the number of switches of each ring main unit;

solving the binocular distribution point planning model to obtain a pareto frontier formed by a plurality of solutions of the binocular distribution point planning model;

acquiring comprehensive deviation degrees of all solutions in the pareto frontier, and selecting an optimized solution from a plurality of solutions of the pareto frontier according to the sequence of the comprehensive deviation degrees from small to large;

and selecting a point distribution switch from the transformable switch according to the optimization solution, and generating and outputting an automatic point distribution scheme according to the point distribution switch.

A feeder automation switch placement system, comprising:

the topological graph analysis module is used for acquiring a network topological graph of the feeder line to be distributed, searching a ring main unit, an overhead line, a plurality of regions of the ring main unit, which can be modified, and acquiring the number of users in each region and the number of switches in each ring main unit;

the data calculation module is used for acquiring the equivalent fault rate of each area according to the preset cable fault rate and the preset overhead line fault rate, and acquiring the power failure time of the repair stage of each area according to the preset cable repair time and the preset overhead line repair time;

the model establishing module is used for establishing a dual-target distribution point planning model which aims at the minimum reconstruction cost and the minimum average power failure time according to the preset time, the power failure time of the repair stage of each area, the preset cost, the reconstructable switch, the ring main unit, the areas, the equivalent failure rate of each area, the number of users of each area and the number of switches of each ring main unit;

the model solving module is used for solving the binocular distribution point planning model to obtain pareto frontier formed by a plurality of solutions of the binocular distribution point planning model;

the optimization solution selection module is used for acquiring the comprehensive deviation degrees of all solutions in the pareto frontier and selecting an optimization solution from a plurality of solutions in the pareto frontier according to the sequence of the comprehensive deviation degrees from small to large;

and the scheme generating module is used for selecting a point distribution switch from the transformable switch according to the optimization solution, generating an automatic point distribution scheme according to the point distribution switch and outputting the scheme.

According to the feeder automatic switch distribution method and the feeder automatic switch distribution system, a dual-target distribution model which aims at the minimum modification cost and the minimum average power failure duration is established according to a ring main unit, an area, an overhead line and a modifiable switch of the ring main unit in a network topological graph of a feeder to be distributed, a pareto frontier is obtained by solving the dual-target distribution model, an optimal solution is selected from the pareto frontier, and a corresponding automatic distribution scheme is generated and output according to the optimal solution to finish automatic distribution. The automatic feeder switch distribution method provided by the invention not only relates to the transformation of an overhead feeder switch, but also relates to the transformation of a switch in a cable ring main unit, and simultaneously considers the cost of switch distribution transformation and the average power failure duration, and the obtained distribution scheme has the advantages of low cost, high reliability and strong practicability.

Drawings

FIG. 1 is a timing diagram of feeder automation fault handling;

FIG. 2 is a flow diagram of a method for feeder automation switch placement in one embodiment;

fig. 3 is a specific flowchart of establishing a binocular landmark point planning model with minimum modification cost and minimum average blackout duration as targets according to preset duration, blackout duration of a repair stage of each area, preset cost, a modifiable switch, a ring main unit, an area, an equivalent failure rate of each area, the number of users of each area, and the number of switches of each ring main unit in one embodiment;

FIG. 4 is a block diagram of a feeder automation switch placement system in one embodiment;

FIG. 5 is a block diagram of feeder lines in an application;

fig. 6 is a graph comparing a Pareto curve obtained by a feeder automation switch placement method and a Pareto curve obtained without the feeder automation switch placement method in an application example.

Detailed Description

There are mainly 3 types of switches involved in feeder automation:

1. the sectional breaker can be matched with a power supply outlet breaker to quickly cut off downstream faults and reduce the influence on the upstream area of the breaker.

2. And the three-remote switch can receive a control command issued by the main station and quickly switch, so as to realize remote signaling and remote measurement of fault information and remote control of the control center. And further subdivided into isolating switches and tie switches according to function.

3. The manual switch refers to an isolating switch which is not provided with an automation function, and can only be switched on and off by operators on site.

The feeder automation transformation content is to replace a manual switch with automation transformation conditions with a three-remote switch or a segmented circuit breaker, which not only relates to the automation technical transformation investment, but also directly influences the feeder automation implementation effect, namely, improves the power supply reliability. Considering the action time sequence matched with the power supply outlet circuit breaker, more than 1 sectional circuit breaker is not suitable to be arranged on the path of the load to the power supply outlet circuit breaker; as a key for automatic implementation, the interconnection switch is transformed into a three-remote switch; the switch setting of cable conductor is in the looped netowrk cabinet, and 1 way passageway is shared in a plurality of switch communication, and the looped netowrk cabinet transformation also is one of the important content of electric power feeder automation transformation promptly.

Further analyzing the influence of various switches on the power supply reliability: when a fault occurs, the sectional circuit breaker can automatically act to isolate the fault, so that the condition that the fault is diffused in a network is avoided, and the fault rate is mainly influenced; the three-remote switch can only be operated after power failure, so that the failure rate cannot be reduced, but the load point outage time can be influenced because the three-remote switch can be quickly switched on and off through remote control operation; the manual switch can isolate the fault area, and the outage time of the non-fault power failure area is greatly reduced.

In summary, the partially modified feeder automation fault handling process is as follows: a circuit breaker trip fault; isolating faults by the three-remote isolating switch; the interconnection switch realizes the recovery of the load side of the downstream three-remote switch with the fault; the manual switch further reduces the fault isolation area; the three-remote switch recovers a non-fault area; and recovering power supply after the failure is repaired. Therefore, the fault occurrence to fault repair can be divided into a plurality of stages: fault tripping phase, automatic isolation phase, manual isolation phase, sound area recovery phase and repair phase. The timing sequence is shown in fig. 1.

