An all-time domain recovery decision method for ac-dc system after fault

Abstract

The present application relates to the technical field of power system safety control and restoration optimization, in particular to a kind of AC-DC system post-fault full-time domain recovery decision method, comprising: obtaining system topology, parameter, initial state and fault set, and processing available topology;According to available generator and topology, the initial live / loss-of-power bus is distinguished, and the power of the equipment connected to the closure loss-of-power bus is closed.Powerful node network is constructed with fixed online line aggregation bus as super node and recoverable line as edge.The topology, initial state, live / loss-of-power bus are used as boundary conditions, and a single-scenario full-time domain optimization model is built by combining parameters and super node network, to obtain recovery path and scheme;Only AC constraints are used when there is no DC power grid, and DC constraints are introduced when there is DC.The present application improves the global coordination and physical feasibility of recovery decision after AC-DC system fault.

Description

Technical Field

[0001] This invention relates to the field of power system safety control and recovery optimization technology, and in particular to a full-time-domain recovery decision-making method for AC / DC system after a fault. Background Technology

[0002] With the expansion of the power grid and the increasing proportion of inter-regional power transmission, DC power transmission, and power electronic equipment, the recovery control after AC / DC system faults faces stronger coupling and complexity. Generators, loads, lines, and converters may be offline or restricted simultaneously, leading to drastic changes in topology and power distribution. Existing recovery methods that rely on manual experience or offline contingency plans cannot fully reflect actual topology changes and power supply support differences. In particular, there is a strong coupling relationship between converter recovery, DC power adjustment, and AC load recovery, making it difficult to ensure global coordination. Although some optimization model-based methods can perform power restoration or dispatch reconfiguration, their objectives are mostly focused on maximizing the restored load or minimizing the lost power. They do not adequately consider the equipment recovery action logic, bus energized connectivity, frequency response characteristics, and safety and stability control factors such as low-frequency load shedding and high-frequency generator tripping. They often simply assume that the frequency of the entire network is consistent or ignore the correlation between the power lost bus and equipment power, making the recovery strategy physically infeasible. In addition, for scenarios where the initial fault is known and no new faults are added in the recovery time domain, rolling optimization methods need to repeatedly update the state, construct multiple scenarios, and introduce constraints on the first step, increasing unnecessary model complexity and computational burden. In summary, existing technologies are unable to provide a unified description and coordinated optimization of the state switching of generators, loads, lines, and converters, the energized state of buses, AC / DC power coupling, and frequency safety control in AC / DC systems within the complete recovery time domain. It is difficult to generate a full-time-domain recovery decision scheme that is both globally coordinated and physically feasible. Summary of the Invention

[0003] This invention provides a full-time-domain recovery decision-making method for AC / DC systems after a fault, which addresses the problems in existing AC / DC system fault recovery processes, such as poor coordination of manual contingency plans, optimization models neglecting safety, stability, and physical feasibility, and redundant calculations in rolling optimization, which prevent the generation of optimal recovery decisions that take into account load recovery, frequency safety, and operational costs across the entire time domain.

[0004] The objective of this invention can be achieved through the following technical solutions:

[0005] The first aspect of this invention is to provide a full-time-domain recovery decision-making method for AC / DC system after a fault, comprising: Obtain the network topology, equipment parameters, initial operating status, and initial fault set of the AC / DC system; process the network topology using the initial fault set to obtain a usable topology map; Determine available generators; based on the available topology and available generators, identify and obtain the initial set of energized buses and the initial set of de-energized buses; then perform state closure processing on the equipment connected to the de-energized buses, setting the power variables of the equipment to zero or constraining them to the energized state of the buses; Obtain the fixed online AC lines of the AC power grid in the AC / DC system within the recovery time domain, establish connectivity through the fixed online AC lines, aggregate the buses in the AC power grid into AC supernodes; establish a mapping from the buses to the supernodes in the AC power grid, and use the recoverable AC lines in the AC power grid as recoverable edges between supernodes to construct an AC supernode network; The available topology, initial operating state, initial set of energized buses, and initial set of de-energized buses are used as initial boundary conditions. A single-scenario full-time domain recovery optimization model is constructed using the initial boundary conditions, equipment parameters, and AC supernode network. By solving the single-scenario full-time domain recovery optimization model, the optimal recovery path and operation scheme within the complete time domain are obtained; When solving the single-scenario full-time domain recovery optimization model, if there is no DC grid in the AC / DC system, the solution is directly obtained by constructing AC constraints; if there is a DC grid in the AC / DC system, both AC constraints and DC constraints are constructed and solved together.

