Power distribution network reconstruction method and long-term time sequence operation simulation method for power distribution network

Abstract

The application belongs to the technical field of electrical engineering, and discloses a power distribution network reconstruction method and a long-term time sequence operation simulation method of the power distribution network. The reconstruction method comprises the following steps: a power distribution network reconstruction optimization model is constructed according to the branches and switches in the power distribution network; state variables of the branches, current direction variables of the branches and switch action variables are introduced into the model, and constraint conditions are constructed based on the above variables; in addition to the constraints related to the operation of the power grid, branch state and power supply direction constraints, switch action constraints and multi-source radiation constraints are also constructed; the multi-source radiation constraint ensures that each non-source node can obtain power supply from a source node through a closed branch, thereby avoiding islanding, and also excludes closed loops, so that the reconstructed network maintains a multi-source radiation structure. Solving the constructed power distribution network reconstruction optimization model can reconstruct the power distribution network for the infeasible scenario of time sequence operation simulation, thereby improving the accuracy of the time sequence operation simulation of the power distribution network.

Description

Technical Field

[0001] This invention belongs to the technical field of electrical engineering, and more specifically, relates to a power distribution network reconfiguration method and a long-term time-series operation simulation method for power distribution networks. Background Technology

[0002] With the continuous growth of new energy power generation such as wind and solar power, a large number of new resources such as distributed renewable energy, electric vehicles, and flexible loads are being integrated into the distribution network, causing the traditional distribution network to gradually evolve into an active distribution network. Compared with the traditional distribution network, which is mainly based on unidirectional power flow and passive power supply, the spatiotemporal coupling relationship of the active distribution network is more complex. Wind and solar power output is significantly affected by weather, seasons, and intraday fluctuations, while load levels are significantly affected by user behavior and meteorological conditions. Therefore, the operation status of the distribution network is no longer a static, single-moment problem, but a medium- to long-term continuous operation problem with obvious time-series characteristics. In order to accurately assess the operational safety, renewable energy absorption capacity, and power supply reliability of the distribution network throughout the year, it is necessary to continuously simulate operating variables such as load, renewable energy, main grid power purchase, line power flow, and node voltage.

[0003] Existing long-term time-series simulation methods for distribution networks typically employ a fixed network topology for year-round operational verification. This means that the distribution network maintains its original wiring configuration throughout the year, and the operational optimization problem for each time period or rolling window is solved sequentially under this topology. While this method is relatively simple in its modeling process, it fails to adequately reflect the switching operation capabilities of actual distribution network dispatching. In real-world distribution networks, when issues such as line overload, localized low voltage, limited distributed renewable energy absorption, or increased load transfer pressure occur during a given period, dispatchers typically do not immediately determine system infeasibility or directly disconnect loads. Instead, they prioritize altering the network topology through coordinated operations of tie switches and sectionalizing switches to achieve power flow transfer, load transfer, and voltage improvement. Therefore, if a fixed topology is consistently used in year-round time-series simulations, it is easy to misjudge some scenarios that could be recovered through network reconfiguration as system infeasibility scenarios. This leads to deviations from actual assessments of distribution network operational risks, renewable energy absorption capacity, and energy storage demand.

[0004] Therefore, how to implement distribution network reconfiguration for scenarios where time-series operation simulation is not feasible in order to improve the accuracy of distribution network time-series operation simulation is a technical problem that urgently needs to be solved. Summary of the Invention

[0005] In view of the above-mentioned defects or improvement needs of the existing technology, the present invention provides a distribution network reconfiguration method and a distribution network long-term time-series operation simulation method. The purpose is to realize distribution network reconfiguration for scenarios where time-series operation simulation is not feasible, so as to improve the accuracy of distribution network time-series operation simulation.

[0006] According to a first aspect of the present invention, a distribution network reconfiguration method for scenarios where time-series operation simulation is infeasible is provided, comprising: Step 1: Construct a distribution network reconfiguration optimization model based on the branch and switch distribution in the distribution network. The distribution network is a radial distribution network, and the branches include physical lines and switch branches. The switch branches are equivalent to zero-impedance virtual branches connecting the two ends of the switch. The distribution network reconfiguration optimization model defines Let k be the state variable of branch k, with values ​​of 0 and 1 representing the branch being open and closed, respectively; defined and Let k be the current direction variable in branch k, and its superscript is... Indicates the direction from the first node to the last node of a branch, indicated by the superscript. This indicates the direction from the end node of the branch to the beginning node. The current direction variable takes values ​​of 0 and 1, representing no current flowing in the corresponding direction and current flowing in the corresponding direction, respectively. The switching action variable for the zero-impedance virtual branch k has a value of 0 and 1, which represent the switch not being switched and the switch being switched, respectively. The reconfiguration objective of the power distribution network reconfiguration optimization model is to minimize the reconfiguration operation cost; The reconfiguration constraints of the distribution network reconfiguration optimization model include: branch status and power supply direction constraints, switch operation constraints, multi-source radiation constraints, branch power flow constraints, voltage drop relationship between the beginning and end of the branch, node power balance constraints, node voltage upper and lower limits constraints, and load shedding constraints. The branch status and power supply direction constraints include: ; The switching action constraints include: , ; To reconstruct the original state of the preceding branch k; The multi-source radiation constraint introduces two virtual connected flow variables in different directions for branch k. and ,Include: ; ; ; ; ; In the formula, i is the node index, and k is the branch index. For the set of distribution network nodes, For the source node set, The set of non-source nodes remaining after removing the source node set from the distribution network node set. Let M be the set of distribution network branches, and M be a preset positive number. Let them represent the first and last nodes of branch k, respectively. Let node i be the first node of branch k. Let node i be the terminal node of branch k.

[0007] Step 2: Solve the power distribution network reconfiguration optimization model to obtain the power distribution network reconfiguration topology and its corresponding operation simulation results.

[0008] According to a second aspect of the present invention, a long-term time-series operation simulation method for a distribution network embedded with a network reconfiguration mechanism is provided, comprising: S1. Divide the entire time period to be simulated into sub-time periods, each sub-time period being a time window. Take the first time window as the current time window and jump to S2. S2. For the current time window, first determine whether there is a risk of exceeding the limit in the original topology of the power grid in the current time window. If there is a risk of exceeding the limit in the current time window, jump to S3. Otherwise, obtain the simulation results of the current time window according to the original topology of the power grid and jump to S4. S3. Execute the above-mentioned distribution network reconfiguration method for time-series simulation of infeasible scenarios to realize network reconfiguration and output the simulation results of the reconfigured distribution network topology under the current time window; S4. Determine if this is the last time window. If not, proceed to the next time window as the current time window and jump to S2. Otherwise, end the simulation and summarize the simulation results for the entire time period.