Referring to fig. 2, a feeder automation switch placement method in an embodiment includes the following steps.

S110: the method comprises the steps of obtaining a network topological graph of a feeder line to be distributed, searching a ring main unit, an overhead line, a plurality of regions of a ring main unit which can be modified, and obtaining the number of users in each region and the number of switches in each ring main unit.

The area refers to the collection of a plurality of interconnected elements of the power distribution network, does not contain a switching device, is composed of a switching node, a power supply node or a tail node, and is the minimum unit of fault influence.

The feeder to be distributed refers to a power grid feeder needing to be subjected to automatic transformation on a switch, and the switch can be transformed into a switch which can be transformed into a three-remote switch or a sectional breaker in the feeder. By analyzing the network topological graph of the feeder line to be distributed, a plurality of areas in the network topological graph and the number of users in each area are obtained according to the position of the switch, and the number of the switch which can be transformed, the ring main unit and the switches in the ring main unit can be searched.

S120: the method comprises the steps of obtaining the equivalent fault rate of each area according to the preset cable fault rate and the preset overhead line fault rate, and obtaining the power failure time of the repair stage of each area according to the preset cable repair time and the preset overhead line repair time.

The method for acquiring the equivalent fault rate of each area according to the preset cable fault rate and the preset overhead line fault rate specifically comprises the following steps: analyzing the cables and cables in the area, and calculating the sum of the total fault rate corresponding to all the cables and the total fault rate corresponding to all the overhead lines in the area according to the cable fault rate and the overhead line fault rate to obtain the equivalent fault rate of the corresponding area.

The method for acquiring the power failure duration of the repair stage of each area according to the preset cable repair duration and the preset overhead line repair duration specifically comprises the following steps: analyzing cables and cables in the area, calculating the sum of the total repair time corresponding to all cables and the total repair time corresponding to all overhead lines in the area according to the cable repair time and the overhead line repair time, and taking the repair time of the repair stage of the corresponding area as the power failure time of the repair stage.

S130: and establishing a dual-target distribution point planning model aiming at the minimum reconstruction cost and the minimum average power failure time according to the preset time, the power failure time of the repair stage of each area, the preset cost, the reconstructable switch, the ring main units, the areas, the equivalent failure rate of each area, the number of users of each area and the number of switches of each ring main unit.

The binocular target distribution point planning model takes the minimum reconstruction expense and the minimum average power failure duration as the binocular targets, and considers the reconstruction problem of the switch from two aspects of cost and power supply reliability.

S140: and solving the binocular distribution point planning model to obtain the pareto frontier formed by a plurality of solutions of the binocular distribution point planning model.

The pareto frontier means that a binocular labeled point planning model is solved to obtain a set of pareto optimal solutions.

S150: and acquiring the comprehensive deviation degrees of all solutions in the pareto frontier, and selecting an optimized solution from a plurality of solutions in the pareto frontier according to the sequence of the comprehensive deviation degrees from small to large.

The selected optimization solution may be one, for example, the solution with the minimum comprehensive offset is used as the optimization solution; the number of the comprehensive deviation degrees can also be multiple, for example, a preset number of the comprehensive deviation degrees are taken according to the sequence from small to large of the comprehensive deviation degrees, and the solution corresponding to the selected comprehensive deviation degrees is the optimized solution.

S160: and selecting a stationing switch from the reconfigurable switch according to the optimization solution, and generating and outputting an automatic stationing scheme according to the stationing switch.

And determining the transformable switch needing to be transformed according to the optimization solution, and using the determined transformable switch as a point distribution switch so as to generate an automatic point distribution scheme. The automatic point distribution scheme is output to a control center or a display screen for display, so that the staff can conveniently carry out automatic switch point distribution according to the point distribution scheme.

In one embodiment, the preset time duration comprises a power failure time duration of a fault tripping stage, a power failure time duration of an automatic isolation stage, a power failure time duration of a manual isolation stage, a power supply transfer time duration of the manual isolation stage and a power failure time duration of a sound area recovery stage; the preset cost comprises three-remote switch transformation cost, breaker transformation cost, communication transformation cost and ring main unit transformation cost. Referring to fig. 3, step S130 includes steps S131 to S135.

S131: and generating a cost function according to the three-remote switch transformation cost, the breaker transformation cost, the communication transformation cost and the ring main unit transformation cost, the transformable switch and the ring main unit.

The cost function is used to calculate the required cost of the automated rebuilding investment.

S132: and acquiring power failure states of the areas in a fault tripping stage, an automatic isolation stage, a manual isolation stage, a healthy area recovery stage and a repair stage respectively during fault.

The power failure state includes power failure and no power failure, and may be represented by "1" and "0", respectively, where "1" represents power failure and "0" represents no power failure, for example.

S133: and acquiring the number of the users in the power failure of each area corresponding to the fault tripping stage, the automatic isolation stage, the manual isolation stage, the sound area recovery stage and the repair stage respectively according to the power failure states of each area corresponding to the fault tripping stage, the automatic isolation stage, the manual isolation stage, the sound area recovery stage and the repair stage respectively, the number of the users in each area, the power failure time length of the fault tripping stage, the power failure time length of the automatic isolation stage, the power failure time length of the manual isolation stage, the power failure time length of the sound area recovery stage and the power failure time length of the repair stage of each area.

S134: and generating a power failure duration function according to the areas, the number of users in each area, the equivalent failure rate of each area and the number of power failure users in each area corresponding to a fault trip stage, an automatic isolation stage, a manual isolation stage, a sound area recovery stage and a restoration stage respectively.

