Dro-based energy storage coordinated emergency dispatching method under typhoon strong uncertainty

Abstract

The method for energy storage collaborative emergency scheduling under strong uncertainty of typhoon provided by the application comprises the following steps: S1, constructing a typhoon wind field model by considering the uncertainty of typhoon path and intensity; S2, describing the failure rate of overhead lines and towers of power transmission lines in the distribution network under typhoon disaster, constructing a power transmission line fault model, and then using a matrix fuzzy set to describe the line fault uncertainty and constructing a power transmission line fault fuzzy set; S3, constructing a two-stage energy storage collaborative emergency scheduling model based on DRO, which optimizes the pre-disaster layout position of MES in the first stage and optimizes the power output scheme of each SES and MES during the typhoon disaster in the second stage; S4, obtaining typhoon forecast information and distribution network information; S5, inputting the obtained information into the two-stage energy storage collaborative emergency scheduling model, solving based on CG and C&CG, and obtaining the energy storage collaborative emergency scheduling strategy under the strong uncertainty of typhoon.

Description

Technical Field

[0001] This invention relates to the field of power distribution network control technology, and in particular to a method for coordinated emergency dispatch of energy storage based on DRO under strong uncertainty of typhoons. Background Technology

[0002] Against the backdrop of global ocean warming, extreme weather events such as typhoons and hurricanes are becoming increasingly frequent, seriously threatening the safe and reliable operation of power distribution systems in coastal areas. The development of energy storage technology provides an effective solution for disaster emergency response. By rationally scheduling energy storage systems, economic losses caused by power shortages during disasters can be mitigated to the greatest extent possible. Depending on the scheduling method, energy storage systems are generally divided into two categories: stationary energy storage (SES) and mobile energy storage (MES). SES has the advantage of long-term, large-scale energy storage, while MES is characterized by flexible scheduling and the ability to adjust deployment locations as needed. However, due to the high uncertainty of typhoon disasters, their path predictions still have errors of tens of kilometers, making it difficult to accurately estimate the wind speed intensity and line fault conditions affecting the power distribution network. This uncertainty significantly increases the complexity of energy storage system scheduling, especially for MES; if the deployment location is inappropriate, its emergency value will decrease sharply.

[0003] To address the strong uncertainty of disasters, existing research mainly employs three modeling approaches: First, scenario-based stochastic programming (SP) generates a large number of scenarios through sampling simulations to characterize uncertainty. It assigns weights to scenarios to guide scheduling schemes. However, it requires a large number of scenarios to ensure representativeness, has a large problem scale, is time-consuming to solve, and does not adequately address scenarios with low probability and high losses. Second, robust optimization (RO) models various uncertainties as multiple uncertainty sets and determines scheduling schemes based on the worst-case scenario. Although robust optimization can effectively improve the safety of scheduling schemes, its drawback is that it only focuses on the worst-case scenario, leading to overly conservative and redundant schemes with significantly compromised economic efficiency, thus often lacking appeal in practical applications for power companies. Third, distributedly robust optimization (DRO) balances stochasticity and robustness, serving as a compromise between the first two methods. It identifies the worst-case probability distribution within a fuzzy set and optimizes scheduling schemes based on this distribution.

[0004] In the research on sub-Bruno bar optimization, Sadra proposed a moment fuzzy set to characterize the fault characteristics of lines under extreme events, and constructed a sub-Bruno bar optimization model for network reconstruction based on this, but did not address the energy storage scheduling problem. Yujia Li established a sub-Bruno bar optimization model for pre-disaster line reinforcement scenarios, and fitted corresponding line vulnerability curves according to different reinforcement strategies. Yuan Yang extracted typical empirical scenarios and their probability distributions, allowed for a certain shift in scenario probabilities, constructed a sub-Bruno bar optimization model based on scenario probability fuzzy sets, and used Benders decomposition and C&CG collaborative algorithms to solve it. The modeling and solving of sub-Bruno bar optimization are highly complex, computationally time-consuming, and the most unfavorable scenarios or distributions are usually difficult to output explicitly, resulting in relatively limited interpretability and guiding value in practical applications. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide a method for coordinated emergency dispatch of energy storage based on DRO under strong uncertainty of typhoon disasters. This method can realize the coordinated dispatch of energy storage systems under strong uncertainty of typhoon disasters, effectively reduce economic losses during typhoons, and improve the emergency response capability and operational resilience of power distribution systems.

[0006] To address the aforementioned technical problems, this invention employs the following technical method: a method for coordinated emergency dispatch of energy storage based on DRO under strong uncertainty during typhoons, comprising:

[0007] Step S1: Considering the uncertainties in the typhoon's path and intensity, construct a typhoon wind field model;

[0008] Step S2: Characterize the fault probability of overhead lines and towers of power distribution network transmission lines under typhoon disaster, construct a fault model of power distribution network transmission lines, and then use the moment fuzzy set method to characterize the uncertainty of power distribution network line faults and construct a fuzzy set of power distribution network transmission line faults.

[0009] Step S3: Construct a two-stage energy storage collaborative emergency dispatch model based on DRO. The model aims to minimize the load loss and system economic loss caused by typhoon disasters. The first stage optimizes the pre-disaster layout location of MES, and the second stage optimizes the power output scheme of each SES and MES during the typhoon disaster.

[0010] Step S4: Obtain typhoon forecast information and power distribution network information;

[0011] Step S5: Input the information obtained in step S4 into the two-stage energy storage collaborative emergency dispatch model, solve the model, and obtain the energy storage collaborative emergency dispatch strategy under the strong uncertainty of typhoon.

[0012] Furthermore, in step S1, the instantaneous typhoon wind field hitting the power distribution network is decoupled into translational velocity and circulation wind speed. The typhoon movement behavior is characterized by the Masae Miyazaki model, and the near-ground wind field distribution is analyzed by the Rankine circulation wind speed model. The resulting typhoon wind field model is as follows:

[0013] (1)

[0014] (2)

[0015] (3)

[0016] (4)

[0017] (5)

[0018] In the formula, , A collection of typhoon paths; It is the first Under the given typhoon path, the distance between the power distribution network area and the typhoon center is determined by the real-time coordinates of the typhoon center. and distribution network coordinates Calculated; The distance from the center of the typhoon is The speed at which the typhoon wind field moves within the power distribution network; It is the circulating wind speed of the power distribution network wind field; The radius of the Earth is taken as 6371 km; This refers to the speed at which the typhoon's center moves. This represents the maximum circulation wind speed of the typhoon. This corresponds to the radius of the maximum wind speed. This is the maximum wind speed near the center of the typhoon; and These represent the wind speed and wind radius corresponding to the Beaufort scale (Lev), respectively.

