A power distribution network topology reconfiguration method based on extreme disaster scenarios

Abstract

The application discloses a power distribution network topology reconstruction method based on an extreme disaster scene, considers the uncertainty of disaster occurrence through a random optimization method, optimizes the power distribution network topology based on the probability distribution of accurately predicted power grid fault scenes, and realizes optimal topology reconstruction of the power distribution network under the extreme disaster fault. The virtual power flow method is used to maintain the radial operation of the power distribution network, the power and voltage balance of the power distribution network under topology reconstruction is considered, and a power distribution network topology reconstruction model based on an extreme disaster scene is constructed. The application considers the current natural environment and the frequent occurrence of severe weather, reconstructs the power distribution network topology through a random optimization method considering the fault occurrence scene, and guarantees the power supply reliability and safety of the power distribution network under the extreme disaster.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of source-network-load-storage collaborative optimization, and particularly relates to a power distribution network topology reconstruction method based on extreme disaster scenarios. BACKGROUND

[0002] In combination with the morphological characteristics of new power systems, the resilience improvement research should consider the following challenges:

[0003] 1) With the widespread application of renewable energy such as wind power and photovoltaic power and the dynamic changes in user electricity consumption patterns, the uncertainty among the source, network, and load of new power systems has significantly increased. This uncertainty runs through all stages of power system planning, operation, and control. In addition, natural disasters can cause damage to intelligent monitoring devices in the power grid, which can weaken the system's ability to perceive real-time operating conditions, making it more difficult to accurately predict power supply and demand. Therefore, to improve the resilience of new power systems, a core challenge is how to achieve fine-grained prediction and accurately reflect the characteristics of different uncertainty factors in modeling to promote the resilience of power systems.

[0004] 2) In new power systems, the temporal and spatial distribution characteristics of natural disasters present an uneven state. Since new power systems are highly dependent on new energy generation, the timing of disasters becomes particularly critical. For example, if a natural disaster occurs during the peak period of wind or solar power generation, it may lead to severe power supply shortages and cause imbalances between power sources and loads. In addition, different power generation resources have different sensitivities to disasters, which means that the system's resistance to various extreme weather events varies, posing additional challenges to the overall resilience of the system. To improve the resilience of new power systems, one of the primary tasks is to accurately identify when and where disasters have the most severe impact on the system under complex spatiotemporal conditions, as well as the interaction mechanism between disasters and power systems.

[0005] 3) New power systems face challenges in dynamic control, which are rooted in the high proportion of renewable energy access. Due to the increase in the proportion of renewable energy, the system's rotational inertia has decreased, and this key stability factor of new power systems has been significantly reduced. In addition, the system's ability to peak-shifting, frequency modulation, and dealing with high-low voltage ride-through is relatively weak, which directly weakens its ability to resist external disturbances. The lack of support for reactive power in the system leads to frequent reverse flow of power flow direction and transient overvoltage problems, which exacerbate the instantaneous impact caused by frequent switching actions, which can lead to large-area power grid splitting. Therefore, ensuring that resilience improvement measures designed for steady-state conditions remain feasible in extreme situations during the transition from N-1 to N-k failures and then to normal operation has become a crucial bottom line. SUMMARY

[0006] The technical problem solved by the present application is to provide a power distribution network topology reconstruction method based on extreme disaster scenarios to address the shortcomings of the prior art. The present application takes into account power distribution network topology reconstruction, source-network-load-storage collaboration, introduces power distribution network tree topology constraints, uses virtual flow methods to ensure that the optimal topology of the power distribution network is a tree structure, and each node except the root node has a parent node and does not form a loop. The power distribution network power balance constraint and the line connection constraint ensure the balance between supply and demand within the system, and realize the collaborative scheduling of source, network, load and storage. The present application not only optimally reconstructs the topology of the power distribution network under extreme disasters, ensuring power supply to the load in harsh environments, but also optimally schedules the various power equipment within the system, ensuring the collaborative operation of the source, network, load and storage.