The power outage duration function is used for calculating the average power outage duration of the user.

S135: and establishing a min function of the cost function and the power failure duration function to obtain a dual-target distribution point planning model, and generating a constraint function of the dual-target distribution point planning model according to the number of switches of the ring main unit, the transformable switches and the ring main unit.

In one embodiment, step S133 includes:

wherein u is_k,ⅠFor the blackout state of the kth zone in the fault trip phase, u_k,ⅡFor the power failure state of the kth zone in the automatic isolation phase, u_k,ⅢFor the power failure state of the kth zone in the manual isolation phase, u_k,ⅣFor the power failure state of the kth zone in the recovery stage of the healthy zone, u_k,ⅤThe power failure state of the kth area in the repair stage; t is t_ⅠFor the duration of the power failure in the fault trip phase, t_ⅡFor the duration of the power outage in the automatic isolation phase, t_ⅢFor the duration of the power failure in the manual isolation phase, t_AFor the transfer of the manual isolation phase, t_ⅣDuration of power outage, t, for recovery phase of sound area_ⅤFor the duration of the power failure in the repair phase, n_kIs the number of users of the kth region, Ω_ds,aA set of downstream regions that are kth regions; v is_k,Ⅰ、ν_k,Ⅱ、ν_k,Ⅲ、ν_k,ⅣV and v_k,ⅤThe number of the users in the k area in power failure at a fault tripping stage, an automatic isolation stage, a manual isolation stage, a sound area recovery stage and a repair stage is respectively.

The derivation processes of equations (7) to (11) are as follows.

And (3) fault tripping stage: after the area has a fault, the sectional breaker trips out of the fault before the three-remote switch, the area where the fault point is located and the area downstream of the fault point have power failure, and the area upstream of the sectional breaker normally supplies power. Let region k and q be the fault point and its upstream neighbors, i.e. region q is upstream of fault region k, switch s links k and q. According to the fault diffusion characteristic, the condition that the switch s is modified into the breaker is related to the power failure state of the region k and the region q by 4 types:

1) the switch s is not transformed into a sectional breaker, and the region k and the region q are both in a power failure state;

2) the switch s is transformed into a sectional breaker, and the downstream area k has power failure; the upstream area q has not been powered off;

3) the switch s is transformed into a sectional breaker, the downstream switch of the area k isolates the fault, and the area k and the area q are both in an uninterrupted state;

4) the switch s is not transformed into a sectionalizer, the downstream switches of the zone k isolate faults, and the zone k and the zone q are in an uninterruptible state.

The truth table of the above relationship is shown in table 1.

TABLE 1

Serial number	u_k	Y_s	u_q
				1	1	0	1
2	1	1	0
				3	0	1	0
4	0	0	0

The model for the column write fault trip phase is as follows:

u_k,Ⅰ＝1,k∈Ω_f∪Ω_ds(12)；

u_k,Ⅰ≥u_q,Ⅰ,k,q∈Ω_f∪Ω_us(13)；

Y_s≥u_k,Ⅰ-u_q,Ⅰ,k,q∈Ω_f∪Ω_us(14)；

Y_s≤1-u_k,Ⅰ+1-u_q,Ⅰ,k,q∈Ω_f∪Ω_us(15)；

in the formula u_k,ⅠAnd u_q,ⅠRespectively showing the power failure states of the areas k and q in the fault tripping stage, wherein 1 shows power failure, and 0 shows that power failure does not exist; omega_f、Ω_dsAnd Ω_usRepresenting the set of zones at the point of failure, downstream of the failure, and upstream of the failure, respectively. Formula (12) represents the power failure state of the area where the fault point is located and the area downstream of the fault point; equation (13) -equation (15) represent the power supply condition in the area upstream of the fault point.

Equation (7) can be derived from equation (12) to equation (15).

And (3) automatic isolation stage: centralized feeder automation master station location fault point, the cooperation of three remote switches of remote control and sectionalized circuit breaker isolated fault, the regional power supply of fault point upper reaches is resumeed in the reclosing tripping operation sectionalized circuit breaker, and the fault point is regional and regional still in the power failure state in lower reaches, and the regional power supply recovery condition in upper reaches relies on the position of three remote switches and sectionalized circuit breaker. According to the fault diffusion characteristics, the relationship between the modification condition of the switch s and the power failure state of the area k and the area q is similar to table 1, except that Ys is modified to Xs + Ys. Thus, the automatic isolation phase model is:

u_k,Ⅱ＝1，k∈Ω_f∪Ω_ds(16)；

u_k,Ⅱ≥u_q,Ⅱ,k,q∈Ω_f∪Ω_us(17)；

X_s+Y_s≥u_k,Ⅱ-u_q,Ⅱ,k,q∈Ω_f∪Ω_us(18)；

X_s+Y_s≤1-u_k,Ⅱ+1-u_q,Ⅱ,k,q∈Ω_f∪Ω_us(19)；

in the formula u_k,ⅡAnd u_q,ⅡRespectively showing the power failure states of the areas k and q in the automatic isolation stage. Equation (16) shows that the area where the fault point is located and the area downstream thereof are still in the power failure state; equation (17) to equation (19) represent the case where the power supply is restored in the area upstream of the failure point.

Equation (8) can be derived from equation (16) to equation (19).

