CN116090840A

CN116090840A - Power distribution network toughness improving method, device, equipment and medium based on energy storage planning

Info

Publication number: CN116090840A
Application number: CN202211239077.5A
Authority: CN
Inventors: 奚鑫泽; 邢超; 李胜男; 马红升; 覃日升; 许守东
Original assignee: Electric Power Research Institute of Yunnan Power Grid Co Ltd
Current assignee: Electric Power Research Institute of Yunnan Power Grid Co Ltd
Priority date: 2022-10-11
Filing date: 2022-10-11
Publication date: 2023-05-09

Abstract

The invention discloses a method, a device, equipment and a medium for improving toughness of a power distribution network based on energy storage planning. Then, based on a random fault scene set, constructing a two-stage robust optimization model by taking the minimum sum of the total energy storage investment cost and the total load loss cost under extreme weather of the power distribution network as a planning target; and converting the matrix form of the two-stage robust optimization model into a main problem model and a sub-problem model, and iteratively solving the main problem model and the sub-problem model to obtain a target energy storage planning result of the power distribution network. The target energy storage planning result includes a planned location distribution, capacity, and power for the energy storage device. In an actual scene, if the energy storage system is invested and built based on the determined position distribution, capacity and power, the capacity of the power distribution network for coping with extreme weather disasters can be effectively enhanced, and meanwhile, the cost of toughness improvement can be reduced.

Description

Power distribution network toughness improving method, device, equipment and medium based on energy storage planning

Technical Field

The invention relates to the technical field of power distribution networks, in particular to a power distribution network toughness improving method, device, equipment and medium based on energy storage planning.

Background

Existing toughness promotion measures can be broadly divided into three categories based on the point in time at which the measure is taken, including: precautions before extreme weather occurs; scheduling operation control measures of the power distribution network in extreme weather; and (5) recovering measures of the power distribution network after extreme weather. Whereas existing toughness promotion methods are primarily short-term in terms of scheduled operation, post-fault recovery, and short-term precautions (e.g., short-term component reinforcement). However, these existing methods have too high economic cost and still have a large optimization space.

Disclosure of Invention

Based on the above, it is necessary to provide a method, a device, equipment and a medium for improving toughness of a power distribution network based on energy storage planning, so as to solve the problem of high economic cost of the existing method.

A method for improving toughness of a power distribution network based on energy storage planning, the method comprising:

constructing a random fault scene set of the power distribution network in extreme weather;

based on the random fault scene set, constructing a two-stage robust optimization model by taking the minimum sum of the total energy storage investment cost and the total load loss cost of the power distribution network in extreme weather as a planning target;

Converting the matrix form of the two-stage robust optimization model into a main problem model and a sub-problem model, and iteratively solving the main problem model and the sub-problem model to obtain a target energy storage planning result of the power distribution network; the target energy storage planning result comprises position distribution, capacity and power planned for the energy storage equipment.

In one embodiment, the two-stage robust optimization model is constructed as follows:

min(C _annual,ESS +βmax min P _annual,Loss )

in the above, C _annual,ESS To the total annual energy storage investment cost, beta max min P _annual,Loss For the total cost of load loss, beta is the annual load loss and economic conversion factor, P _annual,Loss Is a annual load loss in extreme weather.

In one embodiment, the matrix form of the two-stage robust optimization model is:

in the above formula, H is a decision set of energy storage planning, x is a decision variable of a main problem, S is a random fault scene set, S is any random fault scene in S, F is a system variable decision set under the condition of energy storage planning and random fault scene determination, y1 and y2 are decision variables of sub problems, a, b1 and b2 are coefficient vectors of corresponding decision variables respectively, A, B, C, D, D2, E, F1 and F2 are coefficient matrixes, and D, e and F are constant vectors in constraint;

The main problem model after the matrix form of the two-stage robust optimization model is converted is as follows:

in the above formula, η is an auxiliary variable added to the main problem and is used for indicating the result of the sub-problem; m indicates the mth iteration, n is the total iteration number, y ₁ ^m And y ₂ ^m S is a variable added in the mth iteration process ^m To be the negative of the loss found in the sub-problem during the mth iterationThe most loaded random fault scenario.

The sub-problem model after the matrix form of the two-stage robust optimization model is converted is as follows:

in the above, x ^* And solving the decision variables of the obtained main problem.

In one embodiment, the iteratively solving the main problem model and the sub-problem model to obtain a target energy storage planning result of the power distribution network includes:

initializing the optimal solution lower bound of the main problem as negative infinity, the optimal solution upper bound of the sub problem as positive infinity, and the iteration number m as 0;

in the mth iteration, a first optimal solution of the main problem model is obtained, and the optimal solution lower bound is updated based on the first optimal solution;

updating an energy storage planning result according to a decision variable x of a main problem when the first optimal solution is obtained;

solving a second optimal solution of the sub-problem model, and updating the optimal solution upper bound based on the second optimal solution;

Judging whether the upper and lower boundary difference values between the upper boundary of the optimal solution and the lower boundary of the optimal solution are smaller than a preset difference value threshold, if the upper and lower boundary difference values between the upper boundary of the optimal solution and the lower boundary of the optimal solution are larger than or equal to the preset difference value threshold, updating a random fault scene with the largest loss load, enabling m=m+1, and returning to execute the step and the subsequent steps of solving the first optimal solution of the main problem model when the mth iteration is executed, and updating the lower boundary of the optimal solution based on the first optimal solution;

and if the difference value between the upper boundary of the optimal solution and the lower boundary of the optimal solution is smaller than a preset difference value threshold, outputting the currently updated energy storage planning result as the target energy storage planning result.