[0006] Furthermore, the process of obtaining the available topology graph is as follows: The initial fault set is removed from the network topology of the AC / DC system to form the initial system topology after the fault, which is denoted as the available topology graph.

[0007] Furthermore, the process of obtaining the initial set of energized buses and the initial set of de-energized buses is as follows: Using the bus where the available generator is located as the power source, a connectivity search is performed in the available topology map to determine whether each bus is connected to any available generator, and the connectivity determination results of all buses are obtained. Based on the connectivity determination results, the buses connected to available generators are assigned to the initial set of energized buses, and the buses that cannot be connected are assigned to the initial set of de-energized buses. The specific process of busbar division is as follows: In the connectivity determination result, it is determined whether each bus has a connected path with at least one available generator in the available topology graph. If so, the bus is in a energized state; otherwise, the bus is in a de-energized state. All energized buses are assigned to the initial energized bus set, and all de-energized buses are assigned to the initial de-energized bus set.

[0008] Furthermore, the process of acquiring the communication supernode is as follows: Using fixed online AC lines as connecting edges, perform connectivity analysis on all AC buses, divide interconnected buses into the same connected component, and define each connected component as an AC supernode.

[0009] Furthermore, the AC constraints include: equipment status and recovery action logic constraints, bus energized status and equipment power constraints, AC active power balance constraints, AC line power flow constraints, frequency consistency and generation support subgrid constraints, and low-frequency load shedding and high-frequency generator tripping constraints.

[0010] Furthermore, the DC constraint is a coupling constraint between the DC network and the converter.

[0011] Furthermore, the single-scenario full-time domain recovery optimization model includes a comprehensive objective function; the comprehensive objective function consists of load loss cost, frequency deviation cost, recovery action cost, low-frequency load shedding cost, and high-frequency tripping cost; Specifically, the cost of load loss is expressed as follows:

[0012] In the formula, Indicates the time interval for the decision to resume; Indicates load During the period The actual power value, Indicates load The rated power requirement; T represents the load set; T represents the time period set. The frequency deviation cost is specifically expressed as follows:

[0013] In the formula, Supernode During the period The frequency variable is denoted by R, which represents the set of all supernodes. It is the absolute value symbol; The cost of the recovery action is specifically expressed as follows:

[0014] In the formula, Indicates generator During the period The recovery action variable; Indicates load During the period The recovery action variable; Indicates AC line During the period The recovery action variable; This indicates that converter c is in time period The recovery action variable; Represents a set of generators. Represents the load set. Indicates a set of communication lines. Represents a set of converters; The costs of low-frequency load shedding and high-frequency tripping are specifically expressed as follows:

[0015] In the formula, Supernode During the period Low-frequency load reduction power, Supernode During the period High-frequency cutting power; in, Indicates the cost of load loss. Indicates the cost of frequency deviation. Indicates the cost of resuming the action. This indicates the cost of low-frequency load reduction. This indicates the cost of high-frequency switching.

[0016] Furthermore, the comprehensive objective function is specifically expressed by the formula:

[0017] In the formula, The preset weight for load loss cost, The preset weights for the frequency deviation cost, To restore the preset weights of the action cost, Preset weights for the cost of low-frequency load reduction The preset weights for the high-frequency switching cost; J represents the comprehensive objective function value. This represents the function that takes the minimum value.