And (3) manual isolation stage: and after fault positioning and automatic switch isolation in the automatic isolation stage, partial load of the upstream area of the fault point is recovered. The fault area is reduced through a manual switch which is not automatically transformed, so that the load at the upstream of the fault point is completely recovered, and the recovery of the load at the downstream of the fault point depends on whether a transfer power supply and the positions of a three-remote switch and a section breaker are available. Further analysis was done according to the configuration of the downstream switching power supply and the three remote switch:

1) the downstream area has no transfer power supply, and is in a power failure state;

2) the downstream area is provided with a transfer power supply, the downstream switches are all unmodified manual switches, and the load of the downstream area needs to be recovered after the manual isolation stage is completed; the downstream switch is provided with a modified three-remote switch, the three-remote switch s closest to the fault point is taken as a boundary, the load of the downstream area of s is recovered in the manual isolation stage, and other loads are recovered after the manual isolation stage. The manual isolation phase model is as follows:

u_k,Ⅲ＝0,k∈Ω_us(20)；

u_k,Ⅲ＝1,k∈Ω_f(21)；

u_k,Ⅲ＝1,k∈Ω_ds,na(22)；

u_k,Ⅲ≥u_q,Ⅲ,k,q∈Ω_f∪Ω_ds,a(23)；

X_s+Y_s≥u_k,Ⅲ-u_q,Ⅲ,k,q∈Ω_f∪Ω_ds,a(24)；

X_s+Y_s≤1-u_k,Ⅲ+1-u_q,Ⅲ,k,q∈Ω_f∪Ω_ds,a(25)；

in the formula u_k,ⅢAnd u_q,ⅢRespectively representing the power failure states of the areas k and q in the manual isolation stage; omega_ds,naAnd Ω_ds,aIndicating the downstream regions without and with the supply of the diverted power, respectively. Equation (22) represents the case where the downstream area has no secondary power supply; equation (23) -equation (25) represent the case where the downstream area has a transfer power source.

Equation (9) can be derived from equation (20) to equation (25).

A healthy area recovery stage: and fault minimum isolation is realized through a manual isolation stage, and power is supplied to an area which is not recovered. The model for the healthy region recovery phase is as follows:

u_k,Ⅳ＝1,k∈Ω_f(26)；

from this, equation (10) can be derived.

And (3) repairing: and the repairing stage finishes repairing the fault area, and the model is as follows:

u_p,Ⅴ＝u_p,Ⅳ(28)；

this can be inferred to obtain the formula (11).

In one embodiment, the dual-target point planning model is:

minF＝(F₁,F₂) (1)；

the cost function is:

the power outage duration function is:

the constraint function is:

wherein F1 is a cost function, F2 is a power failure duration function, X_sFor the transformation of the s-th reconfigurable switch into a three-remote switch, Y_sFor the transformation of the s-th switch into a transformation variable, Z, of the circuit breaker_cThe transformation variable of the c ring main unit is obtained; c_X、C_Y、C_CAnd C_ZRespectively modifying cost for a three-remote switch, modifying cost for a breaker, modifying cost for communication and modifying cost for a ring main unit; lambda [ alpha ]_kIs the equivalent failure rate of the kth zone, n_kThe number of users in the kth area; omega_ZAnd Ω_DRespectively a set of ring main units and a set of regions; v is_k,Ⅰ、ν_k,Ⅱ、ν_k,Ⅲ、ν_k,ⅣV and v_k,ⅤThe number of the users in the k area in power failure at a fault tripping stage, an automatic isolation stage, a manual isolation stage, a sound area recovery stage and a repair stage is respectively; d_cThe number of switches of the c-th ring main unit is shown; omega_PA set of switches through which a load passes to the mains outlet breaker.

Wherein, formula (4) shows that the manual switch is only allowed to be transformed into a three-remote switch or a sectional breaker; the formula (5) shows that the ring main unit needs to be modified when a switch in the ring main unit is modified; the formula (6) indicates that it is not preferable to provide more than 1 path from the load to the power outlet breaker.

According to the binocular landmark point distribution planning model, each target pays attention to one dimension of automatic transformation of the power distribution network, and transformation consideration is incomplete due to simple addition and solution. For this reason, a normalized normal constraint method is selected to solve the Pareto Frontier (Pareto Frontier) through research and analysis.

For convenience, the dual-target point planning model described by equation (1) -equation (28) is written in a compact form as follows:

minF＝(F₁,F₂) (29)；

st.

h(x)＝0 (30)；

wherein h (x) and g (x) represent vectors formed by equality constraints and inequality constraints,gandrespectively represent the upper and lower limits of the inequality constraint,xand

representing the upper and lower limits of the variable, respectively.

In one embodiment, the binocular landmark planning model includes a first function and a second function. The step S140 includes steps (a1) to (a 4).

Step (a 1): and respectively taking the min function taken by the first function and the min function taken by the second function as target functions, calculating the optimal solution of the target functions, and generating the anchor point of the utopia hyperplane according to the optimal solution and the target functions.

The first function being F₁The second function is F₂. Respectively taking minF1 and minF2 as objective functions and the above formulas (29) - (32) as constraint models to obtain optimal solutions x1 and x2, and obtainingThe corresponding objective functions u1 ═ (F1(x 1), F2(x 1)) and u2 ═ (F1(x 2), F2(x 2)) were obtained and used as the endpoints of the objective function solution space utopia, referred to as anchor points.

Step (a 2): and normalizing the target function according to the anchor point to obtain the normalized target function.

In order to avoid the influence of inconsistent dimensions of the target function, the target function is normalized. Defining Uutopia point vector F^UAnd the worst point vector F^N。

F_i ^U＝min[F_i(x^1*),F_i(x^2*)]＝F_i(x^i*)；

F_i ^N＝max[F_i(x^1*),F_i(x^2*)]；

Based on F^UAnd F^NNormalization processes the objective function:

where i represents the ith function.