[0019] Furthermore, in step S2, adjacent nodes of the distribution network , The failure rate of the overhead lines between them is represented by the following model:

[0020] (6)

[0021] In the formula, For the first Under the typhoon path, nodes , The overhead lines between Failure rate at any given time; For the first Under the typhoon path, nodes , The overhead lines between The constant circulating wind speed; and These are the model coefficients. Take [11,13], Take [-20, -16]; The design wind speed for overhead lines is set; if the wind speed is exceeded, the failure rate of overhead lines increases sharply. It is a node , The length of the overhead line between them;

[0022] adjacent nodes of distribution network , The failure rate of the towers for the overhead lines between them is represented by the following model:

[0023] (7)

[0024] In the formula, For the first Under the typhoon path, nodes , Between Failure rate of the tower; These are model coefficients, set to 0.3; This is the design wind speed for the tower; exceeding this wind speed will cause the tower's failure rate to rise sharply.

[0025] Construct adjacent nodes of the distribution network based on the reliability formula of a series system. , The fault model for the transmission line is as follows:

[0026] (8)

[0027] In the formula, Adjacent nodes of the distribution network , Transmission lines between Failure rate at any given time; Adjacent nodes of the distribution network , The total number of towers between; The sign for product.

[0028] Furthermore, in step S2, the uncertainty of distribution network line faults is characterized using the moment fuzzy set method, as shown in the following equation:

[0029] (9)

[0030] (10)

[0031] In the formula, In order to target the The fuzzy set of fault moments for power distribution lines constructed from the typhoon path contains the set of all possible line fault distributions. This represents a specific probability distribution that satisfies the definition. For the state space of line faults, given by equation Definition, constructed based on the idea of ​​budget uncertainty sets; In state space The domain of all distributions; Indicates in The set of all probability distributions in the above; It is the collection of all lines; Index for scheduling periods; The set of dispatch periods is determined by the duration of typhoon coverage of the power distribution network area by typhoons with winds of at least level 7. This is achieved by solving... This equation yields two roots; the absolute value of the difference between the two roots is... ,in, Indicates the first Typhoon path under the typhoon The distance from the distribution network at any given time is defined as the typhoon's coordinates at the initial time. , For the first The angle of movement of the typhoon under the given typhoon path, then ; Indicates the first Under the typhoon path, the line exist The status at any given moment: a value of 1 indicates that the line is normal, and a value of 0 indicates that the line is faulty. Indicates distribution Down Expectations Indicates distribution Below, the line exist The expected failure rate at any given time should be less than the failure rate. ; Budget for line faults.

[0032] Furthermore, in step S3, the objective function of the constructed two-stage energy storage coordinated emergency dispatch model based on DRO is:

[0033] (11)

[0034] In the formula, For the first Fixed costs for vehicle MES dispatch; It means the first Does the vehicle MES connect to the node? A binary variable, where 1 represents access and 0 represents no access; A set of nodes; A collection of MES; For the first The weights of each typhoon path represent the probability of its occurrence; Node weights; This is the unit loss coefficient for load shedding; For the first Under the typhoon path, nodes exist The amount of load loss at any given moment; , The unit discharge cost for SES and MES are respectively; , SES and MES respectively direct to distribution network nodes The active power generated.

[0035] Furthermore, in step S3, the constructed two-stage energy storage coordinated emergency dispatch model based on DRO includes the following constraints:

[0036] 1) MES pre-disaster deployment constraints: It restricts that at most one MES can be connected to each distribution network node and each MES can be assigned to at most one distribution network node;

[0037] 2) Power balance constraints of the power distribution system: These are established based on the linearized DistFlow power flow model;

[0038] 3) Node voltage constraints;

[0039] 4) Power constraints on transmission lines;

[0040] 5) Node load loss constraints;

[0041] 6) Energy storage dispatch constraints: These include energy constraints during emergency dispatch of SES and MES, as well as power constraints during emergency dispatch of SES and MES.

[0042] Furthermore:

[0043] The constraints for the pre-disaster deployment of the MES are:

[0044] (18)

[0045] (19)

[0046] The power balance constraint of the power distribution system is:

[0047] (20)

[0048] (twenty one)

[0049] In the formula, , For node indexing; , For the first Under the typhoon path, the line exist The constant ebb and flow of merit and demerit; , This indicates the active and reactive power that the distribution network obtains from the upstream power grid; For nodes exist The reactive power gap at any given moment; , It is a node exist The active and reactive power requirements at any given time; , Indicates SES from distribution network node The absorbed active and reactive power, i.e., charging; , This indicates that SES is directed to the distribution network nodes. The active and reactive power generated; , Indicates the first Vehicle MES from distribution network node The absorbed active and reactive power; and It is MES to the distribution network node The active and reactive power generated, i.e., the discharge;

[0050] The node voltage constraint is:

[0051] (twenty two)

[0052] (twenty three)

[0053] In the formula, , They are the first Nodes under the typhoon path , exist Voltage at any given moment; It is a node The upstream node; This is the reference voltage, set to 1.0 pu; , It is a line Resistance and reactance; , It is a node Minimum and maximum allowable voltages; M is the maximum value;

[0054] The power constraint of the transmission line is:

[0055] (twenty four)

[0056] (25)

[0057] In the formula, , They represent the lines respectively. Upper limits for transmitted active and reactive power;

[0058] The node load decompression constraint is:

[0059] (26)

[0060] (27)

[0061] The energy storage scheduling constraints are:

[0062] (28)

[0063] (29)

[0064] (30)

[0065] (31)

[0066] (32)

[0067] (33)

[0068] In the formula, equations (28)-(31) are the energy constraints of energy storage scheduling, and equations (32)-(33) are the power constraints of energy storage scheduling; , They are the first SES and MES under the typhoon path The amount of electricity stored at all times, , These are the lower limits of SES and MES battery levels, respectively. , These are the maximum battery caps for SES and MES, respectively. It refers to the energy storage charging and discharging efficiency; The scheduling time interval; , They are nodes The rated active power and rated reactive power of the SES; , The first The rated active power and rated reactive power of the MES.