[0007] Technical scheme: In order to solve the above technical problems, the present application provides a power distribution network topology reconstruction method based on extreme natural disaster scenarios, comprising the following steps:

[0008] Step 1, obtain disaster information, including disaster occurrence location, disaster intensity, disaster path; obtain historical photovoltaic scenarios; obtain power grid parameters, including generator set power generation coefficient, energy storage system charge-discharge coefficient, line impedance parameters, etc.; obtain the initial topology structure of the power distribution network;

[0009] Step 2, based on the disaster information and photovoltaic historical scenarios obtained in step 1, obtain the failure scenario and its probability distribution under disaster and the expected value of photovoltaic unit output;

[0010] Step 3, for the change of the topology structure of the power distribution network, consider the power distribution network tree topology constraint, power balance constraint under topology reconstruction, voltage balance constraint, and power grid operation constraint, energy storage operation constraint, photovoltaic unit operation constraint, gas turbine operation constraint and gas turbine climbing power constraint each power equipment operation constraint;

[0011] Step 4, based on the system operation constraints and each power equipment operation constraint in step 3, taking the minimum total system operation cost as the objective function, a power distribution network topology reconstruction model based on extreme disaster scenarios is constructed;

[0012] Step 5, based on the power distribution network topology reconstruction model based on extreme disaster scenarios in step 4, combined with the probability distribution of the failure scenario and the expected value of the photovoltaic unit output, the model is solved by using a mixed integer optimization solver, and the power distribution network is scheduled according to the solution to obtain the optimal topology reconstruction strategy of the power distribution network based on extreme disaster scenarios.

[0013] Further, in step 3, the various constraint conditions are:

[0014] 1) Power distribution network tree topology constraint

[0015]

[0016]

[0017]

[0018]

[0019]

[0020] wherein the integer variable represents the radial virtual power flow of the distribution network; F(i) represents the set of terminal nodes with node i as the head node; T(i) represents the set of head nodes with node i as the terminal node; N bus is the set of upper grid nodes; N pcc is the set of nodes where the distribution network is connected to the upper grid; z ij is a 0-1 variable representing the on-off state of the line ij, z ij = 1, indicating that the line is on, z ij = 0, indicating that the line is off; |N|-1 is the virtual power flow out of the root node, which is a positive integer; L is the set of branches of the distribution network; L pcc is the set of branches connected to the nodes connected to the upper grid; equation (1) indicates that a total of |N|-1 units of virtual power flow leave the root node to form a tree network; equation (2) ensures the connectivity of the tree structure of the distribution network, with the virtual power flow value decreasing by 1 after passing through each node; equation (3) indicates that the number of branches in the spanning tree is less than the number of nodes by 1; equation (4) ensures that the virtual power flow on the unconnected branch is 0; equation (5) ensures that the line connected to the root node is in a connected state; N is the total number of nodes of the distribution network;

[0021] 2) Power balance constraint

[0022]

[0023]

[0024]

[0025]

[0026]

[0027] wherein t is the scheduling time; T is the total scheduling period; P i,t,g , Q i,t,g are the injected active power and injected reactive power of node i under the gth fault scenario, respectively; P ij,t,g , Q ij,t,g are the active power and reactive power flowing through the branch ij under the gth fault scenario, respectively; Rij and X ij are the resistance and reactance value of branch ij, respectively; l ij.t,g is the square of the current flowing through branch ij under the gth fault scenario; is the active power purchased from the upper-level grid by the distribution network; is the reactive power purchased from the upper-level grid by the distribution network; and are the active and reactive power output by the gas turbine under the gth fault scenario, respectively; and are the charging and discharging power of the energy storage under the gth fault scenario, respectively; is the active and reactive power injected by the photovoltaic under the gth fault scenario; is the active and reactive load; is the load shedding power under the gth fault scenario; is the reactive load shedding under the gth fault scenario; V i,t,g is the square of the node voltage amplitude; equations (6)-(9) are the power balance constraints of the distribution network; equation (10) is the branch capacity constraint after the second-order cone relaxation;

[0028] 3) Line connectivity constraints

[0029]

[0030]

[0031]

[0032] -Y ij,g z ij M≤Q ij,t,g ≤Y ij,g z ij M (14)

[0033] -Y ij,g z ij M≤P ij,t,g ≤Y ij,g z ij M (15)

[0034]