[0018] A second aspect of the present invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the aforementioned full-time-domain recovery decision method after an AC / DC system fault.

[0019] A third aspect of the present invention is to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned full-time-domain recovery decision method for AC / DC system faults.

[0020] Compared with existing technologies, the beneficial effects of this invention are: acquiring the network topology, equipment parameters, initial operating state, and initial fault set of the AC / DC system; processing the network topology using the initial fault set to obtain a usable topology graph; providing accurate and complete initial boundary conditions and topological foundation for subsequent full-time domain recovery optimization; determining available generators; identifying and acquiring the initial energized bus set and the initial de-energized bus set based on the usable topology graph and available generators; then performing state closure processing on the equipment connected to the de-energized bus, setting the power variables of the equipment to zero or constraining them by the energized state of the bus; avoiding unreasonable power injection or consumption on the de-energized bus, ensuring the physical feasibility of the recovery scheme; acquiring the fixed online AC lines of the AC grid in the AC / DC system within the recovery time domain, establishing connectivity through the fixed online AC lines, and aggregating the buses in the AC grid into AC supernodes; establishing a mapping from buses to supernodes in the AC grid, and using recoverable AC lines in the AC grid as recoverable edges between supernodes to construct an AC supernode network; reducing the scale of model variables while retaining the ability to characterize the frequency differences and power generation support range of different subnets; The available topology, initial operating state, initial set of energized buses, and initial set of de-energized buses are used as initial boundary conditions. A single-scenario full-time-domain recovery optimization model is constructed using these initial boundary conditions, equipment parameters, and an AC supernode network. This model achieves unified and coordinated optimization of load recovery, frequency security, and equipment operation costs across the entire time domain. By solving the single-scenario full-time-domain recovery optimization model, the optimal recovery path and operation scheme within the complete time domain are obtained. A complete equipment recovery sequence and system operation adjustment scheme are output at once, improving the systematic nature of the decision-making. When solving the single-scenario full-time-domain recovery optimization model, if there is no DC grid in the AC / DC system, AC constraints are directly constructed for solving; if there is a DC grid in the AC / DC system, both AC and DC constraints are constructed simultaneously for solving. This makes the method applicable to both pure AC systems and AC / DC hybrid systems, exhibiting good versatility. It solves the problems of poor coordination of manual contingency plans, neglect of safety, stability, and physical feasibility in optimization models, and redundant calculations in rolling optimization during the recovery process of existing AC / DC systems after a fault, which prevent the unified generation of optimal recovery decisions that balance load recovery, frequency security, and operation costs across the entire time domain. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1This invention provides a flowchart illustrating the steps of a full-time-domain recovery decision-making method for AC / DC system after a fault. Figure 2 Here is a diagram of the IEEE 39-node computational example structure; Figure 3 A graph showing the changes in load power supply rate and load loss rate; Figure 4 This is a graph showing the change in the number of power-off buses; Figure 5 This is a graph showing the change in the absolute value of the maximum frequency difference. Detailed Implementation

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0024] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0025] To address the problems existing in the background technology, a full-time domain recovery decision-making method for AC / DC system faults has been researched and designed, which has important practical significance.

[0026] like Figure 1 As shown, the first aspect of the present invention is to provide a full-time domain recovery decision-making method after an AC / DC system fault, comprising the following steps: Step S1: Obtain the network topology, equipment parameters, initial operating status, and initial fault set of the AC / DC system; process the network topology using the initial fault set to obtain an available topology map.

[0027] It should be noted that, because existing methods fail to adequately address the physical fact that the initial state of an AC / DC system has undergone deterministic changes (such as equipment offline, topology changes, and power distribution shifts) after a fault, and accurate initial data is the basis for all subsequent optimizations and judgments, it is necessary to first obtain the system's basic data and initial fault information in order to provide the full-time-domain optimization model with realistic and reliable boundary conditions and initial states, and to avoid the recovery strategy deviating from reality due to incomplete or inaccurate input.