[0069] Preferably, step S5 includes:

[0070] Step S501, the two-stage energy storage coordinated emergency dispatch model is expressed in the following compact form:

[0071] (12)

[0072] In the formula, Indicates the decision variables for the first stage. For the reason The vector formed; These are the coefficients in the objective function of the first stage; This indicates that the second phase is aimed at the first... Decision variables for typhoon paths , , , , , , , , , , , , , , , , They are respectively from , , , , , , , , , , , , , , , , The vector formed; These are the coefficients in the objective function of the second stage; the superscript T indicates transpose; Represents transmission line fault variables; Indicates the inner min model pair The relevant constraints are given by G, h, E, and M, which are model coefficients. inner layer min model Alternative models;

[0073] Step S502: Based on the C&CG algorithm, the two-stage energy storage collaborative emergency dispatch model shown in equation (12) is decomposed into the following C&CG main problem and C&CG sub-problems, and based on the CG algorithm, the C&CG sub-problems are decomposed into the following CG main problem and CG sub-problems:

[0074] ① CG main issue

[0075] In equation (12) Partially equivalent to:

[0076] (13)

[0077] In the formula, This represents an index of power transmission line fault scenarios. For the first Typhoon path number Line fault scenarios in various situations. For the first The probability of each scenario occurring. Total number of scenes;

[0078] (14)

[0079] In the formula, This represents the optimal value for the principal problem of CG. For the first The worst-case scenario set for each typhoon path is obtained by searching the CG subproblem; Represents the transmission line fault rate vector; , The values ​​of the dual variables of the constraints;

[0080] ② CG subproblem

[0081] (15)

[0082] In the formula, Represents the Hadamard product; The optimal value for the CG subproblem;

[0083] Performing a dual transformation on the inner min model of equation (15), equation (15) is reconstructed as follows:

[0084] (16)

[0085] In the formula, For the dual variables of the inner min model constraint of equation (15);

[0086] ③C&CG main problem

[0087] (17)

[0088] In the formula, The optimal value for the C&CG principal problem; It is a variable associated with the target value of the second stage; and It is an index related to the number of iterations. Indicates the first Scenario during the next iteration The probability of;

[0089] ④C&CG subproblem

[0090] The C&CG subproblem is the problem of solving the second stage of the two-stage energy storage coordinated emergency dispatch model. ;

[0091] Step S503: Input the information obtained in step S4 into the two-stage energy storage collaborative emergency dispatch model as shown in equation (12);

[0092] Step S504: Use the CG algorithm to iteratively solve the CG main problem and subproblems, and obtain the worst-case scenario set. and its distribution Add the variables to the main C&CG problem, and then use the C&CG algorithm to solve the main C&CG problem. Iterate in this way until the optimal decision variables of the two-stage energy storage coordinated emergency dispatch model are obtained by gradually converging.

[0093] As another aspect of the present invention, an energy storage collaborative emergency dispatch device based on DRO under strong typhoon uncertainty includes a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement the steps of the aforementioned energy storage collaborative emergency dispatch method based on DRO under strong typhoon uncertainty.

[0094] As another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the aforementioned energy storage collaborative emergency dispatch method based on DRO under strong uncertainty of typhoons.

[0095] As another aspect of the present invention, a computer program product includes a computer program that, when executed by a processor, implements the steps of the aforementioned energy storage collaborative emergency dispatch method based on DRO under strong uncertainty of typhoons.

[0096] In summary, the energy storage collaborative emergency dispatch method based on DRO under strong typhoon uncertainty proposed in this invention mainly consists of three parts. The first part constructs a wind field model and a fuzzy set of transmission line faults under strong typhoon uncertainty. The second part constructs a two-stage energy storage collaborative emergency dispatch model based on DRO. The third part solves the model using algorithms based on CG and C&CG. The first part of this method uses the moment fuzzy set method to characterize the probability of distribution line faults under typhoon disasters, considering the uncertainty of typhoon path and intensity, so that the line fault rate can be more accurately described at the probabilistic level, providing reliable input parameters for the subsequent dispatch model. The second part of this method, based on the aforementioned fuzzy set of transmission line faults, constructs a two-stage energy storage collaborative emergency dispatch model under the DRO framework. The first stage determines the spatial deployment scheme of MES (Mechanical Storage Execution System), and the second stage optimizes the power output strategy of each SES (System-Specific Energy Execution System) and MES during the disaster to minimize the system's economic losses caused by the disaster. The third part of this method addresses the high complexity and difficulty in solving the two-stage energy storage collaborative emergency dispatch model proposed in the second part. It designs a hybrid solution framework combining column generation CG and C&CG algorithms. By iteratively generating key scenarios and constraints, it effectively improves the model's computational efficiency and convergence, thus ensuring its applicability in real-world scenarios. These three parts respectively cover uncertainty modeling, optimization model construction, and algorithm solving. Their sequential connection and coordinated operation enable collaborative dispatch of energy storage systems under conditions of high uncertainty in disasters, effectively reducing economic losses during typhoons and enhancing the emergency response capability and operational resilience of the power distribution system. Attached Figure Description

[0097] Figure 1 This is a flowchart illustrating the solution process for the DRO-based energy storage collaborative emergency dispatch model involved in this invention.

[0098] Figure 2 This is a test distribution network topology diagram in an embodiment of the present invention;

[0099] Figure 3 This is a schematic diagram of a representative typhoon prediction path and corresponding circulation wind speed in an embodiment of the present invention.

[0100] Figure 4 This is a comparison chart of the expected load loss at different nodes in different embodiments of the present invention. Detailed Implementation

[0101] To facilitate understanding by those skilled in the art, the present invention will be further described below with reference to embodiments and accompanying drawings. The content mentioned in the embodiments is not intended to limit the present invention.

[0102] Partial Blue bar optimization can achieve a certain balance between conservatism and economy, but research on its application in disaster-related coordinated energy storage scheduling still needs improvement. To address this, this invention proposes a DRO-based emergency energy storage coordinated scheduling method under strong uncertainty during typhoons. This method constructs a two-stage optimization framework based on partial Blue bar optimization and employs column generation (CG) and C&CG collaborative solution models, significantly shortening computation time and enhancing model applicability. This method can provide decision guidance for emergency scheduling of distribution systems containing SES and MES, effectively reducing economic losses caused by disasters and improving the resilience and reliability of distribution systems.

[0103] Specifically, the energy storage collaborative emergency dispatch method based on DRO under strong uncertainty of typhoons provided by this invention includes:

[0104] Step S1: Considering the uncertainties in the typhoon's path and intensity, construct a typhoon wind field model.

[0105] The instantaneous typhoon wind field striking the power distribution network can be decoupled into two parts: "translational velocity" and "circulation wind speed." The former determines the duration of regional damage, while the latter determines the fault status of overhead lines and towers. Given that the study area is located in the coastal region of China, within the main typhoon path zone of the western Pacific, this invention uses the Masae Miyazaki model to characterize typhoon movement and the Rankine circulation wind speed model to analyze the near-surface wind field distribution. For a single typhoon path... The typhoon wind field model is as follows:

[0106] (1)

[0107] (2)

[0108] (3)

[0109] (4)

[0110] (5)

[0111] In the formula, , A collection of typhoon paths; It is the first Under the given typhoon path, the distance between the power distribution network area and the typhoon center is determined by the real-time coordinates of the typhoon center. and distribution network coordinates Calculated; The distance from the center of the typhoon is The speed at which the typhoon wind field moves within the power distribution network; It is the circulating wind speed of the power distribution network wind field; The radius of the Earth is taken as 6371 km; This refers to the speed at which the typhoon's center moves. This represents the maximum circulation wind speed of the typhoon. This corresponds to the radius of the maximum wind speed. This is the maximum wind speed near the center of the typhoon; and These represent the wind speed and the radius of the wind circle corresponding to the Beaufort scale (Lev), for example... This indicates the radius of the 7-level wind circle.