In one embodiment, the constructing a random fault scenario set of the power distribution network in extreme weather includes:

partitioning a power distribution network to obtain a plurality of power distribution network areas of the power distribution network;

acquiring an initial fault scene set of the power distribution network, and constructing a feature matrix of the initial fault scene set according to sampling data of a power distribution network area at a plurality of moments; the numerical value of the sampling data is used for indicating whether a power distribution network area contains fault loads at the sampling moment, and the feature matrix of the initial fault scene set contains feature matrices of a plurality of initial fault scenes;

Calculating the similar distance between the first fault scene and the second fault scene according to the feature matrix of the first fault scene and the feature matrix of the second fault scene to obtain the similar distance between any two initial fault scenes in the initial fault scene set; wherein the first fault scenario and the second fault scenario are any two of the plurality of initial fault scenarios;

determining a third fault scene and a fourth fault scene, and calculating the minimum probability distance of the third fault scene according to the scene probability of the third fault scene and the similarity distance between the third fault scene and the fourth fault scene so as to obtain the minimum probability distance of each initial fault scene in the initial fault scene set; the third fault scene is any one of the plurality of initial fault scenes, and the fourth fault scene is a fault scene with the smallest similarity distance with the third fault scene in the initial fault scene set;

determining a scene to be removed, and removing the scene to be removed from the initial fault scene set to obtain an updated initial fault scene set; the minimum probability distance of the scene to be removed is the minimum value of the minimum probability distances of all the initial fault scenes;

Judging whether the number of the initial fault scenes in the updated initial fault scene set is equal to a preset threshold value, if the number of the initial fault scenes in the updated initial fault scene set is greater than the preset threshold value, merging the minimum probability distance between the scene to be removed and the fault scene with the minimum similar distance, and returning to execute the step of determining the scene to be removed and the subsequent steps;

and if the number of the initial fault scenes in the updated initial fault scene set is equal to a preset threshold value, taking the updated initial fault scene set as the random fault scene set.

In one embodiment, the method further comprises:

constraining parameters within the main problem model and the sub-problem model based on acquired energy storage constraints during extreme weather durations; the energy storage constraint during the duration of extreme weather comprises a power distribution network installation energy storage quantity constraint, an energy storage running state constraint, an energy storage charging and discharging power constraint, a node installation energy storage capacity and rated power constraint, an energy storage electric quantity balance constraint, a load removal quantity constraint and a load removal duration constraint.

In one embodiment, the method further comprises:

Constraining parameters in the main problem model and the sub problem model based on the acquired power distribution network operation constraint; the operation constraint of the power distribution network comprises active power balance constraint, reactive power balance constraint, line connection state constraint and node voltage constraint of nodes of the power distribution network.

A power distribution network toughness improving device based on energy storage planning, the device comprising:

the random fault scene set construction module is used for constructing a random fault scene set of the power distribution network in extreme weather;

the two-stage robust optimization model construction module is used for constructing a two-stage robust optimization model by taking the minimum sum of the total energy storage investment cost and the total load loss cost under extreme weather of the power distribution network as a planning target based on the random fault scene set;

the target energy storage planning result determining module is used for converting the matrix form of the two-stage robust optimization model into a main problem model and a sub-problem model, and iteratively solving the main problem model and the sub-problem model to obtain a target energy storage planning result of the power distribution network; the target energy storage planning result comprises position distribution, capacity and power planned for the energy storage equipment.

A computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the steps of a power distribution network toughness promotion method based on energy storage planning as described above.

The utility model provides a distribution network toughness promotion equipment based on energy storage planning, includes memory and processor, the memory stores the computer program, the computer program is executed by the processor, makes the processor carry out the step of the distribution network toughness promotion method based on energy storage planning as above.

The invention provides a method, a device, equipment and a medium for improving toughness of a power distribution network based on energy storage planning. Then, based on a random fault scene set, constructing a two-stage robust optimization model by taking the minimum sum of the total energy storage investment cost and the total load loss cost under extreme weather of the power distribution network as a planning target; and converting the matrix form of the two-stage robust optimization model into a main problem model and a sub-problem model, and iteratively solving the main problem model and the sub-problem model to obtain a target energy storage planning result of the power distribution network. The target energy storage planning result comprises position distribution, capacity and power planned for the energy storage equipment. In an actual scene, if the energy storage system is invested and built based on the determined position distribution, capacity and power, the capacity of the power distribution network for coping with extreme weather disasters can be effectively enhanced, and meanwhile, the cost of toughness improvement can be reduced.