Step (a 3): and generating a plurality of points uniformly distributed on the Utobramon hyperplane according to the preset number of distribution points, preset parameters and the normalized target function.

The step (a3) specifically includes: obtaining the Utober plane vector, selecting a normalized step length and generating uniformly distributed points on the Utober hyperplane.

If a certain normalized utopia point is taken as a reference point, and vectors from other normalized utopia points to the reference point are defined as utopia plane vectors, the utopia plane vectors are:

taking m points on the Utoxont surface vector, wherein m is a value corresponding to the number of the distribution points, and the normalization step length is as follows:

δ＝(m-1)^-1；

thus, uniformly distributed points on the meta-plane of utopia are generated, and the following are obtained:

wherein 0 is not more than α_ji≤1，

j denotes the number of evenly distributed points on the Ultrasurface of Utoban, p_j1Representing along a vector

The number of steps is increased. Each point corresponds to a solution.

Step (a 4): and acquiring the pareto front according to a plurality of points uniformly distributed on the superplane of the Utoban.

The step (a4) specifically includes: constructing a single-target optimization problem corresponding to points on the UtoPont hyperplane, and solving a corresponding Pareto solution, wherein the single-target optimization problem is as follows:

s.t.

h(x)＝0；

denormalization Pareto solution:

and screening a global Pareto solution. Because the gradient optimization algorithm may fall into a local optimum, the obtained Pareto solution may be a local Pareto solution, and therefore the obtained Pareto solution needs to be filtered and screened to obtain a global Pareto solution.

In one embodiment, step S150 includes: and acquiring the comprehensive deviation degree of each solution in the pareto frontier, and selecting the solution corresponding to the minimum comprehensive deviation degree as an optimized solution.

The dual-objective landmark planning model includes a first function and a second function. In this embodiment, obtaining the comprehensive shift degrees of each solution in the pareto frontier, and selecting a solution corresponding to the smallest comprehensive shift degree as an optimized solution includes:

wherein, Delta_j(x) Integrated degree of shift, x, for the jth solution of the pareto frontier_optIn order to optimize the solution, the method comprises the following steps of,

and (4) planning the ith function in the model for the binocular distribution point corresponding to the jth solution.

According to the feeder automatic switch stationing method, a binocular stationing planning model which aims at minimum reconstruction cost and minimum average power failure duration is established according to a ring main unit, an area, an overhead line and a reconfigurable switch of the ring main unit in a network topological graph of a feeder to be stationed, the binocular stationing planning model is solved to obtain a pareto front edge, an optimal solution is selected from the pareto front edge, and a corresponding automatic stationing scheme is generated according to the optimal solution and output to finish automatic stationing. The automatic feeder switch distribution method provided by the invention not only relates to the transformation of an overhead feeder switch, but also relates to the transformation of a switch in a cable ring main unit, and simultaneously considers the cost of switch distribution transformation and the average power failure duration, and the obtained distribution scheme has the advantages of low cost, high reliability and strong practicability.

Referring to fig. 4, a feeder automation switch placement system includes a topological graph analysis module 110, a data calculation module 120, a model building module 130, a model solving module 140, an optimized solution selection module 150, and a scenario generation module 160.

The topology map analysis module 110 is configured to obtain a network topology map of the feeder line to be distributed, search for a ring main unit, an overhead line, and a plurality of areas of the ring main unit, which can be modified, according to the network topology map, and obtain the number of users in each area and the number of switches in each ring main unit.

The data calculation module 120 is configured to obtain an equivalent fault rate of each area according to a preset cable fault rate and a preset overhead line fault rate, and obtain a power failure duration of a repair stage of each area according to a preset cable repair duration and a preset overhead line repair duration.

The model establishing module 130 is configured to establish a dual-target distribution point planning model targeting minimum reconstruction cost and minimum average power failure time according to preset time, power failure time of a repair stage of each area, preset cost, the reconstructable switch, the ring main unit, the area, equivalent failure rate of each area, user number of each area, and switch number of each ring main unit.

The model solving module 140 is configured to solve the binocular labeled point planning model to obtain a pareto frontier formed by a plurality of solutions of the binocular labeled point planning model.

The optimization solution selection module 150 is configured to obtain a comprehensive shift degree of each solution in the pareto frontier, and select an optimization solution from the solutions in the pareto frontier according to a descending order of the comprehensive shift degrees.

The scheme generating module 160 is configured to select a point distribution switch from the reconfigurable switch according to the optimization solution, and generate and output an automatic point distribution scheme according to the point distribution switch.

In one embodiment, the preset time duration comprises a power failure time duration of a fault tripping stage, a power failure time duration of an automatic isolation stage, a power failure time duration of a manual isolation stage, a power supply transfer time duration of the manual isolation stage and a power failure time duration of a sound area recovery stage; the preset cost comprises three-remote switch transformation cost, breaker transformation cost, communication transformation cost and ring main unit transformation cost. The model building module 130 includes a first function generating unit (not shown), a stage state analyzing unit (not shown), a power outage subscriber number calculating unit (not shown), a second function generating unit (not shown), and a model combining unit (not shown).

The first function generating unit is used for generating a cost function according to the three-remote switch transformation cost, the breaker transformation cost, the communication transformation cost, the looped network cabinet transformation cost, the transformable switch and the looped network cabinet.

The stage state analysis unit is used for acquiring the power failure states of the regions in a fault tripping stage, an automatic isolation stage, a manual isolation stage, a sound region recovery stage and a repair stage when the regions are in fault.