[0112] Although the aforementioned wind field model can accurately analyze instantaneous wind speeds under a single typhoon path, the meteorological department's forecasts of the typhoon's center location and intensity still suffer from significant uncertainty due to the combined effects of initial field bias and model errors, posing difficulties for subsequent emergency energy storage dispatch. Therefore, this invention introduces multiple possible typhoon path scenarios during the modeling process, namely… ( This involves creating a set of typhoon paths and deriving the potential wind speed levels in the distribution network area under each scenario based on wind field calculation formulas. Furthermore, by incorporating historical data showing the deviation patterns between predicted and actual paths, different path scenarios are assigned corresponding weights to more accurately characterize the impact of typhoon uncertainties on distribution network operation.

[0113] Step S2: Characterize the fault probability of overhead lines and towers of power distribution network transmission lines under typhoon disaster, construct a fault model of power distribution network transmission lines, and then use the moment fuzzy set method to characterize the uncertainty of power distribution network line faults and construct a fuzzy set of power distribution network transmission line faults.

[0114] During typhoons, power outages in the distribution network are mainly caused by damage to overhead lines and towers between nodes. Given the extremely low failure rate of underground cables, this invention only establishes a fault model for the "overhead line-tower" structure. Since the line length between adjacent nodes is relatively short compared to the entire typhoon wind field, it can be reasonably assumed that they experience similar levels of wind damage. Therefore, adjacent nodes in the distribution network... , Faults in multiple sections of overhead lines can be uniformly represented using the following probabilistic model:

[0115] (6)

[0116] In the formula, For the first Under the typhoon path, nodes , The overhead lines between Failure rate at any given time; For the first Under the typhoon path, nodes , The overhead lines between The constant circulating wind speed; and These are the model coefficients. Take [11,13], Take [-20, -16]; The design wind speed for overhead lines is set; if the wind speed is exceeded, the failure rate of overhead lines increases sharply. It is a node , The length of the overhead line between them.

[0117] adjacent nodes of distribution network , The overhead lines between them are usually supported by multiple towers, and the failure of each tower can be considered an independent event. Their failure rates are as follows:

[0118] (7)

[0119] In the formula, For the first Under the typhoon path, nodes , Between Failure rate of the tower; These are model coefficients, set to 0.3; This refers to the design wind speed of the tower. If the wind speed exceeds this, the failure rate of the tower will increase sharply.

[0120] The reliability formula for a series system can be used to construct the adjacent nodes of a distribution network. , The fault model for the transmission line is as follows:

[0121] (8)

[0122] In the formula, Adjacent nodes of the distribution network , Transmission lines between Failure rate at any given time; Adjacent nodes of the distribution network , The total number of towers between; The sign for product.

[0123] To overcome the conservatism of robust optimization, which only focuses on the worst-case scenario, this invention introduces the moment fuzzy set method to characterize the uncertainty of distribution network line faults, as shown in the following equation:

[0124] (9)

[0125] (10)

[0126] In the formula, In order to target the The fuzzy set of fault moments for power distribution lines constructed from the typhoon path contains the set of all possible line fault distributions. This represents a specific probability distribution that satisfies the definition. The state space for line faults is defined by equation (10) and constructed based on the idea of ​​budget uncertainty sets. In state space The domain of all distributions; Indicates in The set of all probability distributions is the complete set of all possible line fault distributions; It is the collection of all lines; Index for scheduling periods; This is a set of scheduling periods, the range of which is determined by the effective impact time of the typhoon on the power distribution network. Considering that outdoor operations cannot be carried out when the wind speed exceeds level 7, this invention assumes that the duration of the typhoon's level 7 wind circle covering the power distribution network area is the typhoon's impact time. This is achieved by solving... This equation yields two roots; the absolute value of the difference between the two roots is... ,in, Indicates the first Typhoon path under the typhoon The distance from the distribution network at any given time is defined as the typhoon's coordinates at the initial time. , For the first The angle of movement of the typhoon under the given typhoon path, then ; Indicates the first Under the typhoon path, the line exist The status at any given moment: a value of 1 indicates that the line is normal, and a value of 0 indicates that the line is faulty. Indicates distribution Down Expectations Indicates distribution Below, the line exist The expected failure rate at any given time should be less than the failure rate. ; Budget for line faults.

[0127] As can be seen from the above formula, the fuzzy set constructed in this invention introduces line failure rate information when characterizing uncertainty, and its conservatism is effectively reduced compared with the traditional robust optimization method that only relies on boundary constraints.

[0128] Step S3: Construct a two-stage energy storage collaborative emergency dispatch model based on DRO.

[0129] 1. Objective function

[0130] The objective function of the two-stage energy storage coordinated emergency dispatch model based on DRO in this invention is:

[0131] (11)

[0132] In the formula, For the first Fixed costs for vehicle MES dispatch; It means the first Does the vehicle MES connect to the node? A binary variable, where 1 represents access and 0 represents no access; A set of nodes; A collection of MES; For the first The weights of each typhoon path represent the probability of its occurrence; Node weights; This is the unit loss coefficient for load shedding; For the first Under the typhoon path, nodes exist The amount of load loss at any given moment; , The unit discharge cost for SES and MES are respectively; , SES and MES respectively direct to distribution network nodes The active power generated.

[0133] As can be seen from the objective function, this invention constructs a two-stage optimization model: the first stage optimizes the pre-disaster layout location of the MES, and the second stage optimizes the power output scheme of each SES and MES during the typhoon disaster. The objective function of this collaborative scheduling model not only focuses on minimizing load loss, but also ensures the rational utilization of the MES to improve the overall economic efficiency of the system.

[0134] 2. Constraints

[0135] 1) MES pre-disaster layout constraints are:

[0136] (18)

[0137] (19)

[0138] Here, constraint (18) restricts that at most one MES can be connected to each distribution network node, and constraint (19) indicates that each MES can be assigned to at most one distribution network node.

[0139] 2) Power balance constraints of the power distribution system

[0140] Based on the linear DistFlow power flow model, the power balance constraints of the power distribution system are as follows:

[0141] (20)

[0142] (twenty one)

[0143] In the formula, , For node indexing; , For the first Under the typhoon path, the line exist The constant ebb and flow of merit and demerit; , This indicates the active and reactive power that the distribution network obtains from the upstream power grid; For nodes exist The reactive power gap at any given moment; , It is a node exist The active and reactive power requirements at any given time; , Indicates SES from distribution network node The absorbed active and reactive power, i.e., charging; , This indicates that SES is directed to the distribution network nodes. The active and reactive power generated; , Indicates the first Vehicle MES from distribution network node The absorbed active and reactive power; and It is MES to the distribution network node The active and reactive power generated, i.e., the discharge.