Drawings

In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Wherein:

FIG. 1 is a flow chart of a method for improving toughness of a power distribution network based on energy storage planning in one embodiment;

FIG. 2 is a schematic flow diagram of constructing a random set of failure scenarios for a power distribution network in extreme weather in one embodiment;

FIG. 3 is a flow chart of iteratively solving a main problem model and a sub-problem model to obtain a target energy storage planning result for a power distribution network in one embodiment;

FIG. 4 is a schematic diagram of a distribution network and typhoon lines in one embodiment;

fig. 5 is an energy storage planning and extreme scenario diagram of the power distribution network at β=50 in one embodiment;

fig. 6 is an energy storage plan and extreme scene graph of the power distribution network at β=80 in one embodiment

Fig. 7 is an energy storage plan and extreme scene graph of the power distribution network when β=100 in one embodiment

FIG. 8 is a schematic diagram of total energy storage investment cost and load loss at different beta in one embodiment;

FIG. 9 is a schematic diagram of a layout of a reinforcement circuit in one embodiment;

FIG. 10 is a schematic structural diagram of a toughness improving device for a power distribution network based on energy storage planning in one embodiment;

fig. 11 is a block diagram of a power distribution network toughness improvement device based on energy storage planning in one embodiment.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

As shown in fig. 1, fig. 1 is a flow chart of a method for improving toughness of a power distribution network based on energy storage planning in an embodiment, where the method for improving toughness of a power distribution network based on energy storage planning in the embodiment includes the following steps:

and s101, constructing a random fault scene set of the power distribution network in extreme weather.

The random fault scene set is a set of scenes simulating random faults of each load of the power distribution network in extreme weather.

In the energy storage planning of the power distribution network energy storage system, the more the randomly generated fault scene sets are, the more accurate the planning scheme is. However, if the number of failure sets is too large, the problem size becomes too large, and the calculation amount also increases drastically. If the fault set is too small, the calculation efficiency is improved, but the accuracy of the planning scheme is poor. In order to be compatible with computational efficiency, as well as the rationality of the planning scheme, in one embodiment, as shown in FIG. 2, the following specific steps are employed:

and s1011, partitioning the power distribution network to obtain a plurality of power distribution network areas of the power distribution network.

The following three principles are mainly observed in the partitioning:

(1) Principle of electric connection partition: the lines containing the electrical connections need to be separated into the same areas.

(2) Geographic location partitioning principle: the geographically proximate lines need to be separated into the same area.

(3) Line number partitioning principle: the number of lines per area needs to be controlled within a certain range, because if the number is too large, the partitioning will be too complex.

The distribution network is partitioned based on the three above-mentioned principles, here assumed to be divided into K distribution network areas (K > 1).

s1012, acquiring an initial fault scene set of the power distribution network, and constructing a feature matrix of the initial fault scene set according to sampling data of the power distribution network area at a plurality of moments.

The numerical value of the sampling data is used for indicating whether a power distribution network area contains fault loads at the sampling moment, the feature matrix of the initial fault scene set contains feature matrices of a plurality of initial fault scenes, and the construction mode of the feature matrix of each initial fault scene is consistent.

By way of example, we assume that the total sampling duration is T, the initial set of failure scenarios is S, where the number of initial failure scenarios S is N. Then for each initial failure scenario S in S, its feature matrix is ζ _s Expressed as:

in the above, xi _s,i,j If the i-th zone of the initial fault scene s contains fault load at j time, if so, xi _s,i,j Equal to 1 otherwise ζ _s,i,j Equal to 0.

It can be understood that the same processing operation is performed on the feature matrices of all the initial fault scenes, so that the feature matrix of the initial fault scene set can be constructed.

s1013, calculating a similar distance between the first fault scene and the second fault scene according to the feature matrix of the first fault scene and the feature matrix of the second fault scene, so as to obtain a similar distance between any two initial fault scenes in the initial fault scene set.

The first fault scene and the second fault scene are any two of a plurality of initial fault scenes. Exemplary, the first fault scenario is denoted as s1 and the second fault scenario is denoted as s2, then a similar distance D between the first fault scenario and the second fault scenario _s1,s2 Expressed as:

it can be understood that the same processing is performed on all the initial fault scenes, so that the similar distance between any two initial fault scenes in the initial fault scene set can be calculated.

s1014, determining a third fault scene and a fourth fault scene, and calculating the minimum probability distance of the third fault scene according to the scene probability of the third fault scene and the similarity distance between the third fault scene and the fourth fault scene, so as to obtain the minimum probability distance of each initial fault scene in the initial fault scene set.

The third fault scene is any one of a plurality of initial fault scenes, and the fourth fault scene is a fault scene with the smallest similarity distance with the third fault scene in the initial fault scene set. Illustratively, the similar distance between the third fault scenario and the fourth fault scenario is expressed as:

in the above formula, the third fault scenario is denoted as s3, the fourth fault scenario is denoted as s4, and the similar distance between s3 and s4 is the smallest.

Here, we set the probability of each failure scenario in S to P (S). For example, P(s) =1/N, but may be set to other values in practice.