[0035] wherein, Y ij,g is a 0-1 parameter indicating that the line is disconnected due to damage caused by each fault, Y ij,g = 0 indicates that the line is damaged due to fault damage, Y ij,g = 1 indicates that the line is not damaged due to the fault; M is a very large number; I max is the maximum transmission current of the line; V min , Vmax Vmin and Vmax are the minimum and maximum node voltage, respectively; Eqs. (11) and (12) calculate the node voltage magnitude, when Y ij z ij = 1, V i,t,g and V j,t,g are equal; when Y ij z ij = 0, V i,t,g and V j,t,g are subject to inequality constraints; Eqs. (13)-(16) constrain the current and power on the lines;

[0036] 4) Energy storage operation constraints:

[0037]

[0038]

[0039]

[0040]

[0041]

[0042]

[0043]

[0044]

[0045] where, is a 0-1 variable representing the state of charge and discharge of the energy storage, represents the energy storage is in charging state, represents the energy storage is in discharging state, represents the energy storage is in not charging state, represents the energy storage is in not discharging state; is the maximum charging and discharging power; Δt is the scheduling interval; is the energy storage capacity under the gth fault scenario; is the upper and lower limit of the energy storage capacity; are the charging and discharging efficiency of the energy storage, respectively; λ max is the maximum number of charging and discharging per day;

[0046] 5) Photovoltaic unit operation constraints:

[0047]

[0048]

[0049]

[0050] wherein, PmaxPV is the photovoltaic unit active power upper limit value; QmaxPV is the photovoltaic unit reactive power upper limit value; CmaxPV is the photovoltaic unit rated capacity;

[0051] 6) upper grid power purchase constraint:

[0052]

[0053] wherein, Pmaxgrid is the maximum power purchased from the upper grid;

[0054] 7) gas turbine output constraint:

[0055]

[0056]

[0057] wherein, PmaxGT is the gas turbine active power upper limit value; QmaxGT is the gas turbine reactive power upper limit value;

[0058] 8) gas turbine ramping power constraint:

[0059]

[0060] wherein, PmaxrampGT is the maximum gas turbine ramping power;

[0061] 9) load shedding constraint:

[0062]

[0063]

[0064] wherein, tan θ L is the load node power factor.

[0065] Further, in step 4, the objective function of the minimum total operation cost of the distribution network is:

[0066]

[0067] wherein, Cpurchase is the power purchase cost; N ess N is the set of nodes connected to the energy storage; Cstorage is the cost coefficient of the energy storage for each charge and discharge; N is the set of nodes connected to the energy storage; g P is the probability of each fault scenario; N busis the set of nodes; η is the cut penalty coefficient; N MT is the set of nodes connected to the gas turbine; is the cost coefficient of the gas turbine; is the cost coefficient of the energy storage charging and discharging; N PV is the set of nodes connected to the photovoltaic; is the penalty cost of the curtailed photovoltaic power; is the curtailed photovoltaic power, which, together with the constraints of equations (1)-(33) in step 2 and the objective function of equation (34), constitutes the distribution network topology stochastic optimization model considering the uncertainty of fault scenarios.

[0068] Further, in step 5, the simplified form of the distribution network topology stochastic optimization model considering the uncertainty of fault scenarios is:

[0069]

[0070]

[0071] Cn g ≤c (37)

[0072]

[0073] Gm+Hn g =w (39)

[0074]

[0075] ||Qn g +P||2≤q T n g +p (41)

[0076] In the formula, is the first-stage scheduling variable; is the second-stage scheduling variable; is the one-stage scheduling cost function; is the two-stage scheduling cost function; is a 0-1 variable representing the connection state of the line; equation (36) is the related constraint of the first-stage scheduling variable, corresponding to equations (22)-(24), (28), wherein E is the unit matrix, equation (37) is the related constraint of the second-stage scheduling variable, corresponding to equations (17)-(21), (25)-(27), (29)-(33), wherein equation (38) is the virtual power flow constraint, corresponding to equations (1)-(5), wherein D= (|N|-1) [E], E=[1] is a matrix with all 1s, and e=[|N|-1]T ; Equation (39) is a coupling constraint of the first stage and the second stage scheduling variables, corresponding to Equations (6)-(9), where G=H=[1], Equation (40) is a line connectivity constraint, corresponding to Equations (11)-(16), where I=J=[2R ij +2X ij ], K=[MY ij ,-MY ij ] T , o=[V i,t,g -V j,t,g ] T ; Equation (41) is a branch power flow second order cone relaxation constraint, corresponding to Equation (10), where Q=q T =2[1], P=p=[l ij,t,g +V i,t,g ] are all coefficient matrices corresponding to second order cone relaxation constraints; the distribution network topology random optimization model considering fault scenario uncertainty is a two-stage optimization model, which belongs to an MIQCP problem, and a solver is directly called for solving.