[0028] Specifically, the network topology of the AC / DC system is obtained, including the connection relationships of AC buses, AC lines, DC nodes, DC lines and converters; Collect parameters of various equipment in the AC / DC system, including generator active power limits, load demand, line reactance and capacity, converter power limits and efficiency, DC network parameters and frequency response parameters; Read the initial operating status of the system, including initial power generation, initial load power supply, initial line status, initial converter power, and initial frequency status; Determine the initial set of faults, identify the generators, loads, lines and converters involved in the faults, and remove them from the set of normal equipment.

[0029] The network topology of the AC / DC system is processed by initial fault set to obtain a usable topology map; the specific processing steps are as follows: The initial set of faults is removed from the network topology (normal state) of the AC / DC system to form the initial system topology after the fault, which is denoted as the available topology diagram (including the energized state of the bus, the online state of the equipment, etc.). It should be noted that the removal is to isolate the faulty line topology (remove the edge), set the online state of the faulty generator, load, and converter to zero and block the power output / input, while the bus nodes connected to them are still retained in the network structure.

[0030] Step S2: Determine available generators; based on the available topology and available generators, identify and obtain the initial set of energized buses and the initial set of de-energized buses; then perform state closure processing on the equipment connected to the de-energized buses, and set the power variables of the equipment to zero or constrain them to the energized state of the buses.

[0031] It should be noted that after the initial fault occurs, it is not intuitive to know which buses in the system can still be powered by available generators through the remaining topology. The energized state of the bus cannot be directly determined by the fault flag of the equipment itself. Therefore, in order to provide accurate initial energized and de-energized boundaries for the subsequent full-time optimization model, and to ensure that all equipment (generators, loads, converters) connected to the de-energized bus will not generate unreasonable power injection or consumption, thereby avoiding physically infeasible recovery schemes in the optimization model, it is necessary to determine the initial energized bus set through connectivity discrimination based on the remaining topology after removing faulty equipment, and to perform closure processing to forcibly set the power variables of the equipment on the de-energized bus to zero.

[0032] Specifically, the set of available generators that have not experienced any faults and are capable of providing power supply is determined. Taking the bus where the available generator is located as the power source, a connectivity search is performed in the available topology map to determine whether each bus is connected to any available generator, and the connectivity determination results of all buses are obtained. Based on the connectivity determination results, buses connected to available generators are assigned to the initial set of energized buses, and buses that cannot be connected are assigned to the initial set of de-energized buses. The specific process of busbar division is as follows: In the connectivity determination result, it is determined whether each bus has a connected path with at least one available generator in the available topology graph. If so, the bus is in a energized state; otherwise, the bus is in a de-energized state. All energized buses are assigned to the initial energized bus set, and all de-energized buses are assigned to the initial de-energized bus set.

[0033] For all equipment (generators, loads, converters) connected to the initial power failure bus, their corresponding power variables are forcibly set to zero or blocked.

[0034] Optionally, the connectivity determination results of the topology graph are checked for consistency with the initial power flow calculation results. The specific process is as follows: when the connectivity determination results are inconsistent with the initial power flow calculation results, it indicates that there is a conflict between the initial fault set, topology state, or power flow verification results. Recovery optimization should be stopped and a prompt should be made to recheck the initial data. When the connectivity determination results are consistent with the initial power flow calculation results, subsequent full-time domain recovery optimization modeling is then performed.

[0035] The initial power flow calculation results were obtained using power flow calculations known in the field of power systems.

[0036] Step S3: Obtain the fixed online AC lines of the AC power grid in the AC / DC system within the recovery time domain, establish connectivity through the fixed online AC lines, aggregate the buses in the AC power grid into AC supernodes; establish a mapping from the buses to the supernodes in the AC power grid, and use the recoverable AC lines in the AC power grid as recoverable edges between supernodes to construct an AC supernode network.

[0037] It should be noted that after an AC / DC system failure, a large number of fixed online AC lines (not involved in recovery decisions) tightly electricalally connect multiple buses. These buses must physically have the same frequency. Setting separate frequency and connectivity variables for each bus would lead to a sharp increase in model size. At the same time, existing methods often ignore the frequency differences between different AC subnets and simply assume that the frequency of the entire network is consistent, which cannot accurately characterize the independent power generation support capacity of each subnet. Therefore, in order to reduce the size of the optimization model variables and improve the solution efficiency while still accurately expressing the independent frequency dynamic characteristics and power generation support range of each AC subnet, so as to provide an accurate physical basis for subsequent frequency safety constraints and load recovery feasibility judgment, the buses are aggregated into AC supernodes based on the connectivity of the fixed online lines, and recoverable lines are regarded as recoverable edges between supernodes.

[0038] Specifically, all fixed-line (i.e., not involved in recovery decisions and always in an operational state) AC lines in the available topology graph are obtained. Using the fixed-line AC lines as connecting edges, connectivity analysis is performed on all AC buses, and interconnected buses are divided into the same connected component; each connected component is defined as an AC supernode, forming a set of supernodes. Establish a mapping relationship between the bus and the supernode, and record the supernode number to which each bus belongs; for a recoverable communication line (i.e. a line that participates in the recovery decision), identify the supernodes to which the bus at both ends of the (recoverable communication line) belongs, and regard the line as a recoverable edge in the supernode network, thereby constructing the communication supernode network.

[0039] Step S4: Use the available topology, initial operating state, initial set of energized buses and initial set of de-energized buses as initial boundary conditions; construct a single-scenario full-time domain recovery optimization model through the initial boundary conditions, equipment parameters and AC supernode network.

[0040] It should be noted that, because the post-fault recovery process requires unified decision-making on the recovery actions, power adjustments, and frequency responses of generators, loads, lines, and converters within the complete time domain, existing methods lack the ability to collaboratively model equipment state logic, bus energization conditions, AC / DC coupling, and safety and stability control (such as low-frequency load shedding and high-frequency tripping). This results in recovery schemes that are either physically infeasible or neglect frequency safety. Therefore, in order to solve the globally optimal recovery path that takes into account topology reconfiguration, power balance, frequency safety, and operational costs in one go, given the initial fault and operating conditions, it is necessary to establish a single-scenario, full-time-domain mixed integer linear programming model that includes equipment state transitions, power constraints, network power flow, frequency consistency, and safety and stability control actions.

[0041] Specifically, the available topology, initial operating state, initial set of energized buses and initial set of de-energized buses are used as initial boundary conditions; a single-scenario full-time domain recovery optimization model is constructed using the initial boundary conditions, equipment parameters and AC supernode network.

[0042] First, it is determined whether a DC power grid exists in the AC / DC system. If no DC power grid exists in the AC / DC system, the single-scenario full-time domain recovery optimization model only includes AC constraints; if a DC power grid exists in the AC / DC system, the single-scenario full-time domain recovery optimization model includes both AC and DC constraints.

[0043] The process of constructing communication constraints is as follows: AC constraints include: equipment status and recovery action logic constraints, bus energization status and equipment power constraints, AC active power balance constraints, AC line power flow constraints, frequency consistency and generation support subgrid constraints, and low-frequency load shedding and high-frequency generator tripping constraints.

[0044] The equipment status and recovery action logic constraints include: generator status transition constraints, load status transition constraints, AC line status transition constraints, converter status transition constraints, exogenous fault equipment mandatory constraints, recovery action quantity limit constraints, and binary variable constraints.

[0045] Among them, a generator is set Connect to AC bus The generator state transition constraints are specifically expressed by the following formula:

[0046] In the formula, Indicates generator During the period The online status variable (1 indicates online, 0 indicates offline); Indicates generator During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery is performed); Indicates generator During the period The high-frequency cutting machine action variable (1 indicates that it has been cut off, and 0 indicates that it has not been cut off); Indicates AC busbar During the period The charged state variables (1 represents charged, 0 represents de-energized); For generator index, For time period index, This is the index for the AC bus. It should be noted that... AC bus in It is a generator The busbar to which it is connected.

[0047] In the generator state transition constraints, the first equation indicates that the generator can only be online if it was already online in the previous time period or if it is currently performing a recovery action; the second equation indicates that the generator's online status depends on its connection to the energized AC bus; the third equation indicates that the generator cannot remain online after the high-frequency tripping action is triggered; the fourth equation is used to ensure that the generator's online status is consistent with the above logic under the conditions that the AC bus is energized, it was online in the previous time period, or it is currently recovering and the high-frequency tripping action has not been triggered; the fifth equation indicates that the recovery action can only be applied to generators that were offline in the previous time period; and the sixth equation indicates that the high-frequency tripping action can only be applied to generators that were online in the previous time period.

[0048] Among them, the load is set Connect to AC bus The load state transition constraint is specifically expressed by the formula:

[0049] In the formula, Indicates load During the period The online status variable (1 indicates online, 0 indicates offline); Indicates load During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery). It should be noted that the load... This refers to connecting to the AC bus. The load on.

[0050] Among them, the load state transfer constraint means that the load cannot remain online on the de-energized AC bus, nor can it perform a restoration action when the AC bus is not energized; after the AC bus is energized again, the load still needs to perform a restoration action to restore power supply.

[0051] Among them, an AC line is provided. Connecting supernodes and supernode The state transition constraints for AC lines are specifically expressed by the following formula:

[0052] In the formula, Indicates AC line During the period The online status variable (1 indicates online, 0 indicates offline); Indicates AC line During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery is performed); Supernode During the period The charged state variables (1 represents charged, 0 represents de-charged). Supernode During the period The charged state variables (1 represents charged, 0 represents de-energized).

[0053] Among them, the AC line state transition constraint means that line restoration needs to meet the condition that the two supernodes at both ends are energized; the line restoration action must start from the supernode that was already energized in the previous time period, and cannot energize an isolated area that has no power support out of thin air.

[0054] Among them, a converter is set Connect to AC bus The converter state transition constraints are specifically expressed by the following formula:

[0055] In the formula, Indicates converter c During the period The online status variable (1 indicates online, 0 indicates offline); Indicates converter c During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery is performed); Indicates AC busbar During the period The charged state variables (1 represents charged, 0 represents de-energized).

[0056] Among them, the converter state transition constraint means that the converter's online and recovery operations both depend on the energized state of its AC side bus; when the AC side bus is de-energized, the converter cannot maintain the power exchange state, nor can it perform recovery actions.

[0057] For any device Then it is in the time period The forced constraints on exogenous fault equipment are specifically expressed by the following formula:

[0058]

[0059] In the formula, Indicates equipment During the period The online status variable (1 indicates online, 0 indicates offline); Indicates equipment During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery).

[0060] Among them, the exogenous fault equipment mandatory constraint means that the faulty equipment cannot be restored to operation in the current time period. For the full-time domain recovery scenario described in this invention, if no new exogenous faults are added within the recovery time domain, the above-mentioned exogenous fault constraints are mainly reflected through the initial fault state and the initial online state.

[0061] The aforementioned state variables satisfy binary variable constraints, which are specifically expressed by the following formula:

[0062] In the formula, Indicates generator During the period The online status variable (1 indicates online, 0 indicates offline); Indicates load During the period The online status variable (1 indicates online, 0 indicates offline); Indicates AC line During the period The online status variable (1 indicates online, 0 indicates offline); This indicates that converter c is in time period The online status variable (1 indicates online, 0 indicates offline).

[0063] To reflect the limitations of on-site operational resources, an upper limit can be set for the number of recovery actions in each time period; time periods are set. The recovery action budget is This constructs a constraint on the number of recovery actions; the constraint on the number of recovery actions is specifically expressed by the formula:

[0064] In the formula, Indicates generator During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery is performed); Indicates load During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery is performed); Indicates AC line During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery is performed); Indicates converter c During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery is performed); Represents a set of generators. Represents the load set. Indicates a set of communication lines. Represents a set of converters; Indicates time period The maximum number of manual recovery actions that can be performed.

[0065] It should be noted that low-frequency load shedding and high-frequency machine switching are safety and stability control actions and are not included in the above-mentioned limit on the number of manual recovery actions.

[0066] Among them, the bus energization state and equipment power constraints include: load power constrained by the bus energization state, generator output constrained by the bus energization state, and converter power constrained by the bus energization state.

[0067] The load power constraint due to the energized state of the busbar can be expressed by the following formula:

[0068] In the formula, Indicates load During the period The actual power value, Indicates AC busbar During the period The charged state variables (1 represents charged, 0 represents de-charged). Indicates load During the period The online status variable (1 indicates online, 0 indicates offline). Indicates load The power rating requirement. If the equipment is connected to a de-energized AC bus, its corresponding power variable should be limited to zero.

[0069] Among them, for generators Connect to AC bus In this situation, the generator output is constrained by the energized state of the AC bus, which can be specifically expressed by the following formula:

[0070] In the formula, Indicates generator During the period Those who have made contributions Indicates generator The lower limit of merit and effort Indicates generator There is a limit to the amount of effort that can be put into practice. Indicates generator During the period The online status variable (1 indicates online, 0 indicates offline).

[0071] Among them, for converters Connect to AC bus In this scenario, the converter power is constrained by the AC bus energization state, specifically as follows:

[0072] In the formula, Indicates converter During the period AC side power, Indicates converter The upper limit of power, Indicates converter c During the period The online status variable (1 indicates online, 0 indicates offline).

[0073] The AC active power balance constraint (i.e., AC bus power balance) is specifically expressed by the following formula:

[0074] In the formula, Indicates generator During the period Those who have made contributions Indicates load During the period The actual power value, Indicates converter During the period AC side power, Indicates AC line During the period The meritorious trend; Indicates connection to AC bus A collection of generators, Indicates connection to AC bus The load set, Indicates connection to AC bus A collection of converters, Indicates AC busbar For the sending end of the line set, Indicates AC busbar This represents the set of lines at the receiving end. This constraint uniformly describes the power balance relationships at the AC bus level for generator injection, load consumption, converter power exchange, and line power flow.

[0075] Among them, for communication lines When the line is recoverable, let the AC busbars at both ends be respectively and The specific formula for AC line power flow constraints is as follows:

[0076] In the formula, Indicates AC line During the period The online status variable (1 indicates online, 0 indicates offline); Indicates AC line The reactance value; Indicates AC line During the period The active power flow, where M is a sufficiently large constant. Indicates AC busbar phase angle, Indicates AC busbar The phase angle. That is, under the DC power flow approximation, the line power flow and the phase angle difference satisfy the constraint of the relationship between line power flow and phase angle. Among them, when When this constraint degenerates into a conventional DC power flow equation; when When the line is disconnected, the constraint is greatly reduced. Relax.

[0077] Among them, frequency consistency and generation support subgrid constraints include: In a communication supernode network, a supernode is set up. During the period The frequency variable is If the line can be restored Connecting supernodes and Therefore, when the line is online, the frequencies of the two supernodes should be consistent:

[0078] when At that time, there were:

[0079] Furthermore, define binary variables. Supernode During the period Is it a network with generator support and electronic grid? If it exceeds the node... If an online generator exists within the memory, it has local power support, and its logical relationship can be expressed as follows:

[0080] In the formula, Supernode During the period The frequency variable, Indicates generator During the period The online status variable (1 indicates online, 0 indicates offline); Indicates AC line During the period The online status variable (1 indicates online, 0 indicates offline); