A flexible interconnected power distribution network collaborative scheduling method for post-disaster recovery

Abstract

The application relates to a flexible interconnected power distribution network collaborative scheduling method for post-disaster recovery, comprising the following steps: respectively establishing a time period decoupling power model of an energy storage type flexible soft switch system and a fault element repair sequence optimization model; taking the time period decoupling power model of the energy storage type flexible soft switch system, the fault element repair sequence optimization model and power distribution network topology reconstruction as pre-decision variables, constructing a collaborative optimization scheduling model considering new energy uncertainty with the lowest comprehensive cost as a target; and solving the collaborative optimization scheduling model by using a double-layer iteration extreme scene generation algorithm based on infinite dimension linear programming duality to obtain scheduling decisions. The application can fully exert a collaborative effect and improve a post-disaster load recovery ratio of a power distribution network.

Description

Technical Field

[0001] This invention relates to the field of power distribution network post-disaster recovery and dispatching technology, and in particular to a collaborative dispatching method for flexible interconnected power distribution networks oriented towards post-disaster recovery. Background Technology

[0002] Frequent extreme weather and natural disasters make power distribution network infrastructure vulnerable to damage, leading to large-scale power outages and severe losses. Flexible open points (SOPs) in energy storage-based stand-alone (E-SOP) systems, with their real-time active / reactive power regulation capabilities, have become key supporting equipment for load transfer and voltage regulation in post-disaster power distribution networks, driving the intelligent upgrade of power distribution networks. Current research has made some progress: some studies have established power models for E-SOP systems and explored their islanding recovery role; some studies have proposed recovery methods that coordinate flexible interconnection devices with traditional tie switches; and other studies have evaluated the reliability of power distribution networks after faults. However, existing technologies have core shortcomings:

[0003] 1) Lack of a coordinated mechanism between fault repair and topology reconfiguration: Existing solutions either optimize the two independently or simply adjust them alternately without considering the dynamic impact of the repair order on the topology reconfiguration path, or the feedback effect of topology reconfiguration on the repair priority, resulting in limited load recovery efficiency.

[0004] 2) Inadequate handling of uncertainties in new energy sources: Traditional stochastic optimization relies on historical distribution assumptions, which are difficult to cover all real-world scenarios and lack robustness; robust optimization ignores probability distribution information, makes overly conservative scheduling decisions, and has poor economic efficiency.

[0005] 3) Insufficient consideration of the time-series coupling characteristics of energy storage systems: The scientific decoupling of energy storage charging and discharging constraints from new energy uncertainties has not been achieved, which increases the complexity of the model and limits its practicality in engineering. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a flexible interconnected distribution network collaborative scheduling method for post-disaster recovery, which can give full play to the synergistic effect and improve the post-disaster load recovery ratio of the distribution network.

[0007] The technical solution adopted by this invention to solve its technical problem is: to provide a flexible interconnected distribution network collaborative scheduling method for post-disaster recovery, comprising the following steps:

[0008] Establish time-period decoupling power model and fault component repair sequence optimization model for energy storage flexible soft switching system respectively;

[0009] Using the time-period decoupled power model of the energy storage flexible soft switch system, the fault component repair sequence optimization model, and the distribution network topology reconfiguration as pre-decision variables, a collaborative optimization scheduling model considering the uncertainty of new energy sources is constructed with the goal of minimizing the overall cost.

[0010] The collaborative optimization scheduling model is solved using a two-level iterative extreme scenario generation algorithm based on the duality of infinite-dimensional linear programming, and the scheduling decision is obtained.

[0011] The time-period decoupling power model of the energy storage flexible soft-switching system satisfies the following constraints:

[0012] The active power balance constraint of the E-SOP system is expressed as: ,in, Let be the active power flowing into the AC side of the m-th converter. Let m be the active power loss of the m-th converter. and These are the discharge power and charging power of the s-th energy storage unit in the DC bus, respectively.

[0013] The loss constraint of the converter is expressed as: , where c m Let be the loss factor of the m-th converter. Let be the reactive power flowing into the AC side of the m-th converter;

[0014] The capacity constraint of the converter is expressed as: ,in, Let m be the apparent power capacity of the m-th converter;

[0015] The allowable boundary constraints for charging and discharging power during the energy storage pre-scheduling phase are expressed as follows: ,in, Let be the lower and upper bound variables of the allowable charging power of the s-th energy storage at time t, respectively. Let be the lower and upper bound variables of the allowable discharge power of the s-th energy storage unit at time t, respectively. All are binary variables. and These are the physical upper bounds of the discharge power and charging power of the s-th energy storage unit at time t, respectively. These are the lower and upper bounds of the allowable capacity of the s-th energy storage unit at time t, respectively. Let represent the charging efficiency and discharging efficiency of the s-th energy storage device, respectively. Represents a time variable. These represent the lower and upper bounds of the energy storage capacity of the s-th energy storage unit, respectively. For adjustment coefficients;

[0016] The allowable boundary constraints for charge and discharge power during the energy storage readjustment phase are expressed as follows: .