The number-of-users-during-power-failure calculation unit is used for acquiring the number of the users during power failure of each area corresponding to the fault tripping stage, the automatic isolation stage, the manual isolation stage, the sound area recovery stage and the repair stage respectively according to the power failure states of each area corresponding to the fault tripping stage, the automatic isolation stage, the manual isolation stage, the sound area recovery stage and the repair stage, the number of the users of each area, the power failure duration of the fault tripping stage, the power failure duration of the automatic isolation stage, the power failure duration of the manual isolation stage, the power failure duration of the sound area recovery stage and the power failure duration of the repair stage of each area.

The second function generation unit is used for generating a power failure duration function according to the areas, the number of users in each area, the equivalent failure rate of each area and the number of users in power failure in the fault tripping stage, the automatic isolation stage, the manual isolation stage, the healthy area recovery stage and the repair stage corresponding to each area.

The model combination unit is used for establishing a min function of the cost function and the power failure duration function to obtain a dual-target distribution point planning model, and generating a constraint function of the dual-target distribution point planning model according to the number of switches of the ring main unit, the transformable switches and the ring main unit.

In one embodiment, the dual-target point planning model includes a first function and a second function, and the model solving module 140 includes an anchor point obtaining unit (not shown), a regularization processing unit (not shown), a distribution point obtaining unit (not shown), and a point processing unit (not shown).

The anchor point acquisition unit is used for calculating the optimal solution of the target function by respectively taking the min function obtained by the first function and the min function obtained by the second function as the target function, and generating the anchor point of the utopia hyperplane according to the optimal solution and the target function.

The regularization processing unit is used for normalizing the target function according to the anchor point to obtain a normalized target function.

The distribution point acquisition unit is used for generating a plurality of points which are uniformly distributed on the Utox hyperplane according to the preset number of distribution points, preset parameters and the normalized target function.

The point processing unit is used for acquiring the pareto front according to a plurality of points uniformly distributed on the Utobond hyperplane.

In one embodiment, the optimized solution selection module 150 is configured to obtain the integrated shift degrees of the solutions in the pareto frontier, and select the solution corresponding to the smallest integrated shift degree as the optimized solution.

According to the feeder automatic switch distribution system, a binocular distribution point planning model which aims at minimum modification cost and minimum average power failure duration is established according to a ring main unit, an area, an overhead line and a modifiable switch of the ring main unit in a network topological graph of a feeder to be distributed, the binocular distribution point planning model is solved to obtain a pareto front, an optimal solution is selected from the pareto front, a corresponding automatic distribution scheme is generated according to the optimal solution and output, and the automatic distribution scheme is used for completing automatic distribution. The automatic feeder switch distribution method provided by the invention not only relates to the transformation of an overhead feeder switch, but also relates to the transformation of a switch in a cable ring main unit, and simultaneously considers the cost of switch distribution transformation and the average power failure duration, and the obtained distribution scheme has the advantages of low cost, high reliability and strong practicability.

To demonstrate the effectiveness of the present invention, a specific application example is described in detail below. An actual overhead-cable mixed feeder line is adopted, and a power grid which contains 17 users, 1 standby power supply, 8 column switches and 6 ring main units and can be used for automatic transformation of the feeder line is used as a sample.

The actual power grid aerial-cable mixed feeder comprises 17 users, 1 standby power supply, 8 column switches and 6 ring main units, and can be used for automatic transformation of the feeder. The switch divides the feeder into 24 areas, the partition map of which is shown in fig. 5, and the partition information is shown in table 2. The parameters involved in the model are listed in table 3, in conjunction with field data and literature. In Table 3,. lambda._cAnd gamma_cIndicating cable failure rate and cable repair duration, lambda_wAnd gamma_wOverhead line failure rate and overhead line repair duration.

TABLE 2

TABLE 3

Parameter(s)	Numerical value	Parameter(s)	Numerical value
				λ_c/(km·a)^-1	0.01	t_I/h	8e-5
γ_c/h	8.0	t_Ⅱ/h	0.08
				λ_w/a^-1	0.04	t_Ⅲ/h	0.50
γ_w/h	4.0	t_A/h	0.10
				C_XTen thousand yuan	2.0	C_CTen thousand yuan	2.0
C_YTen thousand yuan	5.0	C_ZTen thousand yuan	3.0

And setting the number m of uniformly distributed points on the meta-plane of utopia to be 30, and finally obtaining a Pareto curve formed by 12 solutions through 17s by adopting the above dual-target point distribution planning model and the solving method, as shown in a curve 1 of fig. 6. Table 4 lists the three scenarios with the lowest overall offset and their average outage duration, cost of capital investment and overall offset.

TABLE 4

As can be seen from table 4:

the scheme ① is an optimal compromise scheme, the average power failure duration is 3.082h (h), the cost of automatic reconstruction investment is 8 ten thousand, the average power failure duration before reconstruction is 3.153h, and the amplitude is reduced to 2.3%.

If the reliability is further improved, a scheme ② can be adopted, namely the pole-mounted switches 1-17 are transformed into three-remote switches, the pole-mounted switches 1-6 are transformed into circuit breakers, communication transformation is carried out simultaneously, and the average power failure time length after transformation is reduced by 0.001 h.

Scheme ③ is based on scheme ②, the ring network cabinet switches 23-13 are transformed into three-remote switches, the transformation investment cost is increased, but the power supply reliability cannot be further improved, because the normalized normal constraint method obtains the anchor point with the minimum average outage duration as the target, and the limit of the average outage duration of the feeder line is determined to be 3.081 h.

In order to further analyze the influence of the sectional breaker on the automatic reconstruction target, remove the model of the fault trip stage and fix the reconstruction variable Y_sSolving the dual-target point planning model yields a pareto frontier consisting of 9 solutions, as shown in fig. 6, curve 2. Table 5 lists the three schemes with the least aggregate offset and related information.