Further, the formula for calculating the minimum probability distance of the third fault scenario is expressed as:

PD _s3,min ＝P(s3)D _s3,min

In the above formula, P (s 3) is the probability of occurrence of s 3.

s1015, determining a scene to be removed, and removing the scene to be removed from the initial fault scene set to obtain an updated initial fault scene set.

The minimum probability distance of the scene to be removed is the minimum value of the minimum probability distances of all the initial fault scenes. That is, the scene to be removed is expressed as:

and S5 is the scene with the highest similarity in the initial fault scene set, and the updated initial fault scene set can be obtained by removing S5 from the initial fault scene set S.

s1016, determining whether the number of the initial fault scenes in the updated initial fault scene set is equal to a preset threshold, and if the number of the initial fault scenes in the updated initial fault scene set is greater than the preset threshold, executing s1017. If the number of initial failure scenarios in the updated initial failure scenario set is equal to the preset threshold, s1018 is executed.

s1017, merging the scene to be removed and the minimum probability distance of the fault scene with the minimum similar distance, and returning to execute s1015 and subsequent steps.

If s1017 is performed, that is, the scene reduction is not completed, the scene reduction is still performed. The fault scene with the smallest similarity distance with the scene to be removed is expressed as:

And the probability distance is combined as follows:

P(s6)＝P(s5)+P(s6)

this allows s6 to inherit the minimum probability distance of s5 after s5 is deleted.

s1018, taking the updated initial fault scenario set as a random fault scenario set.

Here, if s1018 is performed, that is, the goal of scene cut is completed. At this time, the updated initial fault scene set is used as a random fault scene set.

Therefore, in the above embodiment, the scene with higher similarity in the initial fault scene set can be deleted, and the required random fault scene set can be obtained. Thus, the calculation scale is reduced, and the rationality of planning is ensured.

s102, based on a random fault scene set, constructing a two-stage robust optimization model by taking the minimum sum of the total energy storage investment cost and the total load loss cost under extreme weather of the power distribution network as a planning target.

Specifically, the two-stage robust optimization model constructed here is expressed as:

min(C _annual,ESS +βmax min P _annual,Loss )

For ease of understanding, the two-stage robust optimization model is further explained below: the two-stage robust optimization model is composed of three layers, wherein the first layer is the min of the outermost layer, the sum of the total energy storage investment cost and the total load loss cost of the power distribution network in extreme weather can be guaranteed to be minimum, and the tendency between the total energy storage investment cost and the total load loss cost can be adjusted through beta. The second layer is max, and the worst load loss is considered in the sampled random fault scene set. And the third layer is min, and in each random fault scene, the least weighted load loss is obtained by an energy storage and power supply mode. Such a total constitutes a two-stage robust optimization model of three layers "min-max-min".

Wherein the total cost of energy storage investment C _annual,ESS Mainly comprises three aspects: the equipment cost of energy storage, the site cost of energy storage and the initial operation maintenance cost are expressed as:

in the above, C _{ESS,equipment} For the equipment cost of energy storage, B is the total node set of the power distribution network, c _p Cost coefficient of energy storage per unit rated power, P _ESS,j Rated power for storing energy for jth node c _E Cost factor of energy storage unit capacity, E _ESS,j For the energy storage capacity of the jth node, C _ESS,site Site cost for energy storage c _j For the site cost coefficient, sigma, of the jth node _j An integer variable for whether the j-th node contains energy storage, wherein 1 represents that the energy storage is contained, 0 represents that no energy storage is contained, C _ESS,om Maintenance cost for initial operation r _ESS To be the discount rate, N _ESS For the maximum service life of energy storage, c _om Is a single sheetAnnual operating maintenance costs required for energy storage at rated power.

max min P _annual,Loss Firstly, obtaining a random fault scene set through Monte Carlo simulation, then calculating the minimum loss load of each random fault scene in the random fault scene set, selecting the maximum weighted loss load number with the maximum loss load number as the maximum weighted loss load number in extreme weather, and finally multiplying the maximum weighted loss load number by the number NL of the extreme weather per year, wherein the calculation formula is expressed as follows:

And S103, converting the matrix form of the two-stage robust optimization model into a main problem model and a sub-problem model, and iteratively solving the main problem model and the sub-problem model to obtain a target energy storage planning result of the power distribution network.

The target energy storage planning result comprises position distribution, capacity and power planned for the energy storage device.

Here, however, the matrix form of the two-stage robust optimization model is expressed as:

in the above formula, H is a decision set of energy storage planning, that is, all possible position distributions, capacities and powers; x is the decision variable of the main problem, namely the decision variable of the first stage; s is a random fault scene set, and S is any random fault scene in S; f is a system variable decision set under energy storage planning and random fault scene determination, and y1 and y2 are decision variables of the sub-problem, namely decision variables of a second stage; a. b1 and b2 are coefficient vectors corresponding to decision variables, A, B, C, D1, D2, E, F1 and F2 are coefficient matrices, and D, e and F are constant vectors in constraint.

The main problem is a first layer of min, namely, the energy storage position, capacity and rated power planning with optimal economy under the condition of determining a fault scene are obtained, and the obtained optimal value is an optimal solution lower limit L of the original problem. The transformed main problem model in matrix form of the two-stage robust optimization model is expressed as:

In the above formula, η is an auxiliary variable added to the main problem and is used for indicating the result of the sub-problem; m indicates the mth iteration, n is the total number of iterations,

and->

S is a variable added in the mth iteration process ^m The method is a random fault scene with the highest loss load obtained in the sub-problems in the mth iteration process.

The secondary problem is a second layer max-min, namely under the condition of determining the energy storage position, capacity and rated power planning of the power distribution network, a scene with the most serious loss load and the loss load thereof are obtained, and the obtained optimal value is an optimal solution upper bound U of the original problem. The sub-problem model after the matrix form of the two-stage robust optimization model is converted is expressed as:

Further, to facilitate solution, the sub-problem can be converted into a single-layer optimization problem by using a strong dual principle, and dual variables λ and pi are added, expressed as:

and linearizing each non-convex bilinear term of sTλ by using a large M method, expressed as:

further, considering that various constraint conditions exist in the actual scene, we set various constraint conditions for the main problem model and the sub-problem model before iteratively solving: in one particular embodiment, parameters within the main problem model and the sub-problem model are constrained based on acquired energy storage constraints during extreme weather durations; the energy storage constraint during the duration of extreme weather comprises a power distribution network installation energy storage quantity constraint, an energy storage running state constraint, an energy storage charging and discharging power constraint, a node installation energy storage capacity and rated power constraint, an energy storage electric quantity balance constraint, a load removal quantity constraint and a load removal duration constraint.

Specifically, the energy storage quantity constraint of the power distribution network is expressed as:

in the above, N _ESS,max The maximum value of the energy storage quantity can be installed for the power distribution network.

Energy storage operating state constraints, expressed as:

c _j,t +d _j,t ≤σ _j

the energy storage operating state includes three states: charge state, discharge state, inactive state. In the above, c _j,t And d _j,t Representing the working state of the energy storage equipment of the j node at the moment t, wherein the working states are 0-1 discrete variables, and c is the state of charge if the energy storage equipment works in the state of charge _j,t ＝1，d _j,t =0, if the energy storage works in the discharge state c _j,t ＝0，d _j,t =1, if the j node has no energy storage or the energy storage is in an inactive state c _j,t ＝0，d _j,t ＝0。

The energy storage charge-discharge power constraint is expressed as:

0<＝-c _j,t p _ESS,j,t +d _j,t p _ESS,j,t <＝P _ESS,j

the energy storage is in a charging or discharging state, and the charging and discharging power is smaller than the rated power. In the above, p _ESS,j,t The power at the moment of the j node t is positive in discharging and negative in charging.

Node installation energy storage capacity and rated power constraints, expressed as:

the energy storage capacity and rated power of each node should be lower than the maximum energy storage capacity and maximum rated power allowed by the node. In the above, P _ESS,j,max Maximum rated power for energy storage installed at node j is E _ESS,j,max The maximum capacity of the stored energy can be installed for node j.

The energy storage electric quantity balance constraint is expressed as:

C _S,min E _ESS,j ≤E _ESS,j,t ≤E _ESS,j

the electric quantity at the next moment of the energy storage is related to the electric quantity at the last moment and the charge and discharge state, and simultaneously, the maximum discharge depth C of the energy storage is limited for protecting the energy storage and prolonging the service life of the energy storage _S,min . In the above, E _ESS,j,t And E is _ESS,j,t+1 The energy storage residual quantity, eta of the node j at the time t and the time t+1 respectively _c And eta _d The charge and discharge efficiency of the energy storage is respectively, and delta t is the time interval in Monte Carlo sampling.

Load shedding amount constraint and load shedding duration constraint, expressed as:

when the power distribution network breaks down during extreme events, the load of the low-importance nodes can be properly reduced, the load reduction amount at each moment is smaller than the maximum reduction amount of each node, the total load reduction amount during extreme time is smaller than the total maximum reduction amount, and the load reduction time is smaller than the maximum power failure time of each node. In the above, p _cut,j,t Load amount cut off for j node at t moment, u _cut,j,t Is 0-1 discrete variable, which indicates whether the j node is subjected to load shedding at the moment t, if u _cut,j,t When=1, load shedding is performed at time t, and if u _cut,j,t And =0 indicates that no load shedding is performed at time t. P (P) _cut,j,t,max Maximum load cut-off of j node at time T, T is the duration of the extreme event, P _cut,j,T,max The sum of the maximum load amounts of the j node in the extreme event duration T.

In one particular embodiment, parameters within the main problem model and the sub-problem model are constrained based on the acquired power distribution network operational constraints. The operation constraint of the power distribution network comprises active power balance constraint, reactive power balance constraint, line connection state constraint and node voltage constraint of nodes of the power distribution network.

Specifically, the active power balance constraint and the reactive power balance constraint of the power distribution network node are expressed as:

in the above formula, delta (j) is a lower node of j nodes, pi (j) is an upper node of j nodes, and P _js,t For the active power, P, transmitted by the j node lower branch js at the moment t _L,j,t Active power predicted for j node at time t, P _rj,t For the active power transmitted by the j node upper-level branch rj at the moment of t, Q _js,t For the reactive power, Q, transmitted by the j node lower branch js at the moment t _L,j,t Reactive power predicted for j node at time t, Q _rj,t And the reactive power transmitted by the upper-level branch rj of the j node at the moment t is obtained.

Line connection status constraints, expressed as:

in the above formula, the node s and the node t are two connected nodes, V _s,t And V _r,t Voltages of s node and r node at time t, V ₀ For reference voltage, M is a large positive constant, α _sr,t Representing the break condition of the line sr at the time t, when alpha _sr,t When=1, the representative line sr is closed, when α _sr,t When=0, the representative line sr is disconnected. Wherein alpha is _sr,t The method can be set as continuous variable without setting 0-1 discrete variable, reduces the dividing number of branch delimitation, and reduces the number of the branch delimitation _sr And x _sr The resistance and reactance of the sr branch, respectively.

Node voltage constraints, expressed as:

V _j,min ≤V _j,t ≤V _j,max

in the above, V _j,min And V _j,max The maximum and minimum values of the j node voltages, respectively.

Then, in the scope of the constraint conditions, iteratively solving the main problem model and the sub-problem model to obtain a target energy storage planning result of the power distribution network, and in a specific embodiment, as shown in fig. 3, the specific steps of solving include:

s1031, initializing the optimal solution lower bound of the main problem as negative infinity, the optimal solution upper bound of the sub problem as positive infinity, and the iteration number m as 0.

I.e. let the optimal solution lower bound l= - ≡, on the optimal solution the boundary U = +++ is, m=0.

s1032, at the mth iteration, the first optimal solution of the main problem model is obtained, and the optimal solution lower bound is updated based on the first optimal solution.

That is, a first optimal solution of the main problem model is obtained by using the above formula (1), and an optimal solution lower bound L is updated based on the first optimal solution.

s1033, updating the energy storage planning result according to the decision variable x of the main problem when the first optimal solution is obtained.

And updating an energy storage planning result according to the currently solved decision variable x, wherein the energy storage planning result comprises position distribution, capacity and power planned for the energy storage equipment.

s1034, obtaining a second optimal solution of the sub-problem model, and updating an optimal solution upper bound based on the second optimal solution.

That is, a second optimal solution of the sub-problem model is obtained by using the above formulas (2) to (4), and the optimal solution upper bound U is updated based on the second optimal solution.

s1035, judging whether the difference between the upper boundary of the optimal solution and the lower boundary of the optimal solution is smaller than a preset difference threshold. If the difference between the upper boundary and the lower boundary of the optimal solution is greater than or equal to the preset difference threshold, s1036 is executed; if the difference between the upper boundary of the optimal solution and the lower boundary of the optimal solution is smaller than the preset difference threshold, s1037 is executed.

And judging whether U-L < epsilon is true or not, wherein epsilon is a preset difference threshold value.

s1036, updating the random fault scenario with the most lost load, and returning to s1032 and the subsequent steps by making m=m+1.

If the difference value between the upper boundary and the lower boundary of the optimal solution is greater than or equal to a preset difference threshold value, the two-stage robust optimization model is not converged, and at the moment, a random fault scene with the largest loss load is updated, namely, a random fault scene with the largest loss load is redetermined, so that m=m+1, and iterative solution is continued.

s1037, outputting the currently updated energy storage planning result as a target energy storage planning result.

If the difference value between the upper boundary and the lower boundary of the optimal solution is greater than or equal to a preset difference value threshold, the two-stage robust optimization model is converged, and the current updated energy storage planning result is output as a target energy storage planning result.

Therefore, according to the power distribution network toughness improving method based on energy storage planning, the random fault scene set of the power distribution network under extreme weather is constructed, so that actual extreme weather is simulated. Then, based on a random fault scene set, constructing a two-stage robust optimization model by taking the minimum sum of the total energy storage investment cost and the total load loss cost under extreme weather of the power distribution network as a planning target; and converting the matrix form of the two-stage robust optimization model into a main problem model and a sub-problem model, and iteratively solving the main problem model and the sub-problem model to obtain a target energy storage planning result of the power distribution network. The target energy storage planning result comprises position distribution, capacity and power planned for the energy storage equipment. In an actual scene, if the energy storage system is invested and built based on the determined position distribution, capacity and power, the capacity of the power distribution network for coping with extreme weather disasters can be effectively enhanced, and meanwhile, the cost of toughness improvement can be reduced.

To further illustrate the above planning method, a specific description is provided below in conjunction with a practical example. Exemplary, a schematic diagram of the distribution network and typhoon lines is shown in fig. 4. The power supply circuit comprises a primary load, a secondary load and a tertiary load, wherein each primary load is provided with dual power supply, when the power supply path of the primary load fails, the power supply can be recovered through power supply switching, and the average power supply switching time is 2h. The virtual circles in the figure represent typhoons, and the connecting lines of the circles represent typhoons routes, here assuming that typhoons are experienced 4 times per year.

The whole power distribution network area is divided into four blocks according to different urban area costs, and each area corresponds to different site costs, wherein the area 1 is 100 ten thousand yuan, the area 2 is 90 ten thousand yuan, the area 3 is 80 ten thousand yuan, and the area 4 is 70 ten thousand yuan. Assuming a lifetime of 15 years for all stored energy, the rate of failure is 10%. The maximum rated power and the capacity of the energy storage installation are respectively 350kW and 1000kWh, the cost coefficient is 680 yuan/kW and 2380 yuan/(kWh), and the annual operation and maintenance cost of the energy storage unit power is 40 yuan/(kWa). The maximum value of the energy storage installation quantity of the power distribution network is 7, the charging and discharging efficiency is 0.9, the maximum discharging depth of the energy storage is 0.1, and the initial storage electric quantity is 0.8.

Further, different beta values are set, wherein the values are 50, 80 and 100 respectively, and energy storage plans of the power distribution network under different conditions can be obtained. The most serious fault conditions of energy storage planning positions and load losses under different beta values are shown in figures 5, 6 and 7. Fig. 5 is an energy storage planning and extreme scene diagram of the power distribution network when β=50, where the extreme scenes in the diagram are L6, L11, L13, L18, L22, and L25 are failed, and the installation nodes (location distribution of energy storage devices) are 7, 24, and 28. Fig. 6 is an energy storage planning and extreme scene diagram of the power distribution network when β=80, wherein the extreme scenes in the diagram are that L6, L11, L21, L24, L25, and L29 have faults, and the installation nodes are 6, 11, 24, and 28. Fig. 7 is an energy storage planning and extreme scene diagram of the power distribution network when β=100, where the extreme scenes in the diagram are L7, L12, L21, L24, L25, and L29 are failed, and the installation nodes are 7, 12, 20, 24, and 27.

Further, the planning results obtained by finally summarizing all the data are shown in table 1.

Table 1:

further, as shown in fig. 8, the total energy storage investment cost and the load loss at different beta values are biased to the high investment and the low load loss when beta values are increased, so that the energy storage investment is gradually increased and the load loss is gradually reduced. The effect of the rising energy storage investment on the load loss reduction is gradually reduced.

As can be analyzed from table 1 and fig. 8 above, the energy storage amount of the power distribution network plan increases gradually with increasing β. Since increasing β means an increase in economic loss due to the lost load. And simultaneously, as beta increases, the value of the objective function also increases. No matter what value is taken, because the weights of the primary load and the secondary load are higher, the planning positions of energy storage are positioned on the primary load and the secondary load, the continuous power supply of the primary load and the secondary load in extreme weather is ensured, and the economic loss can be greatly reduced. Meanwhile, a plurality of areas with adjacent primary loads are arranged, so that energy storage is generally planned, and continuous power supply in extreme weather is supported. In comparison to the planning results for four areas, although the site costs for area 4 are lower, the energy storage planning locations are generally not within that area because of the lower load importance. Region 3 is heavily loaded and of high importance, so the energy storage plan is more in this region. Region 1 sites are costly and have a low load density, so when β is small, energy storage is typically not planned. Meanwhile, the most serious fault scenario is generally that a line or a tower at the upstream of a region fails, so that the power distribution network causes larger loss.

Corresponding to this solution, it is assumed that the cost of the reinforced distribution rod is 45000 yuan/rod. Taking beta as 80, and adopting a robust solving algorithm to obtain the optimal result of the reinforced line, wherein the reinforced line is mainly positioned at the upstream of each area of the power distribution network, and 424 electric poles are reinforced together as shown in fig. 9. A comparison of the energy storage planning results with the present invention is shown in table 2 below.

Table 2:

planning method	Cost of investment/yuan	Load loss/kWh	Total cost per unit of load loss	Objective function/element
					Energy storage planning	8776945	70488	5639040	14415985
Reinforced circuit	19035000	57863	4629040	23664040

Therefore, in the practical application scene, the result of the objective function of the energy storage planning method is 14415985 yuan, compared with the result of the objective function of the reinforced line which is 23664040 yuan, the cost can be effectively reduced, and the energy storage can participate in peak clipping and valley filling of the power distribution network in non-extreme weather. Therefore, the energy storage planning method provided by the invention can effectively reduce the load loss in extreme weather and improve the toughness of the power distribution network under the condition of lower investment cost.

In one embodiment, as shown in fig. 10, a power distribution network toughness improving device based on energy storage planning is provided, and the device includes:

the random fault scene set construction module 1001 is used for constructing a random fault scene set of the power distribution network in extreme weather;

the two-stage robust optimization model construction module 1002 is configured to construct a two-stage robust optimization model based on a random fault scene set, with a minimum sum of energy storage investment total cost and load loss total cost of the power distribution network in extreme weather as a planning target;

the target energy storage planning result determining module 1003 is configured to convert a matrix form of the two-stage robust optimization model into a main problem model and a sub-problem model, and iteratively solve the main problem model and the sub-problem model to obtain a target energy storage planning result of the power distribution network; the target energy storage planning result comprises position distribution, capacity and power planned for the energy storage equipment.

FIG. 11 illustrates an internal block diagram of a power distribution network toughness promotion device based on energy storage planning in one embodiment. As shown in fig. 11, the energy storage planning-based power distribution network toughness promotion device includes a processor, a memory, and a network interface connected by a system bus. The memory includes a nonvolatile storage medium and an internal memory. The non-volatile storage medium of the power distribution network toughness improving device based on energy storage planning stores an operating system and can also store a computer program, and when the computer program is executed by a processor, the processor can realize the power distribution network toughness improving method based on energy storage planning. The internal memory may also store a computer program that, when executed by the processor, causes the processor to perform a method for toughness promotion of the power distribution network based on energy storage planning. Those skilled in the art will appreciate that the structure shown in fig. 11 is merely a block diagram of a portion of the structure associated with the present application and is not limiting of the energy storage planning-based power distribution network toughness promotion apparatus to which the present application is applied, and that a particular energy storage planning-based power distribution network toughness promotion apparatus may include more or fewer components than shown, or may incorporate certain components, or may have a different arrangement of components.

A power distribution network toughness promotion device based on energy storage planning, comprising a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor realizes the following steps when executing the computer program: constructing a random fault scene set of the power distribution network in extreme weather; based on a random fault scene set, constructing a two-stage robust optimization model by taking the minimum sum of the total energy storage investment cost and the total load loss cost of the power distribution network in extreme weather as a planning target; and converting the matrix form of the two-stage robust optimization model into a main problem model and a sub-problem model, and iteratively solving the main problem model and the sub-problem model to obtain a target energy storage planning result of the power distribution network.

A computer readable storage medium storing a computer program which when executed by a processor performs the steps of: constructing a random fault scene set of the power distribution network in extreme weather; based on a random fault scene set, constructing a two-stage robust optimization model by taking the minimum sum of the total energy storage investment cost and the total load loss cost of the power distribution network in extreme weather as a planning target; and converting the matrix form of the two-stage robust optimization model into a main problem model and a sub-problem model, and iteratively solving the main problem model and the sub-problem model to obtain a target energy storage planning result of the power distribution network.

It should be noted that the method, the device, the equipment and the computer readable storage medium for improving the toughness of the power distribution network based on the energy storage planning belong to a general inventive concept, and the content in the embodiments of the method, the device, the equipment and the computer readable storage medium for improving the toughness of the power distribution network based on the energy storage planning can be mutually applicable.

Those skilled in the art will appreciate that implementing all or part of the above-described methods in accordance with the embodiments may be accomplished by way of a computer program stored in a non-transitory computer-readable storage medium, which when executed, may comprise the steps of the embodiments of the methods described above. Any reference to memory, storage, database, or other medium used in the various embodiments provided herein may include non-volatile and/or volatile memory. The nonvolatile memory can include Read Only Memory (ROM), programmable ROM (PROM), electrically Programmable ROM (EPROM), electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous Link DRAM (SLDRAM), memory bus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), among others.

The technical features of the above embodiments may be arbitrarily combined, and all possible combinations of the technical features in the above embodiments are not described for brevity of description, however, as long as there is no contradiction between the combinations of the technical features, they should be considered as the scope of the description.

The foregoing examples represent only a few embodiments of the present application, which are described in more detail and are not thereby to be construed as limiting the scope of the present application. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application is to be determined by the claims appended hereto.

Claims

1. The utility model provides a distribution network toughness promotes method based on energy storage planning which characterized in that, the method includes:

2. The method of claim 1, wherein the two-stage robust optimization model is constructed as:

min(C _annual,ESS +βmax min P _annual,Loss )

3. The method according to claim 2, characterized in that the matrix form of the two-stage robust optimization model is:

in the above formula, η is an auxiliary variable added to the main problem and is used for indicating the result of the sub-problem; m indicates the mth iteration, n is the total iteration number, y ₁ ^m And y ₂ ^m S is a variable added in the mth iteration process ^m The random fault scene with the most loss load is obtained in the sub-problem in the mth iteration process;

4. A method according to claim 3, wherein said iteratively solving the main problem model and the sub-problem model to obtain a target energy storage planning result for the power distribution network comprises:

5. The method of claim 1, wherein said constructing a random set of failure scenarios for the distribution network in extreme weather comprises:

6. The method according to claim 1, characterized in that the method further comprises:

7. The method according to claim 1, characterized in that the method further comprises:

8. Distribution network toughness hoisting device based on energy storage planning, characterized in that, the device includes:

9. A computer readable storage medium storing a computer program, which when executed by a processor causes the processor to perform the steps of the method according to any one of claims 1 to 7.

10. A power distribution network toughness promotion device based on energy storage planning, comprising a memory and a processor, characterized in that the memory stores a computer program which, when executed by the processor, causes the processor to perform the steps of the method according to any one of claims 1 to 7.