[0009] According to a third aspect of the present invention, an electronic device is provided, including a memory and a processor, the memory storing a computer program, wherein the processor executes the computer program to implement the steps of the distribution network reconfiguration method for time-series operation simulation of infeasible scenarios as described above, or to implement the steps of the distribution network long-term time-series operation simulation method with embedded network reconfiguration mechanism as described above.

[0010] According to a fourth aspect of the present invention, a computer-readable storage medium is provided having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described distribution network reconfiguration method for scenarios where time-series operation simulation is infeasible, or implements the steps of the above-described distribution network long-term time-series operation simulation method with embedded network reconfiguration mechanism.

[0011] Overall, compared with the prior art, the technical solutions conceived in this invention have the following beneficial effects.

[0012] 1. The proposed distribution network reconfiguration method for time-series operation simulation infeasibility scenarios constructs a distribution network reconfiguration optimization model based on the branch and switch distribution in the distribution network. This model introduces branch state variables, branch current direction variables, and switch action variables as relevant variables for distribution network reconfiguration. Constraints are then constructed based on these variables. In addition to constraints related to network operation such as branch power flow constraints, voltage drop relationship constraints between the beginning and end of a branch, node power balance constraints, node voltage upper and lower limits constraints, and load shedding constraints, constraints related to branch state and power supply direction, switch action constraints, and multi-source radiation constraints are also constructed. The branch state and power supply direction constraints ensure that only one power supply direction can be selected when a branch is closed, and both directions are invalid when a branch is open. The switch action constraints establish the correlation between switch action variables and branch state variables. The multi-source radiation constraint is a crucial constraint for network reconfiguration. On the one hand, it ensures that each non-source node can obtain power from a source node through a closed branch, thus avoiding islanding; on the other hand, it eliminates closed loops, ensuring that the reconfigured network maintains a multi-source radiation structure, meeting the requirements for distribution network reconfiguration. Based on the above-constructed distribution network reconfiguration optimization model, by optimizing the solution with the goal of minimizing the reconfiguration operation cost, distribution network reconfiguration can be achieved for scenarios where time-series operation simulation is not feasible, thereby improving the accuracy of distribution network time-series operation simulation and minimizing costs.

[0013] 2. The long-term time-series operation simulation system for distribution networks with an embedded network reconfiguration mechanism proposed in this invention first identifies out-of-limit windows, then performs network topology reconfiguration for out-of-limit windows and generates simulation operation results for the corresponding windows. After the current window ends, it enters the next window and re-assesses the risk of the original network topology. The simulation is performed window by window until all time-series simulations are completed. The above method forms a closed-loop simulation mechanism that prioritizes the operation of the original topology, automatically triggers reconfiguration for abnormal windows, and restores the normal topology after the risk is mitigated. This improves the realism of the annual operation simulation and provides a more accurate operational basis for subsequent energy storage planning, new energy consumption assessment, and distribution network operation safety analysis. Attached Figure Description

[0014] Figure 1 This is a flowchart of the steps of a power distribution network reconfiguration method for a time-series operation simulation infeasible scenario according to an embodiment of the present invention.

[0015] Figure 2 This is a comparison diagram of the power grid before and after reconfiguration in one embodiment of the present invention. (a) represents the network before reconfiguration, and (b) represents the network after reconfiguration.

[0016] Figure 3 This is a flowchart of the steps of a long-term time-series operation simulation method for a distribution network with an embedded network reconfiguration mechanism in one embodiment of the present invention.

[0017] Figure 4 This is a schematic diagram of time window division in one embodiment of the present invention.

[0018] Figure 5 This is a schematic diagram of the segmented solution of the time-series simulation model in one embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0020] Example 1 like Figure 1 The diagram shows a flowchart of a power distribution network reconfiguration method for a time-series operation simulation infeasible scenario according to an embodiment of the present invention, which mainly includes the following steps.

[0021] Step 1: Construct a distribution network reconfiguration optimization model based on the branch and switch distribution in the distribution network. The distribution network is a radial distribution network. Branches include physical lines and switch branches. Switch branches are equivalent to zero-impedance virtual branches connecting the two ends of the switch. The reconfiguration objective of the distribution network reconfiguration optimization model is to minimize the reconfiguration operation cost. The reconfiguration constraints of the distribution network reconfiguration optimization model include: branch status and power supply direction constraints, switch operation constraints, multi-source radiation constraints, branch power flow constraints, voltage drop relationship between the beginning and end of the branch constraints, node power balance constraints, node voltage upper and lower limits constraints, and load shedding constraints.

[0022] Define the distribution network node set as: ... (1); The above distribution network node set includes ordinary nodes and source nodes. Ordinary nodes represent loads, distributed power sources, or network connection points, while source nodes represent infinite power source access points.

[0023] The set of source nodes is denoted as: (2); Define the set of branches as: ... (3); Among them, each branch road Connect the first node and end nodes : ... (4); In actual modeling, all branches are not entirely equivalent to physical lines, but are composed of two types of branches: ... (5); in: A collection of physical circuits; This is a set of virtual branches for switches.

[0024] For physical line sets The parameters of branch k in the equation include: ; in: Line resistance; For line reactance; This is the maximum capacity of the branch line; This is the upper limit of the branch current.

[0025] Physical lines serve both network connectivity and power flow transmission functions, thus requiring the application of complete distribution network DistFlow constraints, including active power flow, reactive power flow, voltage drop, current square, line loss, and second-order cone constraints.

[0026] For a physical circuit, if it is closed, power flow is allowed; if it is open, active power, reactive power, and current are all forced to zero.

[0027] A virtual branch of a switch is a zero-impedance virtual branch that connects the two ends of a switch; in other words, a virtual branch of a switch is actually a switch. The difference between a switch branch and a physical circuit is: 1) Switch branches primarily describe the connection status, rather than the actual line impedance; 2) The impedance of the switch branch should be zero or at a minimum value, and there should be no actual line loss. 3) The voltages at both ends of the closed switch should be approximately equal; 4) Switching branches are usually not subject to full DistFlow second-order cone constraints to avoid numerical problems caused by zero-impedance branches.