TABLE 5

Viewing fig. 6 and table 5, one can see that:

compared with the curve 1, the curve 2 moves to the upper right while changing in shape; curve 1 contains the Pareto points of curve 2, indicating that the Pareto solution obtained by curve 1 is more comprehensive and can dominate curve 2.

Scheme ① is still the optimal compromise scheme ④ improves ring network cabinet switches 23-13 into three-remote switches on the basis of scheme ①, so that the improvement investment cost is increased, but the power supply reliability cannot be further improved, because the optimal average outage duration of the feeder line is determined to be 3.082h by a normalized normal constraint method.

If only 1 switch is transformed three times, the pole-mounted switches 1-17 are preferentially transformed.

Compared with the mixed scheme and the three-remote scheme, the improved switch has the same position. It can be seen that the head end of the load concentrating branch and the area upstream near the tie switch are key locations for feeder automation reconstruction.

The practical power grid analysis and verification shows that:

1. the feeder automation switch distribution method can effectively solve the feeder automation switch distribution problem, obtain a more comprehensive decision scheme set and provide reference basis for making the distribution network automation transformation scheme.

2. The modified switch positions are the same. Therefore, the head end of the load concentration branch and the upstream area near the interconnection switch are the key positions for feeder automation transformation; the circuit breaker is distributed at the head end of the load concentration branch and is suitable for a feeder line with higher requirement on average power failure time.

The technical features of the above embodiments can be arbitrarily combined, and for the sake of brevity, all possible combinations of the technical features in the above embodiments are not described, but should be considered as the scope of the present specification as long as there is no contradiction between the combinations of the technical features.

The above examples only show some embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. A feeder automation switch placement method, comprising:

2. The feeder automation switch placement method of claim 1, wherein the preset duration includes a power outage duration of a fault trip phase, a power outage duration of an automatic isolation phase, a power outage duration of a manual isolation phase, a transfer duration of a manual isolation phase, and a power outage duration of a health area recovery phase; the preset cost comprises three-remote switch transformation cost, breaker transformation cost, communication transformation cost and ring main unit transformation cost; according to the preset duration, the power failure duration of the repair stage of each region, the preset cost, the transformable switch, the ring main unit, the region, the equivalent failure rate of each region, the number of users of each region and the number of switches of each ring main unit, a dual-target distribution point planning model which aims at the minimum transformation cost and the minimum average power failure duration is established, and the method comprises the following steps:

generating a cost function according to the three-remote switch transformation cost, the breaker transformation cost, the communication transformation cost, the looped network cabinet transformation cost, the transformable switch and the looped network cabinet;

acquiring power failure states of each area in a fault tripping stage, an automatic isolation stage, a manual isolation stage, a sound area recovery stage and a repair stage when the area is in fault;

acquiring the number of power failure time users of each region corresponding to a fault trip stage, an automatic isolation stage, a manual isolation stage, a sound region recovery stage and a repair stage respectively according to the power failure states of each region corresponding to the fault trip stage, the automatic isolation stage, the manual isolation stage, the sound region recovery stage and the repair stage respectively, the number of users of each region, the power failure time length of the fault trip stage, the power failure time length of the automatic isolation stage, the power failure time length of the manual isolation stage, the power failure time length of the sound region recovery stage and the power failure time length of the repair stage of each region;

generating a power failure duration function according to the areas, the number of users in each area, the equivalent failure rate of each area and the number of power failure time users in each area corresponding to a fault trip stage, an automatic isolation stage, a manual isolation stage, a sound area recovery stage and a repair stage respectively;

and establishing a min function of the cost function and the power failure duration function to obtain the dual-target distribution point planning model, and generating a constraint function of the dual-target distribution point planning model according to the number of switches of the ring main unit, the transformable switches and the ring main unit.

3. The method for automatically distributing the feeder switches according to claim 2, wherein the step of obtaining the number of the power failure time of each area corresponding to the fault trip stage, the automatic isolation stage, the manual isolation stage, the health area recovery stage and the restoration stage respectively according to the power failure state of each area corresponding to the fault trip stage, the automatic isolation stage, the manual isolation stage, the health area recovery stage and the restoration stage, the number of the users in the power failure state of each area, the power failure time of each fault trip stage, the power failure time of each area corresponding to the fault trip stage, the automatic isolation stage, the manual isolation stage, the health area recovery stage and the restoration stage respectively comprises:

wherein u is_k,ⅠFor the blackout state of the kth zone in the fault trip phase, u_k,ⅡIn the automatic isolation phase for the k-th zoneIn the state of power failure u_k,ⅢFor the power failure state of the kth zone in the manual isolation phase, u_k,ⅣFor the power failure state of the kth zone in the recovery stage of the healthy zone, u_k,ⅤThe power failure state of the kth area in the repair stage; t is t_ⅠFor the duration of the power failure in the fault trip phase, t_ⅡFor the duration of the power failure in the automatic isolation phase, t_ⅢFor the duration of the power failure in the manual isolation phase, t_AFor the transfer duration, t, of the manual isolation phase_ⅣFor the duration of the power outage in the recovery phase of the healthy area, t_ⅤFor the duration of the power failure of the repair phase, n_kIs the number of users of the kth region, Ω_ds,aA set of downstream regions that are kth regions; v is_k,Ⅰ、ν_k,Ⅱ、ν_k,Ⅲ、ν_k,ⅣV and v_k,ⅤThe number of the users in the k area in power failure at a fault tripping stage, an automatic isolation stage, a manual isolation stage, a sound area recovery stage and a repair stage is respectively.