[0144] 3) Node voltage constraints are:

[0145] (twenty two)

[0146] (twenty three)

[0147] In the formula, , They are the first Nodes under the typhoon path , exist Voltage at any given moment; It is a node The upstream node; This is the reference voltage, set to 1.0 pu; , It is a line Resistance and reactance; , It is a node The minimum and maximum allowable voltages; M is the maximum value, which is a parameter in the Big M method, referring to a sufficiently large constant whose value needs to be large enough to ensure that the constraint does not have a limiting effect under the corresponding conditions, but it cannot be infinitely large to avoid numerical instability. The value of M will be set according to the model parameters, for example, it is an order of magnitude larger than the maximum value that may occur.

[0148] 4) Power constraints for transmission lines are as follows:

[0149] (twenty four)

[0150] (25)

[0151] In the formula, , They represent the lines respectively. Upper limits for transmitted active and reactive power.

[0152] 5) Node unload constraint, is:

[0153] (26)

[0154] (27)

[0155] 6) Energy storage dispatch constraints

[0156] Energy constraints for emergency dispatching of SES and MES are shown in equations (28)-(31), and power constraints are shown in equations (32)-(33):

[0157] (28)

[0158] (29)

[0159] (30)

[0160] (31)

[0161] (32)

[0162] (33)

[0163] In the formula, , They are the first SES and MES under the typhoon path The amount of electricity stored at all times, , These are the lower limits of SES and MES battery levels, respectively. , These are the maximum battery caps for SES and MES, respectively. It refers to the energy storage charging and discharging efficiency; The scheduling time interval; , They are nodes The rated active power and rated reactive power of the SES; , The first The rated active power and rated reactive power of the MES.

[0164] Step S4: Obtain typhoon forecast information through the China Meteorological Administration's typhoon network, and obtain power distribution network information through the power distribution geographic information system and the power distribution automation system.

[0165] Step S5: Input the information obtained in step S4 into the two-stage energy storage collaborative emergency dispatch model, solve the model, and obtain the energy storage collaborative emergency dispatch strategy under the strong uncertainty of typhoon.

[0166] Step S501, the two-stage energy storage coordinated emergency dispatch model is expressed in the following compact form:

[0167] (12)

[0168] In the formula, Indicates the decision variables for the first stage. For the reason The vector formed; These are the coefficients in the objective function of the first stage; This indicates that the second phase is aimed at the first... Decision variables for typhoon paths , , , , , , , , , , , , , , , , They are respectively from , , , , , , , , , , , , , , , , The vector formed; These are the coefficients in the objective function of the second stage; the superscript T indicates transpose; Represents transmission line fault variables; Indicating the inner min model right The relevant constraints are defined by G, h, E, and M, which are model coefficients. For ease of explanation in the subsequent algorithm section, this invention uses... Replace the inner min model of the second stage .

[0169] Step S502: Solving algorithm for the energy storage collaborative emergency dispatch model based on CG and C&CG.

[0170] The two-stage energy storage coordinated emergency dispatch model constructed in this invention cannot be solved directly and needs to be reconstructed. First, the second stage of the model... Partially equivalent to:

[0171] (13)

[0172] In the formula, This represents an index of power transmission line fault scenarios. For the first Typhoon path number Line fault scenarios in various situations. For the first The probability of each scenario occurring. This represents the total number of scenarios. It's worth noting that a line fault scenario refers to a situation where a line fails. For example, if there are 5 lines in the system, then a failure of line 1 at time 12 can be considered one scenario, while failures of lines 2 and 3 at time 13 can also be considered one scenario. The difference between different scenarios lies in the different lines that fail and the different times when the failures occur.

[0173] As can be seen from the above, the meaning of formula (13) is that the expected value can be calculated by weighted sum of a series of discrete scenarios.

[0174] However, enumerating scenarios significantly increases the problem size, leading to increased solution time. To accelerate the solution, this invention introduces the Gardner-Codd (CG) algorithm. The CG algorithm decomposes the original large-scale problem into a main problem and subproblems, iteratively solving each. The main problem optimizes within a given set of candidate scenarios to obtain a temporary solution; while the subproblems, based on the dual information of the main problem, search for and generate new scenarios that may improve the current solution. When a candidate scenario discovered by a subproblem reduces the objective function value, it is added to the main problem, expanding the feasible region of the solution. Through this iterative update, the CG algorithm avoids enumerating all scenarios in the initial stage, significantly reducing computational scale while gradually approaching the optimal solution. Finally, when the subproblems can no longer generate any more improving scenarios, the iteration process terminates, and the solution to the main problem converges to the globally optimal or near-optimal solution.

[0175] For the second phase For any typhoon path Based on the CG algorithm, the following CG main problem and CG subproblems can be written:

[0176] ① CG main issue

[0177] (14)

[0178] In the formula, This represents the optimal value for the principal problem of CG. For the first The worst-case scenario set for each typhoon path is obtained by searching the CG subproblem; Represents the transmission line fault rate vector; , The values ​​of the dual variables of the constraints.

[0179] ② CG subproblem

[0180] (15)

[0181] In the formula, Represents the Hadamard product; This is the optimal value for the CG subproblem. Equation (15) is still a two-layer model and cannot be solved directly, but at this time... and Since the objective function and constraints involved are all linear, the inner min model satisfies the strong duality theorem. By performing a dual transformation on the inner min model, equation (15) can be reconstructed as follows:

[0182] (16)

[0183] In the formula, is the dual variable constrained by the inner min model of equation (15).

[0184] The reconstructed CG subproblem can be solved directly using commercial solvers. Based on the above reconstruction, the second stage is solved using the CG algorithm. The process is described as follows:

[0185] Step 1: Initialize the scene set in the CG algorithm and tolerance Set the number of iterations .

[0186] Step 2: Solve the CG principal problem shown in equation (14) to obtain the first... The optimal value of the main problem in the next iteration of CG The dual variables corresponding to the constraints ( , ).

[0187] Step 3: Solve the CG subproblem shown in equation (16) to obtain the optimal value of the CG main problem. and extreme scenarios .

[0188] Step 4: If ,at this time This is the second-stage model. The iteration terminates when the approximate engineering value is found. Conversely, if the approximate value is not found, the The scene set passed to the main CG problem is: Go back to step 2.

[0189] As can be seen from the above model, this invention can derive the path of any typhoon. Worst-case scenario set and its distribution This provides decision-makers with more information to assist their decision-making.

[0190] After addressing the challenges of solving the second stage of the model, this invention employs the C&CG algorithm to solve the complete model shown in equation (12). The C&CG algorithm is a commonly used decomposition method. Its core idea is to continuously generate dual cuts or key scenarios through iterative solutions to the main problem and subproblems, and dynamically add them to the main problem. Specifically, the main problem provides candidate first-stage solutions under a given set of scenarios, while the subproblems verify whether there are any constraints that are violated under this solution, and generate new valid inequalities or scenarios accordingly. Through this iterative process, the algorithm can gradually converge to the global optimal solution, thereby effectively reducing computational complexity while ensuring accuracy.

[0191] Based on the C&CG algorithm, the staged energy storage coordinated emergency dispatch model is reconstructed, and the main problem and sub-problems are as follows:

[0192] ③C&CG main problem

[0193] (17)

[0194] In the formula, The optimal value for the C&CG principal problem; It is a variable associated with the target value of the second stage; and It is an index related to the number of iterations. Indicates the first Scenario during the next iteration The probability of.

[0195] ④C&CG subproblem

[0196] The C&CG subproblem is the problem of solving the second stage of the two-stage energy storage coordinated emergency dispatch model. The aforementioned CG algorithm can be used to solve this problem, and will not be elaborated further here.

[0197] like Figure 1 As shown, the complete process of solving the two-stage energy storage collaborative emergency dispatch model proposed in this invention based on the CG algorithm and the C&CG algorithm is as follows:

[0198] Step 1: Input the information obtained in step S4 into the two-stage energy storage collaborative emergency dispatch model as shown in equation (12), and initialize the lower bound in the C&CG algorithm. upper limit Number of iterations =0, tolerance =Minimum value.

[0199] Step 2: Solve equation (17) to obtain the first... The first stage decision variables of the next iteration And the optimal value of the C&CG main problem Update LB= .

[0200] Step 3: Solve the second stage based on the CG algorithm , obtained the The optimal value of the main problem in the next iteration of CG Worst-case scenario set and its distribution ,Will , The relevant constraints are added to the C&CG main problem.

[0201] Step 4: Update UB=min{UB, }

[0202] Step 5: If UB−LB≤ Iteration terminated. This is the optimal decision in equation (12). Conversely, let... Return to step 2.

[0203] This concludes the explanation of the solution algorithm for the two-stage energy storage collaborative emergency dispatch model based on DRO.

[0204] On the other hand, based on the same principle as the DRO-based energy storage collaborative emergency dispatch method under strong typhoon uncertainty described in the above embodiments, the present invention also provides a DRO-based energy storage collaborative emergency dispatch device under strong typhoon uncertainty. This device includes a memory, a processor, and a computer program stored in the memory. The processor executes the computer program to implement the steps of the DRO-based energy storage collaborative emergency dispatch method under strong typhoon uncertainty described in the above embodiments. Specifically, the device can be an electronic computer or tablet computer, the processor can be a CPU, GPU, etc., and the memory can be RAM, ROM, EEPROM, CDROM, disk storage medium, or any other medium capable of carrying or storing the computer program and capable of being read by a computer; no limitation is made herein.

[0205] On the other hand, the present invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the energy storage coordinated emergency dispatch method based on DRO under strong typhoon uncertainty as described in the above embodiments. Specifically, the computer-readable storage medium may be RAM, ROM, EEPROM, SSD, CDROM, DVD, USB flash drive, or any other medium capable of carrying or storing a computer program and capable of being read by a computer.

[0206] On the other hand, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the energy storage collaborative emergency dispatch method based on DRO under strong uncertainty of typhoons as described in the above embodiments.

[0207] To verify the feasibility and effectiveness of the method proposed in this invention, a case study analysis will be conducted using the 18-node distribution network test system provided by Matpower.

[0208] I. Test System Data

[0209] exist Figure 2In the test distribution network shown, node 1 is a substation node with a voltage amplitude per unit value of 1.0 pu. The numbers on the lines indicate the line length in km. Assume that nodes 9 and 18 are both equipped with SES (Self-Discharge Execution System), with capacities of 1MW / 2MWh and 0.4kW / 0.8MWh respectively. The system has two MES (Mechanical Discharge Execution System), with capacities of 0.5MW / 1MWh (MES 1, fixed dispatch cost of 3000 yuan / time) and 0.2MW / 0.5MWh (MES 2, fixed dispatch cost of 2000 yuan / time). The unit discharge cost of both SES and MES is set to 100 yuan / MWh. Important loads are distributed at nodes 5, 9, and 18 (…). Ordinary loads are distributed among the remaining nodes ( The system's load shedding factor is ¥10800 / MWh, and the upper and lower voltage limits are set at 1.05 pu and 0.95 pu, respectively. Line fault budget... Set to 3. Tolerance of the CG algorithm. Set to 0.1, the tolerance of the C&CG algorithm Set to 1. The test was performed on a personal computer with an Intel(R) Core i5-11260H CPU, 16.0GB of memory, MATLAB version R2020b, and CPLEX solver version 12.7.

[0210] Regarding typhoon information, this example selects Typhoon "Mangkut" which occurred in September 2024 as the research object, and assumes that the test power distribution network is located in Zhanjiang, Guangdong, and will be affected by "Talim" for about 8 hours. Based on the meteorological department's prediction of the typhoon's impact range, this scheme selects three representative possible paths for analysis, with their occurrence probabilities set at 40% (path 1), 20% (path 2) and 10% (path 3), respectively. The corresponding circulation wind speed every 0.5 hours is calculated using formula (2), such as... Figure 3 As shown. Since the remaining southward-shifting paths prevent the circulating wind speed experienced by the distribution network from exceeding the design wind speed of the overhead lines (Note: The design wind speed of distribution lines mentioned in this invention refers to the basic wind speed used when designing the mechanical strength of lines and towers according to national standards, i.e., the maximum average wind speed over a 30-year return period, at a height of 10 meters, and over 10 minutes, used as a benchmark to characterize the wind resistance capacity of the line structure.), it is not considered in the analysis.

[0211] II. Case Analysis

[0212] 1. Verification of the effectiveness and superiority of the two-stage energy storage coordinated emergency dispatch model

[0213] To verify the effectiveness and superiority of the two-stage energy storage coordinated emergency dispatch model proposed in this invention, five different cases were set up for comparative analysis.

[0214] Case 1: Using DRO modeling method for collaborative scheduling of SES and MES (this invention).

[0215] Case 2: Using the DRO modeling method for SES scheduling only, MES does not participate in pre-disaster scheduling.

[0216] Case 3: Using the DRO modeling method for MES-only scheduling, without considering the SES running status during scheduling.

[0217] Case 4: Using SP modeling method for collaborative scheduling of SES and MES.

[0218] Case 5: Using the RO modeling method for collaborative scheduling of SES and MES.

[0219] It should be noted that Case 4 generates 30 fault scenarios based on Monte Carlo sampling, and then performs collaborative scheduling; Case 5 performs collaborative scheduling based on the uncertain set shown in Equation (10). After executing the scheduling model, the MES access location of each case is shown in Table 1:

[0220] ;

[0221] To verify the effectiveness of the scheduling in each case, this test resampled 100 fault scenarios for each typhoon path and then used... As the objective function, equations (20) to (33) represent the operating costs of the optimal power flow model simulating the system under constraints, yielding the expected load shedding at each node (e.g., ...). Figure 4 As shown in Table 2, the nodes marked with black triangles are critical load nodes, the expected load shedding of the system, and the expected operating cost of the system are as follows:

[0222] ;

[0223] Comparing the results of Cases 1 to 3 reveals that: without scheduling the MES, the system's emergency resources are not fully utilized, and the saved fixed scheduling costs are insufficient to compensate for the economic losses caused by load shedding; if the operating status of the SES is not considered when scheduling the MES, it will lead to redundancy in resource allocation. For example, if an SES is already configured at node 18, the scheduling result still allocates one MES to that node. Compared with Case 1, this redundant configuration increases the expected operating cost of the system by 800 yuan. The above comparison results show that the SES and MES collaborative scheduling method proposed in this invention can significantly improve the overall utilization efficiency of emergency resources and effectively reduce potential economic losses under disaster conditions, thus verifying the rationality and effectiveness of the constructed two-stage energy storage collaborative emergency scheduling model.

[0224] Comparing the results of Cases 1, 4, and 5 reveals that: SP (Service Provider) uses scenario probability as the basis for decision-making, leading to an over-concentration of MES (Management Execution System) on downstream nodes of high-failure-rate lines. While performing reasonably well in common failures, it systematically ignores low-probability, high-loss extreme events, thus increasing expected operating costs. RO (Reverse Optimization), on the other hand, completely abandons probabilistic information, using only the "worst-case line failure" as an anchor point. For example, it determines that a line failure between nodes 3 and 4 has the most severe consequences, thus forcing MES1 to be deployed on node 4. This results in a significant decrease in resource utilization in non-worst-case scenarios, and the expected cost is actually higher than DRO (Distance-Oriented Optimization). In summary, the DRO modeling of this invention incorporates both probability and extreme value information through fuzzy sets, suppressing overreaction to high-frequency scenarios while maintaining appropriate vigilance against extreme tail risks. It achieves an optimal trade-off between economy and conservatism, validating its superiority in highly uncertain typhoon environments.

[0225] 2. Verification of Algorithm Effectiveness and Superiority

[0226] For solving the DRO model based on moment fuzzy sets, the traditional solution algorithm transforms the dualization of the fuzzy set constraints into an equivalent two-stage optimization model containing Lagrange multipliers, as shown in equation (34) below:

[0227] (34)

[0228] In the formula, It's about constraints. The Lagrange multipliers. For this model, the C&CG algorithm can be used to solve it. It can also be seen from equation (34) that the traditional algorithm cannot derive the distribution.

[0229] To verify the superiority of the proposed CG and C&CG combined solution algorithm, the solution performance of this algorithm and the traditional solution algorithm were compared in Matpower 18-node and IEEE 33-node systems, respectively. The results are shown in Table 3.

[0230] ;

[0231] Table 3 shows that the algorithm proposed in this invention significantly improves the solution efficiency compared to traditional methods. The main reason for this is that this invention avoids explicit modeling of the Lagrange multiplier variables in the main problem, thereby effectively reducing the model size and complexity. This advantage becomes increasingly apparent as the system scale increases. It is worth noting that the algorithm is subject to the internal convergence tolerance of CG. The objective function value obtained by the algorithm is slightly higher than that of the traditional precise algorithm, but the relative deviation is less than 0.1%, which is within the tolerance range for engineering applications. Overall, the algorithm proposed in this invention achieves considerable computational speedup at the cost of negligible accuracy loss, verifying its superiority in actual emergency dispatching of power distribution networks.

[0232] In summary, the energy storage collaborative emergency dispatch method based on DRO under strong uncertainty of typhoons proposed in this invention utilizes moment fuzzy sets to characterize the uncertainty of line fault probability distribution, effectively deriving the searched extreme scenarios and the most unfavorable distribution, providing decision-makers with richer scenario information, and enhancing the scientificity and reliability of auxiliary decision-making. Furthermore, this invention establishes a two-stage energy storage collaborative emergency dispatch model based on fully utilizing statistical information. The first stage determines the spatial deployment scheme of MES, and the second stage optimizes the power output strategies of each SES and MES. Through alternating iterative solutions, collaborative dispatch of SES and MES is achieved under strong uncertainty, which effectively improves the overall utilization efficiency of emergency resources and significantly reduces potential economic losses under disaster conditions, enhancing the resilience and reliability of the power distribution system. In addition, the method of combining CG and C&CG algorithms for alternating iterative solutions effectively reduces the size of the two-stage energy storage collaborative emergency dispatch model and improves the model's computational efficiency.

[0233] The above embodiments are preferred implementations of the present invention. In addition, the present invention can be implemented in other ways. Any obvious substitutions without departing from the concept of the present technical solution are within the protection scope of the present invention.

[0234] To facilitate understanding by those skilled in the art of the improvements of this invention over the prior art, some of the accompanying drawings and descriptions have been simplified, and for clarity, some other elements have been omitted from this application. Those skilled in the art should realize that these omitted elements may also constitute the content of this invention.

Claims

1. A method for energy storage collaborative emergency dispatching based on DRO under typhoon strong uncertainty, characterized in that, The method comprises the following steps: Step S1, considering the uncertainty of the typhoon path and intensity, a typhoon wind field model is constructed; Step S2, the failure probability of overhead lines and towers of power transmission lines of the distribution network under typhoon disaster is described, a power transmission line failure model of the distribution network is constructed, and a fuzzy set method is used to describe the uncertainty of the distribution network line failure, and a distribution network power transmission line failure fuzzy set is constructed; The uncertainty of the distribution network line failure is described by using the fuzzy set method, as follows: （9） （10） In the formula, In order to target the The fuzzy set of fault moments for power distribution lines constructed from the typhoon path contains the set of all possible line fault distributions. This represents a specific probability distribution that satisfies the definition. For the state space of line faults, given by equation Definition, constructed based on the idea of ​​budget uncertainty sets; In state space The domain of all distributions; Indicates in The set of all probability distributions in the above; It is the collection of all lines; Index for scheduling periods; The set of dispatch periods is determined by the duration of typhoon coverage of the power distribution network area by typhoons with winds of at least level 7. This is achieved by solving... This equation yields two roots; the absolute value of the difference between the two roots is... ,in, Indicates the first Typhoon path under the typhoon The distance from the distribution network at any given time is defined as the typhoon's coordinates at the initial time. , For the first The angle of movement of the typhoon under the given typhoon path, then ; Indicates the first Under the typhoon path, the line exist The status at any given moment: a value of 1 indicates that the line is normal, and a value of 0 indicates that the line is faulty. Indicates distribution Down Expectations Indicates distribution Below, the line exist The expected failure rate at any given time should be less than the failure rate. ; Budgeting for line faults; Step S3, a two-stage energy storage collaborative emergency dispatching model based on DRO is constructed, the model aims to minimize the loss of load loss and system economic loss caused by typhoon disaster, the first stage optimizes the pre-disaster layout position of MES, and the second stage optimizes the power output scheme of each SES and MES during the typhoon disaster; The objective function of the two-stage energy storage collaborative emergency dispatching model based on DRO is as follows: （11） wherein, is the fixed cost of the th MES dispatching; is a binary variable indicating whether the th MES is connected to the node , 1 means connected, 0 means not connected; is the set of nodes; is the set of MESs; is the weight of the th typhoon path, indicating the occurrence probability; is the weight of the node; is the unit load loss coefficient; is the load loss of the node under the th typhoon path at the th time; , are the unit discharge costs of the SES and the MES, respectively; , are the active power outputs of the SES and the MES to the distribution network node , respectively; Step S4, typhoon forecast information and distribution network information are obtained; Step S5, the information obtained in step S4 is input into the two-stage energy storage collaborative emergency dispatching model, the model is solved, and an energy storage collaborative emergency dispatching strategy under typhoon strong uncertainty is obtained.

2. The DRO-based energy storage collaborative emergency scheduling method under typhoon strong uncertainty according to claim 1, characterized in that: In step S1, the instantaneous typhoon wind field attacking the distribution network is decoupled into translation speed and circulation wind speed, the typhoon moving behavior is described by using the Miyazaki model, and the Rankine circulation wind speed model is used to analyze the near-ground wind field distribution, and the obtained typhoon wind field model is as follows: （1） （2） （3） （4） （5） In the formula, , is a set of typhoon paths; is the first typhoon path, the distance between the power grid area and the typhoon center is calculated by the real-time typhoon center coordinates and the power grid coordinates ; is the moving speed of the typhoon wind field in the power grid with a distance of from the typhoon center; is the wind speed of the power grid wind field circulation; is the radius of the earth, which is 6371 km; is the moving speed of the typhoon center; is the maximum circulation wind speed of the typhoon, is the corresponding maximum wind speed radius; is the maximum wind speed near the typhoon center; and respectively represent the wind speed and the wind circle radius corresponding to the Beaufort wind scale Lev.

3. The DRO-based energy storage collaborative emergency dispatching method under typhoon strong uncertainty according to claim 2, characterized in that: In the step S2, the failure rate of the overhead line between the adjacent nodes of the power distribution network , is expressed by the following model: （6） In the formula, is the first typhoon path, the fault rate of the overhead line between nodes , at the time ; is the first typhoon path, the circulating wind speed borne by the overhead line between nodes , at the time ; and are model coefficients, take [11, 13], take [-20, -16]; is the design wind speed of the overhead line; is the length of the overhead line between nodes , ; Failure rate of a tower of an overhead line between adjacent nodes of a distribution network , is represented by a model as follows: （7） wherein is the 1st is the 1st , is the 1st is the failure rate of the root tower between nodes is the model coefficient, taken as 0.3; is the tower design wind speed; Construct adjacent nodes of the distribution network based on the reliability formula of a series system. , The fault model for the transmission line is as follows: （8） wherein is the failure rate of the transmission line between adjacent nodes of the distribution grid , at time t; is the failure rate of the transmission line between adjacent nodes of the distribution grid , , is the total number of towers between adjacent nodes of the distribution grid is the integral sign.

4. The DRO-based energy storage collaborative emergency scheduling method under typhoon strong uncertainty according to claim 3, characterized in that: In step S3, the two-stage energy storage collaborative emergency dispatching model based on DRO comprises the following constraint conditions: 1) MES pre-disaster layout constraint: it limits that each distribution network node can be connected to at most one MES, and each MES can be allocated to at most one distribution network node; 2) Power balance constraint of the distribution system: it is established based on the linearized DistFlow power flow model; 3) Node voltage constraint; 4) Power constraint of the transmission line; 5) Node load loss constraint; 6) Energy storage dispatching constraint: it includes energy constraints and power constraints of SES and MES during emergency dispatching.

5. The DRO-based energy storage co-optimization method under typhoon strong uncertainty according to claim 4, characterized in that: Step S5 comprises: Step S501, the two-stage energy storage collaborative emergency dispatching model is expressed in the following compact form: （12） wherein denotes the decision variable of the first stage, is a vector consisting of ; is a coefficient in the first stage objective function; denotes the decision variable of the second stage for the th typhoon path, , , , , , , , , , , , , , , , , is a vector consisting of , , , , , , , , , , , , , , , , ; is a coefficient in the second stage objective function; the superscript T denotes transposition; denotes the transmission line fault variable; denotes the constraint related to of the inner min model, G, h, E, M are model coefficients; is a surrogate model of the inner min model ; Step S502, according to the C&CG algorithm, the two-stage energy storage collaborative emergency dispatching model as shown in formula (12) is decomposed into the following C&CG master problem and C&CG sub-problem, and according to the CG algorithm, the C&CG sub-problem is decomposed into the following CG master problem and CG sub-problem: ① CG master problem In equation (12) Partially equivalent to: （13） In the formula, is expressed as a transmission line fault scenario index, is the first is the first is the line fault condition under the first is the first is the occurrence probability of the first is the total number of scenarios; （14） In the formula, is the optimal value of the CG master problem; is the optimal value of the CG master problem; is the worst scenario set of the typhoon track, obtained by the CG sub-problem search; represents the transmission line failure rate vector; , is the dual variable value of the constraint condition; ② CG sub-problem （15） wherein denotes the Hadamard product; is the optimal value of the CG subproblem; The inner min model of formula (15) is transformed by duality, and formula (15) is reconstructed as: （16） wherein is the dual variable of the inner min model constraint of formula (15); ③ C&CG master problem （17） wherein is the optimal value of the C&CG master problem; is a variable associated with the second stage target value; and is an index related to the iteration number, denotes the probability of the scenario at the th iteration. ④ C&CG sub-problem The C&CG sub-problem is to solve the second stage problem of the two-stage energy storage coordinated emergency dispatching model ; Step S503, the information obtained in step S4 is input into the two-stage energy storage collaborative emergency dispatching model as shown in formula (12); Step S504, the CG algorithm is used to iteratively solve the CG master problem and the sub-problem, and the worst scenario set obtained and its distribution is added to the C&CG master problem, and the C&CG algorithm is used to solve the C&CG master problem, and the iteration is performed until the optimal decision variable of the two-stage energy storage collaborative emergency dispatching model is gradually converged.

6. A device for DRO-based energy storage collaborative emergency dispatch under typhoon strong uncertainty, comprising a memory, a processor and a computer program stored in the memory, characterized in that: The processor executes the computer program to realize the steps of the energy storage collaborative emergency dispatching method based on DRO under typhoon strong uncertainty according to any one of claims 1-5.

7. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to implement the steps of the energy storage collaborative emergency scheduling method based on DRO under typhoon strong uncertainty according to any one of claims 1-5.

8. A computer program product comprising a computer program, characterized in that: The computer program is executed by the processor to implement the steps of the energy storage collaborative emergency scheduling method based on DRO under typhoon strong uncertainty according to any one of claims 1-5.

Images