[0017] The loss constraint relaxation of the converter is as follows: .

[0018] The optimization model for the repair sequence of faulty components includes:

[0019] Spatial transfer constraints are expressed as: Where ℛ represents the set of locations consisting of the maintenance center and the faulty component, and ℛ\r represents the set of other locations excluding the faulty component r. This indicates whether the repair team d has moved from location s' to the location of the faulty component r. This indicates whether the repair team d has moved from the location of the faulty component r to location s. This indicates whether the repair team d passed through the location of the faulty component r. Indicate whether the emergency repair team d has passed through the maintenance center;

[0020] Temporal association constraints are expressed as: , where t r,d t is the time when the maintenance team d arrives at the location of the faulty component r. s,d Δt is the time when maintenance team d arrives at location s. r,d Δt represents the repair time taken by maintenance team d to repair the faulty component r. r,s,d The travel time for maintenance team d to move from the location of the faulty component r to location s. It is a constant. This indicates whether the faulty component r was repaired during time period t.

[0021] The objective function of the collaborative optimization scheduling model is: ,in, For the active power output of unit n, g n,t D is the power generation cost coefficient for unit n. i,t The original load of node i, The actual load after load abandonment. This is the load abandonment penalty coefficient. e represents the active power of the bus. t For electricity price, u is the predicted value of new energy k. k,t For the actual fluctuation of new energy k, For the actual contribution of new energy k, This is the penalty coefficient for losses in new energy sources.

[0022] The constraints of the cooperative optimization scheduling model are: ,in, These are the lower and upper bounds of the active power output of gas turbine n, respectively. The reactive power output of gas turbine n are respectively The lower and upper bounds, The predicted active power output of new energy source k is given. For the active power fluctuation of new energy k, The reactive power output of new energy k is respectively The lower and upper bounds, and Let represent the active load and reactive load of node i, respectively. The original active load of node i before it discards the load. For power factor, A i and B i Let p be the set of upstream nodes and the set of downstream nodes of node i, respectively. ij,t and q ij,t These represent the active and reactive power transmitted by line ij, respectively. i,t and q i,t These represent the net active power injection and net reactive power injection at node i, respectively, δ i v is a binary constant. i and v j These are the voltages at both ends of line ij, where v0 is the rated voltage and r is the voltage across the line. ij and x ij These are the resistance and reactance of line ij, respectively. Let be the on / off state variable of line ij. Let be the lower and upper bounds of the voltage at node i, respectively. These are the lower and upper bounds of the active power transmitted by line ij, respectively. These are the lower and upper bounds of the reactive power transmitted by line ij, respectively.

[0023] The collaborative optimization scheduling model adopts new energy fuzzy sets. To describe the probability distribution P of new energy sources in different time periods t The feasible region, the new energy fuzzy set Represented as: , where p(u t ) for scene u t The probability density, This represents the expected value for all scenarios. Center of expected value For the threshold, These are the lower and upper bounds of the expected value, respectively. For the uncertainty set of new energy;

[0024] The compact form of the expected operating cost of the collaborative optimization scheduling model is expressed as follows: Where c is the cost coefficient of the first stage, x is the decision of the first stage, u is the uncertainty variable, and Q is the cost coefficient of the first stage. t(x t ,u t () represents the optimal value for the second stage readjustment cost in each time period.

[0025] The method of solving the collaborative optimization scheduling model using a two-level iterative extreme scenario generation algorithm based on the duality of infinite-dimensional linear programming to obtain scheduling decisions specifically includes:

[0026] Based on the dual transformation of infinite-dimensional linear programming, the probability distribution uncertainty in the collaborative optimization scheduling model is transformed into scenario uncertainty;

[0027] By focusing on the extreme scenarios that have the greatest impact on scheduling constraints, the collaborative optimization scheduling model, after being transformed into scenario uncertainty, is simplified.

[0028] A two-layer iterative extreme scenario generation algorithm is used to solve the simplified cooperative optimization scheduling model. Specifically, the two-layer iterative extreme scenario generation algorithm includes:

[0029] Initialize the set of extreme scenarios;

[0030] Solve the cooperative optimization scheduling model after transforming it into a scenario with uncertainty, optimize the pre-scheduling decision, and obtain the lower bound of the cost;

[0031] Initialize the set of extreme scenarios for this iteration as the center of the expected value, fix the optimized pre-scheduling decision, determine the dual variable corresponding to the new energy fuzzy set, generate the upper bound of the cost, and solve the simplified collaborative optimization scheduling model to generate a cluster of extreme scenarios. Repeat this step until the target value is reached.

[0032] When the difference between the upper and lower bounds of the cost is greater than a preset value, the generated extreme scenario is added to the extreme scenario set, and the solution is returned to the collaborative optimization scheduling model after the scenario uncertainty is converted.

[0033] The technical solution adopted by the present invention to solve its technical problem is: to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the above-mentioned flexible interconnected distribution network collaborative scheduling method for post-disaster recovery.