4. The feeder automation switch placement method of claim 2, wherein the cost function is:

the power outage duration function is:

the dual-target distribution point planning model is as follows:

minF＝(F₁,F₂)；

the constraint function is:

wherein F1 is the cost function, F2 is the power failure duration function, X_sFor the transformation of the s-th reconfigurable switch into a three-remote switch, Y_sFor the transformation of the s-th switch into a transformation variable, Z, of a sectionalizer_cThe transformation variable of the c ring main unit is obtained; c_X、C_Y、C_CAnd C_ZRespectively modifying cost for a three-remote switch, modifying cost for a breaker, modifying cost for communication and modifying cost for a ring main unit; lambda [ alpha ]_kIs the equivalent failure rate of the kth zone, n_kThe number of users in the kth area; omega_ZAnd Ω_DRespectively a set of ring main units and a set of regions; v is_k,Ⅰ、ν_k,Ⅱ、ν_k,Ⅲ、ν_k,ⅣV and v_k,ⅤThe number of the users in the k area in power failure at a fault tripping stage, an automatic isolation stage, a manual isolation stage, a sound area recovery stage and a repair stage is respectively; d_cThe number of switches of the c-th ring main unit is shown; omega_PA set of switches through which a load passes to the mains outlet breaker.

5. The feeder automation switch placement method of claim 1, wherein the dual target placement planning model comprises a first function and a second function, and the solving the dual target placement planning model to obtain pareto frontiers formed by a plurality of solutions of the dual target placement planning model comprises:

respectively taking a min function taken by a first function and a min function taken by a second function as target functions, calculating an optimal solution of the target functions, and generating anchor points of the utopia hyperplane according to the optimal solution and the target functions;

normalizing the target function according to the anchor point to obtain a normalized target function;

generating a plurality of points uniformly distributed on the Utobramon hyperplane according to the number of preset distribution points, preset parameters and a normalized target function;

and acquiring the pareto front according to a plurality of points uniformly distributed on the superplane of the Utoban.

6. The feeder automation switch placement method of claim 1, wherein the obtaining a composite degree of shift for each solution in the pareto frontier, and the selecting an optimized solution from the plurality of solutions for the pareto frontier in order of the composite degree of shift from small to large comprises:

and acquiring the comprehensive deviation degrees of all solutions in the pareto frontier, and selecting the solution corresponding to the minimum comprehensive deviation degree as the optimized solution.

7. The feeder automation switch placement method of claim 6, wherein the dual-target placement planning model includes a first function and a second function, the obtaining of the integrated skewness of each solution in the pareto frontier, and the selecting of the solution corresponding to the smallest integrated skewness as the optimized solution includes:

wherein, Delta_j(x) Integrated degree of shift, x, for the jth solution of the pareto frontier_optIn order to solve the above-mentioned optimization,and (4) planning the ith function in the model for the binocular distribution point corresponding to the jth solution.

8. A feeder automation switch placement system, comprising:

9. The feeder automation switch placement system of claim 8 wherein the preset durations include a blackout duration of a fault trip phase, a blackout duration of an automatic isolation phase, a blackout duration of a manual isolation phase, a transfer duration of a manual isolation phase, and a blackout duration of a health area recovery phase; the preset cost comprises three-remote switch transformation cost, breaker transformation cost, communication transformation cost and ring main unit transformation cost; the model building module comprises:

the first function generating unit is used for generating a cost function according to the three-remote switch transformation cost, the breaker transformation cost, the communication transformation cost, the looped network cabinet transformation cost, the transformable switch and the looped network cabinet;

the stage state analysis unit is used for acquiring power failure states of the regions in a fault tripping stage, an automatic isolation stage, a manual isolation stage, a sound region recovery stage and a repair stage when the regions are in fault;

the power outage time household number calculation unit is used for acquiring the power outage time household numbers of the areas corresponding to the fault tripping stage, the automatic isolation stage, the manual isolation stage, the sound area recovery stage and the restoration stage respectively according to the power outage states of the areas corresponding to the fault tripping stage, the automatic isolation stage, the sound area restoration stage and the restoration stage respectively, the number of users of the areas, the power outage duration of the fault tripping stage, the power outage duration of the automatic isolation stage, the power outage duration of the manual isolation stage, the sound area restoration stage and the restoration stage;

the second function generation unit is used for generating a power failure duration function according to the areas, the number of users in each area, the equivalent fault rate of each area and the number of users in power failure in the fault tripping stage, the automatic isolation stage, the manual isolation stage, the healthy area recovery stage and the repair stage corresponding to each area;

and the model combination unit is used for establishing a min function of the cost function and the power failure duration function to obtain the dual-target distribution point planning model, and generating a constraint function of the dual-target distribution point planning model according to the number of switches of the ring main unit, the transformable switches and the ring main unit.

10. The feeder automation switch placement system of claim 8, wherein the dual objective placement plan model comprises a first function and a second function, the model solving module comprising:

the anchor point acquisition unit is used for calculating the optimal solution of the target function by respectively taking the min function obtained by the first function and the min function obtained by the second function as target functions, and generating the anchor point of the Utobramon hyperplane according to the optimal solution and the target functions;

the regularization processing unit is used for normalizing the target function according to the anchor point to obtain a normalized target function;

the distribution point acquisition unit is used for generating a plurality of points which are uniformly distributed on the Utobramon hyperplane according to the preset number of distribution points, preset parameters and a normalized target function;

and the point processing unit is used for acquiring the pareto front according to a plurality of points uniformly distributed on the Utobond hyperplane.