[0028] Therefore, for virtual branches of switches, the model focuses on constraining their on / off states, power flow start / stop, and voltage consistency during closure, rather than treating them as ordinary physical lines.

[0029] For each branch road Define a binary state variable: 1 indicates a closed branch; 0 indicates an open branch, i.e.: ... (6); ... (7); Define the initial state of the branch as The original state of the branch is denoted as the state of the branch before reconstruction, and is a known quantity; For uncontrollable branches: ... (8); For controllable branches: ... (9); in It is a set of controllable branches, including virtual branches of switches that are allowed to participate in reconfiguration, such as feeder outlet switches, tie switches, and sectionalizing switches.

[0030] In real-world scenarios, automated switches in power distribution networks are typically categorized by function into feeder outlet switches, tie switches, sectionalizing switches, and branch switches.

[0031] The node where the feeder outlet switch is located, or its equivalent connection point, can be considered as a source node. The modeling principle for the feeder outlet switch is: directly model the feeder outlet switch as a source node, and define the main grid power purchase variables at each source node, including the main grid active power and reactive power. Define the voltage at the source node as a fixed value. ,in It is usually set to 1.0 pu.

[0032] Interchange switches are typically located between different feeders, zones, or power supply paths, and are usually open during normal operation. Their main function is to close in case of faults, maintenance, heavy overloads, or voltage exceeding limits, thus establishing a load transfer path. For interchange switches... Typically, there are: the original state If the contact switch is closed during the reconfiguration process, then At this point, a new power supply path will be added to the network. To avoid forming a loop, the model must simultaneously disconnect at least one other switch so that the reconstructed network still maintains a radial structure. Therefore, the typical reconstructing function of a tie switch is to transfer the power supply path.

[0033] Section switches are typically installed on the main feeder line to divide the feeder into multiple power supply sections. For section switches... Typically, the initial state is a closed state, that is... During reconfiguration, the sectionalizing switch can be disconnected to change the power supply source of a feeder section or isolate a section of the network. Its main function is: 1) When the tie switch is closed, the original power supply path is cut off to avoid network loops; 2) Transfer the load from one feeder to another; 3) Change the power supply zone boundaries; 4) To alleviate overload or voltage overrun on certain branches.

[0034] Branch switches are typically located at feeder branches or tributaries and are used to control whether a specific branch load area is connected to the main network. Under normal circumstances, branch switches are mostly closed, and their initial state is... .

[0035] In the reconfiguration model, branch switches can be used for local isolation or changing branch access paths. However, in strict network reconfiguration mode, branch switches cannot be disconnected because load shedding is not allowed; therefore, branch switch switching is not considered in strict network reconfiguration mode.

[0036] To adapt to the multi-source distribution network and the changes in power supply direction after reconfiguration, each branch... Introduce two directional variables: , ;in, This indicates whether current flows from the beginning node to the end node of a branch. This indicates whether there is current flowing from the end node to the beginning node. A value of 1 indicates that there is current, and a value of 0 indicates that there is no current. The above direction variables and branch states satisfy: ... (10); This constraint means: 1) If the branch is disconnected, i.e. If both directional variables are 0, then both directional variables are 0; 2) If the branch is closed, that is Then only one power supply direction can be selected; 3) This direction is determined by the optimization model based on the reconstructed power supply path.

[0037] In addition, switch action variables are defined: , indicating a branch Whether the state has changed relative to the original state is represented by a value of 0 indicating no change and a value of 1 indicating a change. The relationship between the switch action variable and the binary state variable can be expressed as: ... (11); Converted to linear representation: (12); ... (13); For uncontrollable branches: ... (14); In addition, for each branch road and each time period Define power flow variables in two directions: ; ,in, Indicates a branch exist The contributing current in the direction; Indicates a branch exist Reactive power flow in the direction of the current; This represents the square of the current in that direction; , , The corresponding variable represents the opposite direction.

[0038] The purpose of introducing bidirectional power flow variables is to enable the same branch to determine the power flow direction based on the actual power supply path under different reconfiguration topologies.

[0039] For each node and each time period Define the squared variable of the node voltage: The model uses a voltage squared variable instead of a voltage magnitude variable to simplify the model and to maintain consistency with the second-order cone power flow form of DistFlow.

[0040] For the source node Define the main grid power purchase variables: active power and reactive power of main grid power purchase. .

[0041] Non-source nodes are not allowed to inject power directly from the main network: ... (15); For wind-solar nodes, the available wind and solar power output is defined as follows: The variable of abandoning wind and solar power is The actual wind and solar power input is: ... (16); Define active load shedding variables Define reactive load reduction variables. : Its constraints are: ... (17); ... (18); In the formula, Indicates the upper limit of active power load; This indicates the upper limit for reactive power load reduction.

[0042] The state variables of the above branches Directional variables , Switch action variable All power flow variables are decision variables. ; All of these are variables to be decided.

[0043] The model aims to minimize the overall system operating cost while satisfying distribution network operation constraints and topology constraints. The objective function includes the main grid power purchase cost, line loss cost, switching operation cost, renewable energy curtailment penalty, active power load shedding penalty, and reactive power load reduction penalty.

[0044] First, the definitions of the key parameters in the model are given, where the subscript i represents the node index, the subscript k represents the branch index, and the subscript t represents the time period index; The following sets will be involved in the modeling: : Set of distribution network nodes; : The set of source nodes; Branch set; : Collection of physical circuits; : Set of virtual branch circuits for switches; : Set of controllable branches; , : These represent the first and last nodes of branch k, respectively; : The state variable of branch k; : These represent the power supply direction variables of branch k along the ft and tf directions, respectively; : The switching action variable of branch k; : These represent the virtual connected flows of branch k along the ft and tf directions, respectively; M: Represents a sufficiently large positive number used for constraint linearization and constraint relaxation; : Active and reactive power directly injected into node i from the main network; : Represents the available active power output of photovoltaic power and the available active power output of wind power at node i in time period t; : Represents the curtailment of solar and wind power at node i during time period t; : Represents the active power load demand of node i during time period t; : Represents the reactive load demand of node i in time period t; , : Represents the active power load shedding demand and reactive power load shedding demand at node i during time period t; The maximum active power load shedding limit for node i; : The upper limit of reactive power load shedding at node i; : The outflow active power of node i; : The active power flowing into node i; : Net injected active power at node i; : The reactive power flowing out of node i; : The reactive power flowing into node i; : Net injected reactive power at node i; , : These are the squares of the voltage at the first node and the squares of the voltage at the last node of branch k, respectively; The resistance of branch k; Reactance of branch k; The maximum capacity of branch k; The upper limit of the current in branch k; , : Active power flow of branch k in the corresponding direction; , : Reactive power flow of branch k in the corresponding direction; , : The square of the current in the corresponding direction of branch k.

[0045] In one embodiment, the reconstruction objective function can be written as: ... (19); The main grid electricity purchase cost is: ... (20); The cost of line loss is: ........... (twenty one); The cost of switching action is: ........... (twenty two); The penalty for abandoning renewable energy is: ........... (twenty three); The load shedding penalty is: ........... (twenty four); In the formula, For time index, For the length of time period, This represents the total number of simulation moments. For line capacity, Penalty for line loss, As a penalty for the switching action, As punishment for abandoning light, For the purpose of effective load reduction penalty, This is a penalty for reactive power load shedding.

[0046] In the strict mode of network reconfiguration, since load shedding is not allowed and the upper limits in equations (17) and (18) above are all forced to be zero, the active load shedding penalty and reactive load reduction penalty in the objective function are both zero.

[0047] The reconfiguration constraints include: branch status and power supply direction constraints, switch operation constraints, multi-source radiation constraints, branch power flow start and stop constraints, physical line second-order cone power flow constraints, physical line voltage drop relationship constraints, switch branch voltage drop relationship constraints, node active power balance constraints, node reactive power balance constraints, node voltage upper and lower limits constraints, main grid power purchase upper and lower limits constraints, new energy output and curtailment upper and lower limits constraints, and load shedding constraints.

[0048] Constraint 1_1: Branch status and power supply direction constraints.

[0049] For any branch ,have: ... (25); ... (26); This constraint ensures that only one power supply direction can be selected when the branch is closed, and both directions are invalid when the branch is open.

[0050] For uncontrollable branches, we have: (27); For controllable branches, we have: (28).

[0051] Constraint 1_2: Switch action constraint.

[0052] For controllable branches To simplify the switching action into a linear constraint, the switching action variable satisfies: ... (29); ... (30); (31).

[0053] Constraint 1_3: Multi-source radiation constraint.

[0054] A radial distribution network refers to a distribution network in which all branches are connected in a loop-free tree structure during normal operation, power is transmitted unidirectionally from the power source to each load node along a single path, and any non-power source node is powered by only one upstream branch.

[0055] Therefore, the reconfigured distribution network should maintain a multi-source radial structure. Due to the presence of multiple infinitely large power source nodes, the network is not necessarily a single tree, but rather a radial forest rooted at multiple source nodes. Let the set of source nodes be... Then the number of closed branches should satisfy: ... (32); This constraint states that in a multi-source radial forest, the number of closed branches equals the number of nodes minus the number of source nodes.

[0056] This constraint alone cannot completely guarantee the absence of islands and loops, therefore, a virtual connected flow variable is further introduced: ... (33); It satisfies the following with the direction variable: ... (34); ... (35); in, It is a sufficiently large constant.

[0057] For non-source nodes It is required to obtain a unit virtual supply current from the source node: ...(36); For the source node set, the total output of the virtual stream satisfies: ... (37); The meaning of virtual flow constraint is that each non-source node must be able to obtain power from a source node through a closed branch, thereby avoiding islanding; combined with the closed branch number constraint, closed loops can be eliminated, so that the reconstructed network maintains a multi-source radial structure.

[0058] Constraint 1_4: Branch power flow start / stop constraint.

[0059] For each branch road and time period ,have: ... (38); ... (39); ... (40); The same logic applies in the opposite direction: ... (41); ... (42); ... (43); This set of constraints guarantees: 1) When the branch is disconnected, The active power, reactive power, and current in both directions of the branch are all zero; 2) When a branch is closed, power flow is only allowed to be transmitted in the selected direction; 3) Branch flow cannot exceed the capacity limit.

[0060] Constraint 1_5: Second-order cone power flow constraint for physical lines.

[0061] For physical circuits The DistFlow model contains: ... (44); Its standard second-order cone form is: ... (45); For the reverse direction of the branch, the DistFlow second-order cone constraint is: ... (46); Its standard second-order cone form is: (47).

[0062] The aforementioned constraint transforms the nonlinear branch power, voltage, and current relationships into a second-order cone form, giving the model a mixed-integer second-order cone programming structure. This constraint is the core of the model's mixed-integer second-order cone programming nature. It is important to note that this second-order cone constraint is primarily applied to the physical circuitry; for zero-impedance switch virtual branches, the complete second-order cone constraint is not applied.

[0063] Constraint 1_6: Constraint on the voltage drop relationship between the beginning and end of the physical circuit.

[0064] For physical circuits If the direction of the branch road is The voltage drop relationship of DistFlow is as follows: ... (48); In the formula, , These are the voltages at the first and last nodes of branch k, respectively. This represents the square of the current in branch k in the direction of ft.

[0065] After considering the direction variable, the Big-M form is adopted: ...(49); If the direction of the branch road is Then we have: ...(50); When a direction is selected, the corresponding voltage drop constraint takes effect; when the direction is not selected, the constraint is relaxed by Big-M.

[0066] Constraint 1_7: Constraint on the voltage drop relationship between the beginning and end of the switch branch.

[0067] For switch virtual branches A complete voltage drop formula should not be applied as to a physical circuit. If the switch is closed, the voltages at both ends should be approximately equal. ... (51); Big-M format can be used: ... (52); This constraint means: 1) When When the switch is closed, the voltages at both ends are equal; 2) When When the switch is open, the voltage at both ends is no longer forced to be the same; 3) When the switch is open, the power flow start-stop constraint has already ensured that the branch power is zero.

[0068] This approach aligns with the physical meaning of a switch as a zero-impedance connection element and avoids numerical instability caused by zero-impedance branches in second-order conical power flow constraints.

[0069] Constraint 1_8: Node active power balance constraint.

[0070] Define nodes in time periods The net active power injection is: ... (53); in: This indicates that the node absorbs active power load; This indicates that the node has a net active power injection; After considering branch losses, the active power balance of a node can be written as: ... (54); Among them, the active power outflow from the node is: ... (55); The active power flowing into the node is: ... (56); This balance means that the power sent out by a node minus the power received from adjacent branches equals the node's net injected power.

[0071] In the formula, it is stipulated that if This indicates that node i injects active power into the network. This indicates that node i absorbs active power from the network; This represents the outflow active power of node i. This represents the inflow active power at node i. This represents the net injected active power at node i. , Let represent the active power flow of branch k in the ft and tf directions, respectively. , These represent the active power and reactive power directly injected into node i from the main network, respectively. and These represent the available active power output of photovoltaic and wind power at node i during time period t, respectively. and These represent the curtailed solar power and curtailed wind power at node i during time period t, respectively.

[0072] Constraint 1_9: Nodal reactive power balance constraint.

[0073] Define nodes During the period The net reactive power injection is: ... (57); in: This indicates that the node absorbs inductive reactive power; This indicates that the node is injecting capacitive reactive power.

[0074] The reactive power balance of the nodes is as follows: ... (58); in: ... (59); ... (60).

[0075] In the formula, if This indicates that node i is injecting reactive power into the network. This indicates that node i absorbs reactive power from the network. This represents the outflow reactive power at node i. This represents the inflow active power at node i. This represents the net injected reactive power at node i. , These represent the reactive power flow of branch k in the ft and tf directions, respectively.

[0076] Constraint 1_10: Upper and lower limits of node voltage constraints.

[0077] For all nodes and time period ,have: ... (61); In the formula, These are the lower and upper limits of the node voltage, respectively.

[0078] For the source node It can be fixed as: (62).

[0079] Constraint 1_11: Upper and lower limits of electricity purchase from the main grid.

[0080] For the source node The main grid electricity purchase meets the following requirements: ... (63); ... (64); In the formula, These represent the lower and upper limits of the active power directly injected into node i from the main network, respectively. These represent the lower and upper limits of reactive power directly injected into node i from the main grid, respectively. If backfeeding to the upstream grid is not allowed, then: If the upstream power grid is allowed to absorb the reverse power, then the following settings can be configured: .

[0081] For non-source nodes ,have: , .

[0082] Constraint 1_12: Upper and lower limits of renewable energy output and curtailment.

[0083] For new energy nodes: ... (65); ... (66).

[0084] Constraint 1_13: Load shedding constraint.

[0085] The active load shedding constraint is: ... (67); The reactive load reduction constraint is: ... (68); It should be noted that more constraints can be set according to the actual network architecture, such as energy storage constraints.

[0086] Step 2: Solve the distribution network reconfiguration optimization model to obtain the reconfigured distribution network topology and its corresponding operation simulation results.

[0087] Specifically, based on the distribution network reconfiguration optimization model constructed above, the commercial solver Gurobi can be called to solve it and obtain all decision quantities. Among them, the reconfigured distribution network topology can be obtained based on the state variables of each branch. In addition, the operating variables will also be output.

[0088] In one embodiment, the load shedding constraint has load shedding constraints in strict network reconfiguration mode and load shedding constraints in emergency network reconfiguration mode: the load shedding constraint in strict network reconfiguration mode limits the load shedding amount to zero, and the upper limit of the load shedding is set to zero; the load shedding constraint in emergency network reconfiguration mode relaxes the load shedding constraint in strict network reconfiguration mode, allowing load shedding within a preset range, and the upper limit of the load shedding is set to be greater than zero; the distribution network reconfiguration method first performs network reconfiguration in strict network reconfiguration mode, and if it cannot be solved in this mode, it switches to emergency network reconfiguration mode for network reconfiguration.

[0089] Figure 2 The diagram shows a comparison of the power grid reconfiguration before and after in one embodiment of the present invention. The diagram illustrates a typical process of load transfer achieved through the cooperation of one tie switch and multiple sectionalizing switches. Before reconfiguration, node 11 is powered by source A via the 7-3-12-11 path, where sectionalizing switches S2 (3-12) and S1 (12-11) are closed, and tie switch S0 (12-51) is open, with the upper and lower feeders operating independently. When the power flow exceeds the limit in branch L5 (7-3), it indicates that the power supply pressure from source A to the downstream load is too high, requiring topology reconfiguration to reduce the load on this branch. During reconfiguration, the system closes tie switch S0 (12-51) and simultaneously opens sectionalizing switch S2 (3-12). Thus, node 11 is no longer powered by source A via 7-3, but instead is powered by source B via the 44-43-51-12-11 path. Through this operation, the load at node 11 is transferred from feeder A to feeder B, reducing the power flow that originally passed through 7-3, thereby alleviating the over-limit problem of branch L5. The key to this reconfiguration method is to maintain the radial operation of the distribution network. If only the tie switch is closed without opening the sectionalizing switch, a loop network will be formed; if only the sectionalizing switch is opened without closing the tie switch, node 11 may lose power. Therefore, the typical reconfiguration logic of a type A switch cluster is: close a normally open tie switch and simultaneously open a normally closed sectionalizing switch within the same cluster, changing the load power supply path without interrupting power supply, thus achieving branch load reduction and load transfer.

[0090] Example 2 like Figure 3 The diagram shows a flowchart of the long-term time-series operation simulation method for a distribution network with an embedded network reconfiguration mechanism according to an embodiment of the present invention, which includes the following steps.

[0091] S1. Divide the entire time period to be simulated into sub-time periods, each sub-time period being a time window. Take the first time window as the current time window and jump to S2.

[0092] Specifically, considering the complexity of solving medium- and long-term time-series simulations and the difficulty of direct solutions, a piecewise solution method is adopted, such as... Figure 4 As shown, the entire year is divided into several short segments for sequential solving. The result of the last segment of the previous sub-segment is used as the starting state of the next sub-segment, thus satisfying the time-segment coupling constraint and ensuring a complete and feasible solution. Figure 5 Solving each sub-period sequentially and then summing the results of each sub-period yields the complete annual time-series simulation results. Sub-periods are typically set to a day or a week, which is acceptable in terms of computational complexity, but the simulation accuracy can be affected by too many sub-periods. However, rural power systems have fewer lines and units, so longer sub-periods can be set to improve the accuracy of the solution and reduce the number of decompositions, provided that computing power allows.

[0093] In one embodiment, the time-series operation of the distribution network throughout the year (8760 hours) can be simulated, which is divided into 48 time windows.

[0094] S2. For the current time window, first determine whether there is a risk of exceeding the limit in the original topology of the power grid in the current time window. If there is a risk of exceeding the limit in the current time window, jump to S3. Otherwise, obtain the operation simulation results of the current time window according to the original topology of the power grid and jump to S4.

[0095] Specifically, if there is an operational risk such as load reduction, branch power flow exceeding the limit, or node voltage exceeding the limit, it indicates that there is an operational risk of exceeding the limit in the current time window, and the network reconfiguration mechanism needs to be triggered.

[0096] S3. Execute the distribution network reconfiguration method for time-series simulation of infeasible scenarios, realize network reconfiguration, and output the simulation results of the reconfigured distribution network topology under the current time window.

[0097] Specifically, the power grid reconfiguration method for time-series operation simulation of infeasible scenarios, as described in Example 1, is used to realize power grid reconfiguration and output the operation simulation results of the reconfigured power grid topology under the current time window. For details, please refer to the above description, which will not be repeated here.

[0098] S4. Determine if this is the last time window. If not, proceed to the next time window as the current time window and jump to S2. Otherwise, end the simulation and summarize the simulation results for the entire time period.

[0099] Specifically, when entering the next time window, the original power grid topology is used again for verification, and steps S2 to S3 are executed to complete the simulation of the next time window. This achieves a year-round closed-loop operation simulation with the original topology running first, automatic reconfiguration triggered by infeasible scenarios, and restoration of the original topology after risk mitigation.

[0100] Through the above methods, the present invention can extend the network reconstruction scheme obtained within a single window to the long-term time-series operation simulation process throughout the year. While ensuring the continuous transmission of operation results between windows, it generates a year-round rolling window reconstruction action plan that can be used for scheduling execution and operation analysis, thereby avoiding the direct misjudgment of over-limit windows or load-cutting windows that can be recovered through network reconstruction as infeasible scenarios.

[0101] In one embodiment, in step S2, the original topology of the power grid may be determined to have an over-limit risk in the current time window by solving the soft constraint risk diagnosis model. If the soft constraint risk diagnosis model has no solution in the current time window or any of the solved load shedding penalty term, branch power flow over-limit penalty term, and node voltage over-limit penalty term is greater than zero, it indicates that there is an over-limit risk in the current time window, and the process jumps to S3. Otherwise, the solved operating parameters are used as the operating simulation result of the current time window and the process jumps to S4.

[0102] The diagnostic constraints of the soft constraint risk diagnosis model include node power balance constraints, voltage drop relationship constraints at the beginning and end of the line, soft constraints for node voltage exceeding limits, soft constraints for line active power flow exceeding limits, and soft constraints for load reduction.

[0103] It should be noted that the soft-constraint risk diagnosis model and the distribution network reconfiguration optimization model are two different models. The soft-constraint risk diagnosis model does not need to introduce variables that characterize network reconfiguration, so the model form will be different.

[0104] Constraint 2_1: Node power balance constraint.

[0105] Node power balance constraints are used to ensure that the injected power at each node in the power grid is equal to the algebraic sum of the power flowing out of / into the branches of that node, in order to satisfy the energy conservation requirement of Kirchhoff's Current Law. Specifically: Node power balance constraints include node active power balance constraints and node reactive power balance constraints. The net injected active power of the node is: ... (69); The net injected reactive power of the node is: ... (70); The active power balance constraint at the node is: ……….(71); The reactive power balance constraint at the node is: ………..(72); In the formula, it is stipulated that if This indicates that node i injects active power into the network. This indicates that node i absorbs active power from the network; if This indicates that node i is injecting reactive power into the network. This indicates that node i absorbs reactive power from the network. and Let these represent the first and last nodes of branch k, respectively. , Let represent the net injected active power and net injected reactive power at node i, respectively. , Let K be the active power and reactive power flowing out of branch k. This represents the square of the current in branch k; Let K be the resistance of branch k. For the reactance of branch k, , Let these represent the active and reactive power directly injected into node i from the main network, respectively. For non-source nodes, take... , . and These represent the available active power output of photovoltaic and wind power at node i during time period t, respectively. and These represent the curtailed solar power and wind power at node i during time period t, respectively. This represents the active power load demand at node i during time period t; This represents the reactive load demand of node i during time period t. , This represents the active power load shedding demand and reactive power load shedding demand at node i during time period t.

[0106] Constraint 2_2: Constraint on the voltage drop relationship between the beginning and end of the line.

[0107] Voltage drop relationship constraint at the beginning and end of a line: This means that the voltage phasor difference between the two ends of each branch is equal to the complex impedance of that branch multiplied by the complex current flowing through the branch. It is a constraint condition obtained by coupling voltage drop with the branch transmission power / current according to Ohm's law based on the branch power flow model constructed by Kirchhoff's laws.

[0108] First, construct a branch power flow model based on Kirchhoff's laws. Then, combine this model with Ohm's law to derive the Ohm's law constraints as shown in the following equation. For physical circuits, we have: ... (73); in, and These are represented as the squares of the node voltage values.

[0109] For a closed virtual branch of a switch in the original topology, we have: (74).

[0110] Constraint 2_3: Second-order cone power flow constraint for the line.

[0111] It can be expressed as follows: ... (75); Equation (75) shows the relationship between node voltage, branch current and branch power.

[0112] Constraint 2_4: Soft constraint for node voltage exceeding limits.

[0113] Node voltage over-limit constraints refer to the requirement that the operating voltage amplitude of each node in a distribution network must be limited within its specified lower and upper safety thresholds to ensure system power quality and safe equipment operation. Specifically, this can be expressed as: ... (76); in, Represents a node The minimum value taken by the square of the voltage. For nodes The maximum value taken by the square of the voltage. Set the variable to This can reduce the dimensionality of variables, thereby simplifying the calculation process of the model.

[0114] The node voltage over-limit soft constraint introduces voltage relaxation variables on top of the original node voltage over-limit constraint. ... (77); The node voltage over-limit soft constraint can be expressed as: ... (78); In the formula, This represents the low-voltage over-limit relaxation amount of node j in time period t. This represents the overvoltage relaxation amount at node j during time period t.

[0115] Constraint 2_5: Soft constraint for line active power flow exceeding the limit.

[0116] Active power flow overload constraint refers to the requirement that the active power transmitted by each line in a power system must be limited to its maximum permissible range to prevent line overload from causing tripping or system cascading failures. Specifically, it can be expressed as: ... (79); In the formula, This represents the maximum active power transmission value allowed for line k.

[0117] The active power flow exceeding the limit soft constraint is to introduce an active power flow exceeding the limit relaxation variable on the basis of the original active power flow constraint: ... (80); The active power flow exceeding the limit soft constraint of the line can be expressed as: ... (81); In the formula, This represents the active power flow over-limit relaxation variable of line k in time period t, which reflects the degree of active power flow over-limit.

[0118] Constraint 2_6: Load reduction soft constraint.

[0119] Load shedding constraints prevent load shedding during normal operation. Specifically, this can be expressed as: ... (82); In the formula This indicates the amount of active power shearing.

[0120] The load reduction soft constraint introduces a load demand slack variable on top of the original load reduction constraint: The soft constraint for load reduction can be expressed as: ... (83); In the formula, This represents the load reduction amount during time period t. The constraint represents the load demand slack variable for time period t. Therefore, the constraint means that the load reduction amount cannot be negative, nor can it exceed the total load demand at that time.

[0121] The goal of the soft-constraint risk diagnosis model is to minimize the diagnostic objective function, which includes the main grid power purchase cost, line loss cost, renewable energy curtailment penalty and load shedding penalty, branch power flow exceeding limit penalty and node voltage exceeding limit penalty. The load shedding penalty is the spatiotemporal accumulation of load demand slack variables, the branch power flow exceeding limit penalty is the spatiotemporal accumulation of active power flow exceeding limit slack variables, and the node voltage exceeding limit penalty is the spatiotemporal accumulation of voltage slack variables.

[0122] Among them, the penalty for exceeding the limit of branch line flow Voltage over-limit penalty Load shedding penalty The calculation formula is as follows.

[0123] Branch road flow exceeding the limit penalty : ... (84); In the formula, The penalty coefficient for branch road exceeding the limit. This is the baseline capacity.

[0124] Voltage over-limit penalty : ... (85); In the formula, This is the penalty coefficient for voltage exceeding the limit.

[0125] load shedding penalty : ... (86); In the formula, This is the load shedding penalty coefficient.

[0126] In summary, this invention embeds a network reconfiguration mechanism into the long-term time-series simulation process of distribution networks, constructing a complete simulation framework that includes soft constraint risk diagnosis, candidate reconfiguration topology generation, and a full-time rolling window reconfiguration action plan. By identifying risk scenarios and automatically triggering network reconfiguration when the original topology is infeasible, this invention effectively avoids the misjudgment by traditional fixed topology simulation methods of abnormal scenarios that can be recovered through reconfiguration as system infeasibility. It focuses on refined time-series production simulation of distribution networks, with a particular emphasis on identifying infeasible operating conditions, and then uses topology optimization to better simulate the actual situation of the distribution network to evaluate its operation. This provides a more accurate, continuous, and logically consistent data foundation for subsequent assessments of renewable energy absorption capacity, configuration of distribution automation switches, and expansion and upgrading of distribution networks, demonstrating significant engineering application value.

[0127] Example 3 The present invention also relates to an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0128] The electronic device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The memory can be used to store computer programs and / or modules. The processor implements various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory, and by accessing data stored in the memory.

[0129] Example 4 The present invention also relates to a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0130] Specifically, the memory may include high-speed random access memory, as well as non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0131] The technical features of the embodiments described above can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. It should be noted that the terms "in one embodiment," "for example," and "again" in this invention are intended to illustrate the invention and are not intended to limit the invention.

[0132] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A distribution network reconfiguration method for time-series operational simulation infeasibility scenarios, characterized in that, include: Step 1: Construct a distribution network reconfiguration optimization model based on the branch and switch distribution in the distribution network. The distribution network is a radial distribution network, and the branches include physical lines and switch branches. The switch branches are equivalent to zero-impedance virtual branches connecting the two ends of the switch. The distribution network reconfiguration optimization model defines Let k be the state variable of branch k, with values ​​of 0 and 1 representing the branch being open and closed, respectively; defined and Let k be the current direction variable in branch k, and its superscript is... Indicates the direction from the first node to the last node of a branch, indicated by the superscript. This indicates the direction from the end node of the branch to the beginning node. The current direction variable takes values ​​of 0 and 1, representing no current flowing in the corresponding direction and current flowing in the corresponding direction, respectively. The switching action variable for the zero-impedance virtual branch k has a value of 0 and 1, which represent the switch not being switched and the switch being switched, respectively. The reconfiguration objective of the power distribution network reconfiguration optimization model is to minimize the reconfiguration operation cost; The reconfiguration constraints of the distribution network reconfiguration optimization model include: branch status and power supply direction constraints, switch operation constraints, multi-source radiation constraints, branch power flow constraints, voltage drop relationship between the beginning and end of the branch, node power balance constraints, node voltage upper and lower limits constraints, and load shedding constraints. The branch status and power supply direction constraints include: ; The switching action constraints include: , ; To reconstruct the original state of the preceding branch k; The multi-source radiation constraint introduces two virtual connected flow variables in different directions for branch k. and ,Include: ； ； ； ； ； In the formula, i is the node index and k is the branch index. For the set of distribution network nodes, For the source node set, The set of non-source nodes remaining after removing the source node set from the distribution network node set. Let M be the set of distribution network branches, where M is a preset positive number. Let them represent the first and last nodes of branch k, respectively. Let node i be the first node of branch k. Let node i be the terminal node of branch k; Step 2: Solve the power distribution network reconfiguration optimization model to obtain the power distribution network reconfiguration topology and its corresponding operation simulation results.

2. The distribution network reconfiguration method for time-series operation simulation infeasibility scenarios as described in claim 1, characterized in that, The branch power flow constraints include branch power flow start and stop constraints and physical line second-order cone power flow constraints. The branch power flow start / stop constraints apply to each branch. Constraints include: ； ； ； ； ； ； In the formula, Represents the set of branches. , Let represent the active power flow of branch k in the ft and tf directions, respectively. , Let represent the reactive power flow of branch k in the ft and tf directions, respectively. , Let represent the squares of the currents in branch k in the ft and tf directions, respectively. This represents the upper limit of the capacity of branch k. This indicates the upper limit of the current in branch k; The second-order cone power flow constraint of the physical line only applies to the physical line. Constraints include: ； ； In the formula, This represents the set of physical lines, and t represents the time period index; , Let represent the active power flow of branch k in the ft and tf directions, respectively. , Let represent the reactive power flow of branch k in the ft and tf directions, respectively. , Let represent the squares of the current in branch k in the ft and tf directions, respectively; and Let represent the square of the voltage at the first node and the square of the voltage at the last node of branch k, respectively; This represents the L2 norm.

3. The distribution network reconfiguration method for time-series operation simulation infeasibility scenarios as described in claim 1, characterized in that, The voltage drop relationship constraint between the beginning and end of the branch includes the voltage drop relationship constraint between the beginning and end of the physical line and the voltage drop relationship constraint between the beginning and end of the switch branch. The voltage drop relationship constraint between the beginning and end of the physical circuit is for the physical circuit. Constraints include: ； ； The voltage drop relationship constraint between the first and last ends of the switching branch is for the switching branch. Constraints include: ； In the formula, This represents the set of physical lines, and t represents the time period index. Represents the set of switch branches. , Let be the squares of the voltages at the first and last nodes of branch k, respectively. , Let represent the active power flow of branch k in the ft and tf directions, respectively. , Let represent the reactive power flow of branch k in the ft and tf directions, respectively. , Let represent the squares of the currents in branch k in the ft and tf directions, respectively. Let K be the resistance of branch k. Let K be the reactance of branch k.

4. The distribution network reconfiguration method for time-series operation simulation infeasibility scenarios as described in claim 1, characterized in that, The node power balance constraints include node active power balance constraints and node reactive power balance constraints. The active power balance constraints at the nodes include: ； ； ； ； The node reactive power balance constraint includes: ； ； ； ； In the formula, Let t represent the set of branches, and t represent the time period index. , Let these represent the outflow of active power and the outflow of reactive power at node i, respectively. , Let and represent the inflow active power and inflow active power at node i, respectively. , Let represent the net injected active power and net injected reactive power at node i, respectively. , Let represent the active power flow of branch k in the ft and tf directions, respectively. , Let represent the reactive power flow of branch k in the ft and tf directions, respectively. , Let these represent the active load demand and reactive load demand of node i, respectively. , Let represent the squares of the currents in branch k in the ft and tf directions, respectively. Let K be the resistance of branch k. For the reactance of branch k, , These represent the active power and reactive power directly injected into node i from the main network, respectively. and Let i represent the available active power output of photovoltaic and wind power, respectively. and These represent the curtailed solar power and curtailed wind power at node i, respectively.

5. The distribution network reconfiguration method for time-series operation simulation infeasibility scenarios as described in claim 1, characterized in that, The load shedding constraints include load shedding constraints under strict network reconstruction mode and load shedding constraints under emergency network reconstruction mode: In the strict mode of network reconstruction, the load shedding constraint limits the load shedding amount to zero, and the upper limit of load shedding is set to zero. The load shedding constraint in the emergency mode of network reconstruction relaxes the load shedding constraint in the strict mode of network reconstruction, allowing load shedding within a preset range, with the upper limit of load shedding set to be greater than zero. The power distribution network reconfiguration method first performs network reconfiguration in strict network reconfiguration mode. If the solution cannot be found in this mode, it switches to emergency network reconfiguration mode for network reconfiguration.

6. The distribution network reconfiguration method for time-series operation simulation infeasibility scenarios as described in claims 1 to 5, characterized in that, The reconfiguration operation costs include the main grid power purchase cost, line loss cost, switching operation cost, renewable energy curtailment penalty, and load shedding penalty.

7. A long-term time-series operation simulation method for distribution networks with embedded network reconfiguration mechanisms, characterized in that, include: S1. Divide the entire time period to be simulated into sub-time periods, each sub-time period being a time window. Take the first time window as the current time window and jump to S2. S2. For the current time window, first determine whether there is a risk of exceeding the limit in the original topology of the power grid in the current time window. If there is a risk of exceeding the limit in the current time window, jump to S3. Otherwise, obtain the simulation results of the current time window according to the original topology of the power grid and jump to S4. S3. Execute the distribution network reconfiguration method for time-series operation simulation of infeasible scenarios as described in any one of claims 1 to 6, realize network reconfiguration and output the operation simulation results of the reconfigured distribution network topology under the current time window; S4. Determine if this is the last time window. If not, proceed to the next time window as the current time window and jump to S2. Otherwise, end the simulation and summarize the simulation results for the entire time period.

8. The long-term time-series operation simulation method for distribution networks with embedded network reconfiguration mechanism as described in claim 7, characterized in that, The soft-constraint risk diagnosis model is used to determine whether there is a risk of exceeding the limit in the original topology of the power grid in the current time window. If the soft-constraint risk diagnosis model has no solution in the current time window or any of the load shedding penalty term, branch power flow limit exceeding penalty term, and node voltage limit exceeding penalty term obtained by the solution are greater than zero, it indicates that there is a risk of exceeding the limit in the current time window, and the process jumps to S3. Otherwise, the solved operating parameters are used as the operating simulation result of the current time window and the process jumps to S4. The diagnostic constraints of the soft constraint risk diagnosis model include node power balance constraints, voltage drop relationship constraints at the beginning and end of the line, second-order cone power flow constraints of the line, soft constraints for node voltage exceeding limits, soft constraints for active power flow exceeding limits of the line, and soft constraints for load reduction. The soft constraints for node voltage exceeding limits are achieved by introducing non-negative voltage relaxation variables based on the node voltage range that the system can withstand. The line active power flow exceeding limit soft constraint is to introduce a non-negative active power flow exceeding limit relaxation variable based on the line active power flow range that the system can bear, so as to realize the line active power flow exceeding limit relaxation treatment; the load reduction soft constraint is to introduce a non-negative load demand relaxation variable, so as to realize the load shedding exceeding limit relaxation treatment. The objective of the soft-constraint risk diagnosis model is to minimize the diagnostic objective function, which includes the main grid power purchase cost, line loss cost, renewable energy curtailment penalty and load shedding penalty, branch power flow exceeding limit penalty and node voltage exceeding limit penalty. The load shedding penalty is the spatiotemporal accumulation of load demand slack variables, the branch power flow exceeding limit penalty is the spatiotemporal accumulation of active power flow exceeding limit slack variables, and the node voltage exceeding limit penalty is the spatiotemporal accumulation of voltage slack variables.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the distribution network reconfiguration method for time-series operation simulation infeasible scenarios as described in any one of claims 1 to 6, or the steps of the distribution network long-term time-series operation simulation method with embedded network reconfiguration mechanism as described in claim 7 or 8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the distribution network reconfiguration method for time-series operation simulation infeasible scenarios as described in any one of claims 1 to 6, or the steps of the distribution network long-term time-series operation simulation method with embedded network reconfiguration mechanism as described in claim 7 or 8.

Images