[0034] The technical solution adopted by the present invention to solve its technical problem is: to provide a computer-readable storage medium on which a computer program is stored, wherein when the computer program is executed by a processor, the steps of the above-mentioned flexible interconnected distribution network collaborative scheduling method for post-disaster recovery are implemented.

[0035] Beneficial effects

[0036] By adopting the above-mentioned technical solutions, this invention has the following advantages and positive effects compared with the prior art: Through the coordinated optimization of fault repair and topology reconfiguration, and the improvement of power flow distribution by flexible interconnection devices, this invention can reduce load losses in the distribution network and improve the scheduling robustness under renewable energy fluctuations. Simultaneously, the model employs biblical optimization to handle renewable energy uncertainties, and the proposed two-layer iterative extreme scenario generation algorithm reduces probability distribution uncertainty to scenario uncertainty, which can reduce the number of iterations in the mixed integer master problem, resulting in high computational efficiency (only about 1 minute), and can support rapid computation in practical systems. Attached Figure Description

[0037] Figure 1 This is a flowchart of the first embodiment of the flexible interconnected distribution network collaborative scheduling method for post-disaster recovery of the present invention;

[0038] Figure 2 This is a schematic diagram of solving the collaborative optimization scheduling model in the first embodiment of the present invention. Detailed Implementation

[0039] The present invention will be further illustrated below with reference to specific embodiments. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. Furthermore, it should be understood that after reading the teachings of this invention, those skilled in the art can make various alterations or modifications to the invention, and these equivalent forms also fall within the scope defined by the appended claims.

[0040] The first embodiment of this invention relates to a collaborative scheduling method for flexible interconnected distribution networks oriented towards post-disaster recovery. This method first establishes a two-stage sub-Bruker collaborative model, then proposes a two-layer iterative extreme scenario generation algorithm to solve the model, thereby obtaining the optimal scheduling plan for distribution network recovery. In the model establishment stage, E-SOP and emergency repair sequence modeling are first performed, and collaborative constraints for emergency repair and reconfiguration are established to describe the deep coupling relationship between the two. Subsequently, a fuzzy set of the probability distribution of new energy sources is constructed to describe the feasible region of the probability distribution of new energy output, and then the above content is integrated to establish a two-stage sub-Bruker collaborative model. In the model solving stage, dimensionality reduction through infinite-dimensional duality is first used to transform the uncertainty of the probability distribution into scenario uncertainty. Then, a two-layer iterative extreme scenario generation process is entered: the inner layer, the batch scenario generation step, generates extreme scenarios for each time period in batches through sub-problems under a fixed pre-scheduling decision; while the outer layer, the optimization scheduling decision step, incorporates the generated extreme scenarios into the main problem, continuously optimizing the scheduling decision until convergence.

[0041] like Figure 1 As shown, the flexible interconnected distribution network collaborative scheduling method for post-disaster recovery in this embodiment specifically includes the following steps:

[0042] Step 1: Establish the time-period decoupling power model and the fault component repair sequence optimization model for the energy storage flexible soft switching system.

[0043] This step first establishes a time-decoupled power model for the energy storage-type flexible soft-switching system. In a flexible interconnected distribution network, each feeder is connected to a common DC bus via a converter, and the energy storage system can also be connected to the DC bus. The energy storage and converter together constitute the E-SOP system. During normal operation, the E-SOP system can adjust the active and reactive power on the tie lines to achieve optimal interconnection system efficiency. Simultaneously, after a fault, the energy storage system can also act as a backup power source, enabling load transfer. The E-SOP system must satisfy active power balance constraints, loss constraints, and capacity constraints, namely:

[0044] (1)

[0045] (2)

[0046] (3)

[0047] Wherein, equation (1) is the active power balance constraint of the E-SOP system, Let be the active power flowing into the AC side of the m-th converter. This represents the active power loss of the m-th converter. and Let c be the discharge power and charging power of the s-th energy storage unit in the DC bus, respectively. Equation (2) represents the loss constraint of converter m, and c is the power of the s-th energy storage unit in the DC bus. m Its loss coefficient, This represents the reactive power flowing into the AC side. Equation (3) represents the capacity constraint of converter m. Its apparent power capacity.

[0048] In this embodiment, equation (2) is relaxed to:

[0049] (4)

[0050] Since the lower the loss, the higher the operating efficiency of the flexible interconnected distribution network, the above-mentioned relaxation is generally tight.