[0077] Advantages: Compared with the prior art, the technical scheme of the present application has the following advantages:

[0078] Compared with the conventional scheme of fixing the microgrid structure before natural disasters occur, the present application reconstructs the distribution network topology by introducing a virtual power flow method, and combines a photovoltaic system and an energy storage system to form an integrated light-storage model into the power grid, so as to perform collaborative optimization with the minimum total system cost as an objective function. The example test results show that the method proposed in the present application can reduce the total system operation cost, reduce the load shedding amount caused by disasters, and improve the stability and safety of the power system compared with the prior art. BRIEF DESCRIPTION OF DRAWINGS

[0079] Figure 1 is a method flowchart of the present application;

[0080] Figure 2 is an example diagram of the distribution network topology reconstruction model based on extreme disaster scenarios;

[0081] Figure 3 is a comparison diagram of the influence of whether to perform topology reconstruction on the operation results. DETAILED DESCRIPTION

[0082] The present application will be further illustrated below in conjunction with the drawings and specific embodiments, and it should be understood that these embodiments are only used to illustrate the present application and not to limit the scope of the present application, and various modifications of the present application by those skilled in the art after reading the present application all fall within the scope defined by the appended claims.

[0083] AsFigure 1 As shown, the present application provides a power distribution network topology reconfiguration method based on extreme natural disaster scenarios, comprising the following steps:

[0084] Step 1, obtain disaster information, including disaster occurrence location, disaster intensity, disaster path; obtain historical photovoltaic scenarios; obtain power grid parameters, including generator set power generation coefficient, energy storage system charge and discharge coefficient, line impedance parameters, etc.; obtain the initial topology structure of the power distribution network;

[0085] Step 2, based on the disaster information and photovoltaic historical scenarios obtained in step 1, obtain the failure scenario and its probability distribution under the disaster, as well as the expected value of the photovoltaic unit output;

[0086] Step 3, for the change of the topology structure of the power distribution network, consider the power distribution network tree topology constraint, the power balance constraint under topology reconfiguration, the voltage balance constraint, and the operation constraints of the power grid, the energy storage operation constraint, the photovoltaic unit operation constraint, the gas turbine operation constraint and the gas turbine climbing power constraint Various operation constraints of electric power equipment;

[0087] Step 4, based on the system operation constraints and the operation constraints of each electric power equipment in step 3, taking the minimum total system operation cost as the objective function, a power distribution network topology reconfiguration model based on extreme disaster scenarios is constructed;

[0088] Step 5, based on the power distribution network topology reconfiguration model based on extreme disaster scenarios in step 4, combined with the probability distribution of the failure scenario and the expected value of the photovoltaic unit output, the model is solved by using a mixed integer optimization solver, and the power distribution network is dispatched according to the solving result to obtain the optimal topology reconfiguration strategy of the power distribution network based on extreme disaster scenarios.

[0089] Further, in step 3, various constraint conditions are:

[0090] 1) Power distribution network tree topology constraint

[0091]

[0092]

[0093]

[0094]

[0095]

[0096] Wherein, the integer variable represents the power distribution network radial virtual flow; F(i) represents the set of terminal nodes with node i as the first node; T(i) represents the set of head nodes with node i as the terminal node; N busN is the set of superior grid nodes; N pcc z is the set of nodes connected to the superior grid; z ij is a 0-1 variable representing the on-off state of line ij; z ij = 1, indicating that the line is on; z ij = 0, indicating that the line is off; |N|-1 is the virtual flow out of the root node, which is a positive integer; L is the set of distribution network branches; L pcc is the set of branches connected to the grid-connected nodes; equation (1) indicates that a total of |N|-1 units of virtual flow leave the root node to form a tree network; equation (2) ensures the connectivity of the distribution network tree structure, with the virtual flow value decreasing by 1 every time a node is passed through; equation (3) indicates that the number of branches in the spanning tree is less than the number of nodes by 1; equation (4) ensures that the virtual flow on the unconnected branch is 0; equation (5) ensures that the line connected to the root node is in a connected state; N is the total number of nodes in the distribution network;

[0097] 2) Power balance constraint

[0098]

[0099]

[0100]

[0101]

[0102]

[0103] where t is the scheduling time; T is the total scheduling period; P i,t,g and Q i,t,g are the injected active power and injected reactive power of node i under the gth fault scenario, respectively; P ij,t,g and Q ij,t,g are the active power and reactive power flowing through branch ij under the gth fault scenario, respectively; R ij and X ij are the resistance and reactance values of branch ij, respectively; l ij.t,g is the square of the current flowing through branch ij under the gth fault scenario; is the active power purchased by the distribution network from the superior grid; is the reactive power purchased by the distribution network from the superior grid; and are the active and reactive power output by the gas turbine under the gth fault scenario, respectively; and are the charging and discharging power of the energy storage under the gth fault scenario, respectively; is the active and reactive power injected by the photovoltaic under the gth fault scenario; For active and reactive loads; is the load shedding power under the g-th fault scenario; is the reactive load shedding under the g-th fault scenario; V i,t,g is the square of the node voltage amplitude; Equations (6) to (9) are the power balance constraints of the AC nodes in the distribution network; Equation (10) is the branch capacity constraint after second-order cone relaxation;

[0104] 3) Line connectivity constraints

[0105]

[0106]

[0107]

[0108] -Y ij,g z ij M≤Q ij,t,g ≤Y ij,g z ij M (14)

[0109] -Y ij,g z ij M≤P ij,t,g ≤Y ij,g z ij M (15)

[0110]

[0111] Among them, Y ij,g Y is a 0-1 parameter indicating the line is damaged and disconnected due to each fault. ij,g =0 means the line is damaged due to fault damage, Y ij,g =1 means the line is not damaged due to a fault; M is a large number; I max is the maximum value of the line transmission current; V min , V max are the minimum and maximum node voltages respectively; Equations (11) and (12) calculate the node voltages. ij z ij =1, V i,t,g and V j,t,g There is an equality relationship between them; when Y ij z ij =0, V i,t,g and V j,t,g There are inequality constraints between them; Equations (13)-(16) constrain the current and power on the line;

[0112] 4) Energy storage device operation constraints:

[0113]

[0114]

[0115]

[0116]

[0117]

[0118]

[0119]

[0120]

[0121] where, is a 0-1 variable representing the state of charge and discharge of the energy storage, represents that the energy storage is in the charging state, represents that the energy storage is in the discharging state, represents that the energy storage is in the non-charging state, represents that the energy storage is in the non-discharging state; is the maximum charging and discharging power; Δt is the scheduling interval; is the storage power of the energy storage under the gth fault scenario; is the upper and lower limits of the energy storage capacity; are the charging and discharging efficiencies of the energy storage, respectively; λ max is the maximum number of daily charging and discharging times;

[0122] 5) Photovoltaic unit operation constraints:

[0123]

[0124]

[0125]

[0126] where, is the upper limit value of the active power output of the photovoltaic unit; is the upper limit value of the reactive power output of the photovoltaic unit; is the rated capacity of the photovoltaic unit;

[0127] 6) Upper-level grid power purchase constraints:

[0128]

[0129] where, is the maximum power purchased by the grid;

[0130] 7) Gas turbine output constraints:

[0131]

[0132]

[0133] wherein, is the upper limit of the active power of the gas turbine; is the upper limit of the reactive power of the gas turbine;

[0134] 8) the ramping power constraint of the gas turbine:

[0135]

[0136] wherein, is the maximum value of the ramping power of the gas turbine;

[0137] 9) the load shedding constraint:

[0138]

[0139]

[0140] wherein, tanθ L is the power factor of the load node.

[0141] Further, in step 4, the objective function of the minimum total operation cost of the power distribution network is:

[0142]

[0143] wherein, is the cost of purchasing electricity; N ess is the set of nodes connected to the energy storage; is the cost coefficient of the energy storage for each charge and discharge; is the charge and discharge coefficient of the energy storage in the total scheduling period; G is the fault scenario; p g is the probability of occurrence of each fault scenario; N bus is the set of nodes; η is the load shedding penalty coefficient; N MT is the set of nodes connected to the gas turbine; is the cost coefficient of the gas turbine; is the cost coefficient of the charge and discharge of the energy storage; N PV is the set of nodes connected to the photovoltaic; is the penalty cost of abandoned light; is the abandoned light power of the photovoltaic, and the random optimization model of the power distribution network topology considering the uncertainty of the fault scenario is formed by the objective function of formula (34) and the constraint conditions of formula (1)-formula (33) in step 2.

[0144] Further, in step 5, the random optimization model of the power distribution network topology considering the uncertainty of the fault scenario is as follows:

[0145]

[0146]

[0147] Cn g ≤c (37)

[0148]

[0149] Gm+Hn g =w (39)

[0150]

[0151] ||Qn g +P||2≤q T n g +p (41)

[0152] In the formula, is the first-stage scheduling variable; is the second-stage scheduling variable; is the one-stage scheduling cost function; is the two-stage scheduling cost function; is a 0-1 variable representing the line connection state; formula (36) is the related constraint of the first-stage scheduling variable, corresponding to formula (22)-(24), formula (28), wherein E is a unit matrix, formula (37) is the related constraint of the second-stage scheduling variable, corresponding to formula (17)-(21), formula (25)-(27), formula (29)-(33), wherein formula (38) is a virtual flow constraint, corresponding to formula (1)-(5), wherein D= (|N|-1) [E], E=[1] is a matrix full of 1s, and e=[|N|-1] T ; formula (39) is the coupling constraint of the first-stage and second-stage scheduling variables, corresponding to formula (6)-(9), wherein G=H=[1], formula (40) is the line connection constraint, corresponding to formula (11)-(16), wherein I=J=[2R ij +2X ij ] K=[MY ij ,-MY ij ] T , o=[V i,t,g -V j,t,g ] T ; formula (41) is the branch flow second-order cone relaxation constraint, corresponding to formula (10), wherein Q=q T=2[1]、P=p=[l ij,t,g +V i,t,g ] are all coefficient matrices corresponding to the second-order cone relaxation constraints; the distribution network topology stochastic optimization model considering the uncertainty of fault scenarios is a one- or two-stage optimization model, which belongs to the MIQCP problem and is solved by directly calling the solver.

[0153] Case Analysis

[0154] The following example illustrates the superiority of the distribution network topology reconstruction method based on extreme disaster scenarios described in the present invention. Figure 2 The improved IEEE 33-bus power system is shown. To compare the superiority of the proposed method, the distribution network system was dispatched with and without topology reconstruction. The present invention is implemented using the GAMS optimization platform and the Gurobi solver for solving nonlinear programming problems.

[0155] Table 1 Failure scenarios and probabilities

[0156]

[0157] This example assumes that the failure scenarios caused by extreme weather and their probability distributions are known (see Table 1). Based on this example, the prediction performance and dispatch results of the distribution network topology reconstruction scheduling scheme based on extreme disaster scenarios are given (see the results in Table 1). Figure 2 ) and a comparison of the scheduling results with and without topology reconstruction (see Table 2 for the results). In the scheduling scheme with topology reconstruction performed before the disaster, in fault scenario g1, line 13-14 was actively disconnected by the upper-layer topology optimization. After line 28-29 was disconnected due to the fault, the grid formed a partial island and was supplied by distributed power generation. The load reduction in this scenario was the most severe. In fault scenario g2, lines 18-33 and 21-22 were both actively disconnected by topology optimization before the disaster, and the load reduction in this scenario was the least. In fault scenario g3, line 9-10 was disconnected due to the fault, and the distribution network formed a partial island. The load reduction in this scenario was in the middle. Comparing the optimized scheduling results before and after topology reconstruction, it can be seen that the topology reconstruction before the disaster reduced the total operating cost by 2.67% and the load reduction by 24.28%, effectively ensuring the distribution network in the event of a fault and reducing the overall operating cost of the system.

[0158] Table 2 Comparison of the impact of topology reconstruction on running results

[0159]

[0160] The above merely illustrates the specific embodiments of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of the changes or replacements within the technical range disclosed by the present application, which should be covered in the protection scope of the present application.

Claims

1. A distribution network topology reconstruction method based on extreme natural disaster scenarios, characterized in that: The following steps are involved: Step 1: Obtain disaster information, including the location, intensity, and path of the disaster; obtain historical photovoltaic scenarios; obtain grid parameters, including the power generation coefficient of the generator set, the charge and discharge coefficient of the energy storage system, and the line impedance parameters; and obtain the initial topology of the distribution network. Step 2: Based on the disaster information and historical photovoltaic scenarios obtained in step 1, the fault scenarios and their probability distribution under the disaster and the expected value of the photovoltaic unit output are obtained; Step 3: Consider the tree-like topology constraints of the distribution network, the power balance constraints under topology reconstruction, the voltage balance constraints, the grid operation constraints, the energy storage operation constraints, the photovoltaic unit operation constraints, the gas turbine operation constraints and the gas turbine ramp power constraints, and the operation constraints of various power equipment in response to changes in the distribution network topology. Step 4: Based on the system operation constraints and the operation constraints of each power equipment in step 3, and taking the minimum overall system operation cost as the objective function, a distribution network topology reconstruction model based on extreme disaster scenarios is constructed; Step 5: Based on the distribution network topology reconstruction model based on extreme disaster scenarios in step 4, combined with the probability distribution of fault scenarios and the expected value of photovoltaic unit output, the model is solved using a mixed integer optimization solver. The distribution network is dispatched according to the solution to obtain the optimal distribution network topology reconstruction strategy based on extreme disaster scenarios; In step 4, the objective function for minimizing the overall operating cost of the system is: in, is the electricity purchase cost; N ess A collection of nodes connected to energy storage; is the cost coefficient of each charge and discharge of energy storage; is the energy storage charge and discharge coefficient in the total dispatch period; G is the fault scenario; p g is the probability of each failure scenario; N bus is a set of nodes; η is the load shedding penalty coefficient; N MT is the set of nodes connected to the gas turbine; is the gas turbine cost coefficient; is the cost coefficient of energy storage charging and discharging; N PV is the set of nodes connected to photovoltaics; Penalty cost for abandoned light; is the photovoltaic curtailment power; t is the scheduling time; T is the total scheduling period, To purchase active power from the upper power grid for the distribution network, is the active power output of the gas turbine under the g-th fault scenario, and are the energy storage charging and discharging power under the g-th fault scenario, is the load shedding power under the g-th fault scenario, and Δt is the scheduling interval.

2. A distribution network topology reconstruction method based on extreme natural disaster scenarios according to claim 1, characterized in that: In step 3, the various constraints are: 1) Distribution network tree topology constraints Among them, integer variables Represents the radial virtual power flow of the distribution network; F(i) represents the set of terminal nodes with node i as the first node; T(i) represents the set of first-end nodes with node i as the last node; N bus is the set of upper-level power grid nodes; N pcc is the node set of the distribution network connected to the upper power grid; ij is a 0-1 variable representing the on-off state of line ij, z ij =1, indicating that the line is conductive, z ij =0, indicating that the line is disconnected; |N|-1 is the virtual power flow out of the root node, which is a positive integer; L is the set of distribution network branches; L pcc is the set of branches connected to the grid-connected node; Formula (1) indicates that a total of |N|-1 units of virtual power flow leave the root node to form a tree network; Formula (2) ensures the connectivity of the distribution network tree structure, and the virtual power flow value decreases by 1 every time it passes through a node; Formula (3) indicates that the number of branches in the spanning tree is 1 less than the number of nodes; Formula (4) ensures that the virtual power flow on the unconnected branch is 0; Formula (5) ensures that the line connected to the root node is in a connected state; N is the total number of nodes in the distribution network; 2) Power balance constraints Among them, t is the scheduling time; T is the total scheduling period; P i,t,g , Q i,t,g are the injected active power and injected reactive power of node i under the g-th fault scenario; P ij,t,g , Q ij,t,g are the active power and reactive power flowing through branch ij under the g-th fault scenario; R ij and X ij are the resistance and reactance of branch ij respectively; l ij,t,g is the square of the current flowing through branch ij under the g-th fault scenario; Purchase active power from the upper power grid for the distribution network; Purchase reactive power from the upper power grid for the distribution network; and are the active and reactive power output of the gas turbine under the g-th fault scenario respectively; and are the energy storage charging and discharging power under the g-th fault scenario respectively; The active power and reactive power injected by PV in the g-th fault scenario; For active and reactive loads; is the load shedding power under the g-th fault scenario; is the reactive load shedding under the g-th fault scenario; V i,t,g is the square of the node voltage amplitude; Equations (6) to (9) are the power balance constraints of the AC nodes in the distribution network; Equation (10) is the branch capacity constraint after second-order cone relaxation; 3) Line connectivity constraints Among them, Y ij,g Y is a 0-1 parameter indicating the line is damaged and disconnected due to each fault. ij,g =0 means the line is damaged due to fault damage, Y ij,g =1 means the line is not damaged due to a fault; M is a large number; I max is the maximum value of the line transmission current; V min , V max are the minimum and maximum node voltages respectively; Equations (11) and (12) calculate the node voltages. ij z ij =1, V i,t,g and V j,t,g There is an equality relationship between them; when Y ij z ij =0, V i,t,g and V j,t,g There are inequality constraints between them; Equations (13)-(16) constrain the current and power on the line; 4) Energy storage operation constraints: in, is a 0-1 variable representing the energy storage charge and discharge state, Indicates that the energy storage is in charging state. Indicates that the energy storage is in the discharge state. Indicates that the energy storage is not charging. Indicates that the energy storage is in a non-discharging state; is the maximum charge and discharge power; Δt is the scheduling interval; is the stored energy capacity under the g-th fault scenario; The upper and lower limits of energy storage capacity; are the energy storage charging and discharging efficiency; max The maximum number of charge and discharge times per day; 5) PV unit operation constraints: in, The upper limit of active power output of photovoltaic units; The upper limit of reactive power output of photovoltaic units; is the rated capacity of the photovoltaic unit; 6) Constraints on power purchase from higher-level power grids: in, The maximum power for purchasing electricity from the grid; 7) Gas turbine operating constraints: in, The upper limit of the active power output of the gas turbine; is the upper limit of reactive power output of gas turbine; 8) Gas turbine ramp power constraints: in, is the maximum value of the gas turbine climbing power; 9) Load shedding constraints: Among them, tanθ L is the load node power factor.

3. The method for reconfiguring the distribution network topology based on extreme natural disaster scenarios according to claim 2, characterized in that: The abbreviated form of the distribution network topology stochastic optimization model considering the uncertainty of fault scenarios is: Cn g ≤c (37) Gm+Hn g =w (39) ||Qn g +P||2≤q T n g +p (41) Where, Schedule variables for the first stage; Scheduling variables for the second phase; is the one-stage scheduling cost function; is the two-stage scheduling cost function; is a 0-1 variable representing the line connectivity status; Equation (36) is the related constraint of the first-stage scheduling variable, corresponding to Equations (22)-(24) and (28), where E is the identity matrix, Equation (37) is the relevant constraint of the second-stage scheduling variables, corresponding to Equations (17)-(21), (25)-(27), and (29)-(33), where Equation (38) is the virtual power flow constraint, corresponding to Equations (1)-(5), where D = (|N|-1)E, e=[|N|-1] T ; Equation (39) is the coupling constraint of the scheduling variables in the first and second stages, corresponding to Equations (6)-(9), where G = H = [1], Equation (40) is the line connectivity constraint, corresponding to equations (11)-(16), where I = J = [2R ij +2X ij ],K=[MY ij ,-MY ij ] T , o=[V i,t,g -V j,t,g ] T ; Equation (41) is the second-order cone relaxation constraint of the branch flow, corresponding to Equation (10), where Q = q T =2[1]、P=p=[l ij,t,g +V i,t,g ] are all coefficient matrices corresponding to the second-order cone relaxation constraints; the distribution network topology stochastic optimization model considering the uncertainty of fault scenarios is a one- or two-stage optimization model, which belongs to the MIQCP problem and is solved by directly calling the solver.

Images