[0051] In addition, all energy storage systems (including DC bus energy storage and energy storage in AC distribution networks) must meet charging and discharging power constraints and capacity constraints. The allowable boundary constraints for charging and discharging power during the energy storage pre-dispatch phase are as follows:

[0052] (5)

[0053] (6)

[0054] (7)

[0055] (8)

[0056] (9)

[0057] (10)

[0058] (11)

[0059] (12)

[0060] (13)

[0061] Equations (5)-(8) represent the allowable and upper bound variables of the charging power of energy storage s. And the allowable and upper limits of discharge power variables. The constraints that need to be satisfied, among which, Let be the lower and upper bound variables of the allowable charging power of the s-th energy storage at time t, respectively. Let be the lower and upper bound variables of the allowable discharge power of the s-th energy storage unit at time t, respectively. All are binary variables. and Let be the physical upper bounds of the discharge power and charging power of the s-th energy storage unit at time t, respectively. Equation (8) is used to form a unique and continuous allowable boundary for the energy storage charging and discharging power, so as to avoid simultaneous charging and discharging during the readjustment phase as much as possible. Equations (10) and (11) represent the allowable boundary for the energy storage charging power and the lower and upper limits of the capacity allowance. The relationship. Equation (12) indicates that the allowable boundary of energy storage capacity cannot exceed the actual physical boundary. Equation (13) indicates that the energy storage capacity after the scheduling ends should not deviate too far from the preset value. ε is a small constant, which can be taken to be around 0.2.

[0062] During the readjustment phase, energy storage only needs to constrain its charging and discharging power within permissible limits:

[0063] (14)

[0064] Then, an optimization model for the repair sequence of faulty components is established. Optimizing the repair sequence of faulty components is a typical path planning problem. Repair teams start from the maintenance center, repair each faulty component step by step according to a certain repair sequence, and finally return to the maintenance center. The maintenance center and the faulty components together constitute the location set ℛ, where the index r=0 represents the maintenance center, and r≥1 represents the faulty component. The set of other locations excluding r is denoted as ℛ\r. The distribution network has multiple repair teams, each with an index d, and the team set is... This implementation defines two sets of binary variables to represent the stay and transfer status of each maintenance team at various locations, where the stay variable x s,d The value of x indicates whether the repair team d passed through location s; 1 indicates yes, 0 indicates no. The transfer variable x... r,s,d This indicates whether the emergency repair team d has moved from location r to location s; 1 indicates yes, 0 indicates no.

[0065] The above-mentioned residence variable x s,d With the transfer variable x r,s,d Satisfy spatial transfer constraints:

[0066] (15)

[0067] (16)

[0068] (17)

[0069] Equation (15) indicates that maintenance team d can only enter from one location to location r and move from location r to another location. Equation (16) indicates that each maintenance team passes through the maintenance center. Equation (17) indicates that each faulty component can only be repaired by one emergency repair team.

[0070] Let the time taken by the repair team to repair component r be Δt. r,d The transfer (i.e. travel) time from element r to s is Δt. r,s,d Then the arrival times of the maintenance team at components r and s satisfy the timing correlation constraint:

[0071] (18)

[0072] (19)

[0073] In the formula, the variable t r,d Let d be the time when the maintenance team arrives at the faulty component r, and let Δt be the maintenance time. r,d and travel time Δt r,s,d M is a constant, and M is a very large constant.

[0074] Define binary variable z r,tThis indicates whether the faulty component r was repaired within time period t; 1 indicates yes, 0 indicates no. Since each component is repaired within a single time period, then:

[0075] (20)

[0076] If the repair team leaves immediately after fixing the fault, then the time when the repair team d leaves component r is related to the time of the aforementioned repair of the binary variable z. r,t satisfy:

[0077] (twenty one)

[0078] Distribution network topology reconfiguration controls the on / off states of each line to change the topology. The on / off state α of the faulty line ij. ij,t It is related to its fault state. When the line is repaired, the line is allowed to be open; however, when the line is not repaired, it can only be in an open state, hence the open state α. ij,t The above repair of binary variable z ij,t satisfy:

[0079] (twenty two)

[0080] Step 2: Using the time-period decoupled power model of the energy storage flexible soft switch system, the fault component repair sequence optimization model, and the distribution network topology reconfiguration as pre-decision variables, a collaborative optimization scheduling model considering the uncertainty of new energy sources is constructed with the goal of minimizing the overall cost.

[0081] The aforementioned fault component repair sequence, distribution network topology reconfiguration, and allowable boundary variables for energy storage charging and discharging power are pre-decision variables, meaning they remain unchanged after being determined and will not change during subsequent real-time adjustments. Specifically, the optimization of the fault component repair sequence is shown in equations (15)-(22), and the optimization of the allowable boundary variables for energy storage system charging and discharging power are shown in equations (5)-(13). The distribution network topology reconfiguration controls the switching variables α of each line. ij,t This ensures the system conforms to a radial network. A virtual power flow method can be used to construct the radial network constraints, forming a set of constraints regarding α. ij,t The linear constraints. The first-stage pre-decision variable x includes the energy storage allowable boundary variable. Variables for the order of repair of faulty components and topology reconstruction variable α ij,t In the second phase, after observing actual fluctuations in new energy sources, the power of each component was adjusted to ensure the safe operation of the system.

[0082] The objective function and constraints of the collaborative optimization scheduling model established in this embodiment can be expressed as:

[0083] (23a)

[0084] (23b)

[0085] (23c)

[0086] (23d)

[0087] (23e)

[0088] (23f)

[0089] (23g)

[0090] (23h)

[0091] (23i)

[0092] (23j)

[0093] (23k)

[0094] (23l)

[0095] (23m)

[0096] (23n)

[0097] In this equation (23a), the optimization objective is defined, and the first term represents the gas turbine power generation cost. For the active power output of unit n, g n,t Its power generation cost coefficient, the second term is the load curtailment penalty cost, D i,t The original load of node i, The actual load after load abandonment. The third item is the load abandonment penalty coefficient, and the fourth item is the power purchase cost of the distribution network. e represents the active power of the bus. t The first item is the electricity price, and the fourth item is the penalty cost for losses of renewable energy. u is the predicted value of new energy k. k,t For its actual fluctuations, To actually contribute to it, This represents the penalty coefficient for new energy losses. Equations (23b)-(23n) are the constraints for the readjustment stage, where equation (23b) represents the active power output of gas turbine n. and doing nothing Upper and lower bound constraints, These are the lower and upper bounds of the active power output of gas turbine n, respectively. The reactive power output of gas turbine n are respectively The lower and upper bounds. Equation (23c) represents the active power output of new energy k. and doing nothing Upper and lower bound constraints, The predicted active power output of new energy source k is given. For the active power fluctuation of new energy k, The reactive power output of new energy k is respectively The lower and upper bounds. Equation (23d) represents the nodal load constraints. and These represent the active and reactive loads at node i, respectively. This represents the original active load of node i before it discards the load. For power factor. Equations (23e) and (23f) are the active and reactive power balance constraints at each node, A i and B i Let p be the set of upstream nodes and the set of downstream nodes of node i, respectively. ij,t and q ij,t These represent the active and reactive power transmitted by line ij, respectively. i,t and q i,t These represent the net active power injection and net reactive power injection at node i, respectively, and their structures are shown in formulas (23g) and (23h), δ i The constant is a binary number; the bus node is 1, and the other nodes are 0. Constraints (23i) and (23j) together describe the voltage v across line ij. i and v j The descent relationship, v i and v j These are the voltages at both ends of line ij, where v0 is the rated voltage and r is the voltage across the line. ij and x ij Let M be a very large constant, representing the resistance and reactance of line ij respectively. The on / off state variables of line ij. Equation (23k) represents the node voltage v. i Upper and lower bound constraints, These represent the lower and upper bounds of the voltage at node i, respectively. Equations (23l) and (23m) represent the upper and lower bound constraints on the active and reactive power transmission of line ij. These are the lower and upper bounds of the active power transmitted by line ij, respectively. These represent the lower and upper bounds of the reactive power transmitted by line ij, respectively. Equation (23n) represents the constraints on the flexible interconnection device E-SOP system and the energy storage charging and discharging power. The second-stage readjustment variables include the active / reactive power output of the gas turbine, the active / reactive load of the node, the active / reactive power output of the new energy source, the energy storage charging / discharging power, the active / reactive power of the bus gate, the active / reactive power of the E-SOP system, the active / reactive power transmitted by the branch, and the node voltage.

[0098] In the second-stage readjustment problem, the new energy source is uncertain, and this implementation method considers the uncertainty of the probability distribution of the new energy source. Since the decision-making is independent in each time period of the second-stage problem, the following fuzzy set of new energy sources can be constructed. To describe the probability distribution P of new energy sources in different time periods t Feasible domain:

[0099] (twenty four)

[0100] In the formula, p(u t ) for scene u t The probability density, For the uncertainty set of new energy, i.e., the feasible domain of the scenario The first line of the above formula indicates that the sum of the probability densities of all scenarios is 1, and the second line indicates that the expected value of all scenarios is... The third line indicates that the 1-norm expectation of new energy is no more than the threshold from the center of the expected value. The fourth line indicates that the expected value of new energy is within a certain range. The fifth line indicates that the probability density is greater than 0 and all scenarios are located within the uncertainty set.

[0101] Given the uncertainties of new energy sources, the expected cost of optimized operation for disaster recovery and dispatching of flexible interconnected distribution networks can be summarized in a compact form as follows:

[0102] (25)

[0103] In the formula, c is the cost coefficient for the first stage (c=0 in this model), x is the decision for the first stage, u is the uncertainty variable, and Q is the value of Q. t (x t ,u t ) represents the optimal value of the second-stage readjustment cost for each time period, and is the value of the decision variable y. t The minimum value of the optimization problem, in compact form, is:

[0104] (26)

[0105] In the formula, U t V t W t G t,i Let a be a constant matrix. t bt d t g t It is a constant vector.

[0106] Step 3: Solve the collaborative optimization scheduling model using a two-level iterative extreme scenario generation algorithm based on the duality of infinite-dimensional linear programming to obtain the scheduling decision.

[0107] This step first addresses the distribution uncertainty.

[0108] In the proposed scheduling problem (25), the problem of the expected value of extreme distribution costs. In this problem, both the new energy scenarios and their probability densities are infinitely and continuously distributed, making it an infinite-dimensional linear programming problem, which is extremely difficult to handle. However, based on duality theory, this infinite-dimensional linear programming problem can be transformed into its dual problem. The detailed form is as follows:

[0109] (27)

[0110] In the formula, α t ,β t , , , For each row constraint, there is a dual variable.

[0111] According to the duality theory of infinite-dimensional linear programming, the dual problem of the original problem (27) is:

[0112] (28a)

[0113] st (28b)

[0114] (28c)

[0115] Equation (28c) is a robust cost constraint that must hold for any new energy scenario. That is, through dual transformation, the probability distribution uncertainty is reduced to scenario uncertainty.

[0116] Replacing the second-stage extreme distribution cost expectation problem with its dual problem, the post-disaster recovery scheduling of flexible interconnected distribution networks (25) is transformed into:

[0117] (29)

[0118] Referring to the two-stage RO, extreme scenarios for new energy can be extracted, and constraint (28c) can be transformed into a deterministic piecewise constraint:

[0119] (30)

[0120] In the formula, A collection of extreme scenarios for each time period.

[0121] The key to transforming the robust cost constraint into a deterministic constraint lies in identifying all the extreme scenarios that have the greatest impact on it, i.e., minimizing the feasibility of the constraint:

[0122] (31)

[0123] Re-adjust the optimal cost value Q t (x t ,u t ) is about the decision variable y t The minimization problem, problem (31) is actually a two-stage problem, namely the uncertainty scenario u t To minimize the feasibility of the constraints, while the decision variable y t To maximize the feasibility of the constraints. Therefore, Q can be... t (x t ,u t If we replace it with its dual problem, then problem (31) can be transformed into a single-stage problem. t (x t ,u t The dual problem of ) is:

[0124] (32)

[0125] The constraints in the formula constitute dual variables. feasible domain .

[0126] Q t (x t ,u t Replacing it with its dual problem, the problem of finding extreme scenarios (31) is transformed into:

[0127] (33)

[0128] Since the extreme scenario occurs at the endpoint of the new energy uncertainty set, and the 1-norm can be linearized, equation (33) can be linearized into the MISOCP problem based on the big M method.

[0129] Therefore, the post-disaster recovery scheduling of flexible interconnected distribution networks can also be solved by referring to the traditional two-stage RO. That is, the main problem (29) incorporates all extreme scenarios to optimize the pre-scheduling decision x and the fuzzy set dual variable. The subproblem (33) finds the extreme scenarios that have the greatest impact on the robust cost constraint (28c) after the above variables are determined. The main and subproblems are iterated alternately, and the algorithm converges when no more scenarios violate the robust cost constraint. However, the disaster recovery scheduling of flexible interconnected distribution networks contains a large number of binary variables for repair order and topology reconstruction, and the scale of extreme scenarios is relatively large. The above extreme scenario generation method can only generate one extreme scenario per iteration, which requires iteratively solving the mixed integer main problem multiple times, resulting in a heavy computational burden. To this end, this implementation proposes a two-layer iterative extreme scenario generation method, which aims to reduce the number of times the mixed integer main problem is solved, so as to improve the computational efficiency of the algorithm.

[0130] This implementation uses a two-layer iterative extreme scenario generation method for solution. The outer iteration of this method incorporates all extreme scenarios into the main problem to optimize the pre-scheduling decision x, while the subproblems are fixed at x, generating a cluster of extreme scenarios at each time period. The inner iteration generates subproblems from the extreme scenarios, and its approach is to alternately optimize the fuzzy set dual variables. The proposed two-layer iterative extreme scenario generation method, along with extreme scenario variables, achieves higher overall computational efficiency because the inner iteration no longer optimizes the binary pre-scheduling decision x. The main difference between the proposed method and the traditional single-layer iterative algorithm lies in the subproblem: the proposed method fixes the pre-scheduling decision x and generates a cluster of extreme scenarios in each (outer) iteration. In contrast, the traditional method fixes the pre-scheduling decision x and the fuzzy set dual variable in the subproblem and generates only one extreme scenario in each iteration. The proposed algorithm reduces the number of iterations required to generate all extreme scenarios, thus reducing the number of times the MISOCP main problem needs to be solved. Furthermore, by decoupling the energy storage system from the new energy fuzzy set time period, the scale of extreme scenarios in each time period is reduced, and each time period can independently generate and upload extreme scenarios, further improving computational efficiency.

[0131] Therefore, the problem of post-disaster recovery of distribution networks can be solved using a master-subproblem two-level iterative algorithm, such as... Figure 2 As shown. In the outer iteration, the subproblem searches for extreme distributions based on the determined scheduling decision x, while the main problem incorporates all extreme distributions and forms the primal cut, thereby optimizing the scheduling decision x. The two problems iterate alternately until convergence, outputting the scheduling decision x. The algorithm steps are as follows:

[0132] Algorithm 1: Outer layer iterative solution for post-disaster scheduling of distribution network:

[0133] 1) Initialize the set of extreme scenarios for new energy sources Assuming the center of the desired value, initialize the iteration step n=1 and the lower bound LB and upper bound UB of the target value.

[0134] 2) Solve the main problem (29) and determine the pre-scheduling decision x. Set the objective value of the main problem as the lower bound LB.

[0135] 3) Fix x, and use inner iteration to generate a family of extreme scenarios for the subproblem. Set the target value of the subproblem as the upper bound UB.

[0136] 4) If UB-LB≤ε1, stop the iteration; otherwise, add the generated extreme scenarios for each time period to the iterative process. And repeat step 2).

[0137] Algorithm 2: Inner layer iteratively generates a cluster of extreme scenarios (fixed pre-scheduling decision x).

[0138] 1) Initialize the set of extreme scenarios for this iteration Let the center of expectation be the robust cost constraint (28c), and set its feasibility to negative infinity.

[0139] 2) Solve problem (28) and determine the dual variables corresponding to the new energy fuzzy set. The target value is D t * Then the inner iteration generates the upper bound of the target value UB = .

[0140] 3) Solve problem (33) to determine the extreme scenario that has the greatest impact on the robust cost constraint, with an objective value of f. feas (t).

[0141] 4) If f feas (t)≥-ε2, Stop iterating if the extreme scenarios are not found. Otherwise, add the newly obtained extreme scenarios to the set. And repeat step 2).

[0142] It is easy to see that this invention, through the coordinated optimization of fault repair and topology reconfiguration, and the improvement of power flow distribution by flexible interconnection devices, enables the distribution network to reduce load losses and improve dispatch robustness under renewable energy fluctuations. Simultaneously, the model employs biblical optimization to handle renewable energy uncertainties, and the proposed two-layer iterative extreme scenario generation algorithm reduces probability distribution uncertainty to scenario uncertainty, reducing the number of iterations in the mixed integer master problem and achieving high computational efficiency (taking only about 1 minute), thus supporting rapid computation in practical systems.

[0143] The second embodiment of the present invention relates to an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the first embodiment of the flexible interconnected distribution network collaborative scheduling method for disaster recovery.

[0144] The third embodiment of the present invention relates to a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the first embodiment of the method for coordinated scheduling of flexible interconnected distribution networks for disaster recovery.

[0145] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage and optical storage) containing computer-usable program code.

[0146] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0147] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction methods implemented in a process. Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0148] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0149] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for collaborative scheduling of flexible interconnected distribution networks for post-disaster recovery, characterized in that, Includes the following steps: Establish time-period decoupling power model and fault component repair sequence optimization model for energy storage flexible soft switching system respectively; Using the time-period decoupled power model of the energy storage flexible soft switch system, the fault component repair sequence optimization model, and the distribution network topology reconfiguration as pre-decision variables, a collaborative optimization scheduling model considering the uncertainty of new energy sources is constructed with the goal of minimizing the overall cost. The collaborative optimization scheduling model is solved using a two-level iterative extreme scenario generation algorithm based on the duality of infinite-dimensional linear programming, and the scheduling decision is obtained.

2. The method for coordinated dispatching of flexible interconnected distribution networks for post-disaster recovery as described in claim 1, characterized in that, The time-period decoupling power model of the energy storage flexible soft-switching system satisfies the following constraints: The active power balance constraint of the E-SOP system is expressed as: ,in, Let be the active power flowing into the AC side of the m-th converter. Let m be the active power loss of the m-th converter. and These are the discharge power and charging power of the s-th energy storage unit in the DC bus, respectively. The loss constraint of the converter is expressed as: , where c m Let be the loss factor of the m-th converter. Let be the reactive power flowing into the AC side of the m-th converter; The capacity constraint of the converter is expressed as: ,in, Let m be the apparent power capacity of the m-th converter; The allowable boundary constraints for charging and discharging power during the energy storage pre-scheduling phase are expressed as follows: ,in, Let be the lower and upper bound variables of the allowable charging power of the s-th energy storage at time t, respectively. Let be the lower and upper bound variables of the allowable discharge power of the s-th energy storage unit at time t, respectively. All are binary variables. and These are the physical upper bounds of the discharge power and charging power of the s-th energy storage unit at time t, respectively. These are the lower and upper bounds of the allowable capacity of the s-th energy storage unit at time t, respectively. Let represent the charging efficiency and discharging efficiency of the s-th energy storage device, respectively. Represents a time variable. These represent the lower and upper bounds of the energy storage capacity of the s-th energy storage unit, respectively. For adjustment coefficients; The allowable boundary constraints for charge and discharge power during the energy storage readjustment phase are expressed as follows: .

3. The method for coordinated dispatching of flexible interconnected distribution networks for post-disaster recovery as described in claim 2, characterized in that, The loss constraint relaxation of the converter is as follows: .

4. The method for coordinated dispatching of flexible interconnected distribution networks for post-disaster recovery as described in claim 1, characterized in that, The optimization model for the repair sequence of faulty components includes: Spatial transfer constraints are expressed as: Where ℛ represents the set of locations consisting of the maintenance center and the faulty component, and ℛ\r represents the set of other locations excluding the faulty component r. This indicates whether the repair team d has moved from location s' to the location of the faulty component r. This indicates whether the repair team d has moved from the location of the faulty component r to location s. This indicates whether the repair team d passed through the location of the faulty component r. Indicate whether the emergency repair team d has passed through the maintenance center; Temporal association constraints are expressed as: , where t r,d t is the time when the maintenance team d arrives at the location of the faulty component r. s,d Δt is the time when maintenance team d arrives at location s. r,d Δt represents the repair time taken by maintenance team d to repair the faulty component r. r,s,d The travel time for maintenance team d to move from the location of the faulty component r to location s. It is a constant. This indicates whether the faulty component r was repaired during time period t.

5. The method for coordinated dispatching of flexible interconnected distribution networks for post-disaster recovery as described in claim 2, characterized in that, The objective function of the collaborative optimization scheduling model is: ,in, For the active power output of unit n, g n,t D is the power generation cost coefficient for unit n. i,t The original load of node i, The actual load after load abandonment. This is the load abandonment penalty coefficient. e represents the active power of the bus. t For electricity price, u is the predicted value of new energy k. k,t For the actual fluctuation of new energy k, For the actual contribution of new energy k, This is the penalty coefficient for losses in new energy sources.

6. The method for coordinated scheduling of flexible interconnected distribution networks for post-disaster recovery according to claim 5, characterized in that, The constraints of the cooperative optimization scheduling model are: ,in, These are the lower and upper bounds of the active power output of gas turbine n, respectively. These are the reactive power outputs of gas turbine n. The lower and upper bounds, The predicted active power output of new energy source k. For the active power fluctuation of new energy k, The reactive power output of new energy k is respectively The lower and upper bounds, and Let represent the active load and reactive load of node i, respectively. The original active load before node i discards the load. For power factor, A i and B i Let p be the set of upstream nodes and the set of downstream nodes of node i, respectively. ij,t and q ij,t These represent the active and reactive power transmitted by line ij, respectively. i,t and q i,t These represent the net active power injection and net reactive power injection at node i, respectively, δ i v is a binary constant. i and v j These are the voltages at both ends of line ij, where v0 is the rated voltage and r is the voltage across the line. ij and x ij These are the resistance and reactance of line ij, respectively. Let be the on / off state variable of line ij. Let be the lower and upper bounds of the voltage at node i, respectively. These are the lower and upper bounds of the active power transmitted by line ij, respectively. These are the lower and upper bounds of the reactive power transmitted by line ij, respectively.

7. The method for coordinated scheduling of flexible interconnected distribution networks for post-disaster recovery according to claim 5, characterized in that, The collaborative optimization scheduling model adopts new energy fuzzy sets. To describe the probability distribution P of new energy sources in different time periods t The feasible region, the new energy fuzzy set Represented as: , where p(u t ) for scene u t The probability density, This represents the expected value for all scenarios. Center of expected value For the threshold, These are the lower and upper bounds of the expected value, respectively. For the uncertainty set of new energy; The compact form of the expected operating cost of the collaborative optimization scheduling model is expressed as follows: Where c is the cost coefficient of the first stage, x is the decision of the first stage, u is the uncertainty variable, and Q is the cost coefficient of the first stage. t (x t ,u t () represents the optimal value for the second stage readjustment cost in each time period.

8. The method for coordinated dispatching of flexible interconnected distribution networks for post-disaster recovery as described in claim 1, characterized in that, The method of solving the collaborative optimization scheduling model using a two-level iterative extreme scenario generation algorithm based on the duality of infinite-dimensional linear programming to obtain scheduling decisions specifically includes: Based on the dual transformation of infinite-dimensional linear programming, the probability distribution uncertainty in the collaborative optimization scheduling model is transformed into scenario uncertainty; By focusing on the extreme scenarios that have the greatest impact on scheduling constraints, the collaborative optimization scheduling model, after being transformed into scenario uncertainty, is simplified. A two-layer iterative extreme scenario generation algorithm is used to solve the simplified cooperative optimization scheduling model. Specifically, the two-layer iterative extreme scenario generation algorithm includes: Initialize the set of extreme scenarios; Solve the cooperative optimization scheduling model after transforming it into a scenario with uncertainty, optimize the pre-scheduling decision, and obtain the lower bound of the cost; Initialize the set of extreme scenarios for this iteration as the center of the expected value, fix the optimized pre-scheduling decision, determine the dual variable corresponding to the new energy fuzzy set, generate the upper bound of the cost, and solve the simplified collaborative optimization scheduling model to generate a cluster of extreme scenarios. Repeat this step until the target value is reached. When the difference between the upper and lower bounds of the cost is greater than a preset value, the generated extreme scenario is added to the extreme scenario set, and the solution is returned to the collaborative optimization scheduling model after the scenario uncertainty is converted.

9. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the collaborative scheduling method for flexible interconnected distribution networks for disaster recovery as described in any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the collaborative scheduling method for flexible interconnected distribution networks for disaster recovery as described in any one of claims 1-8.

Images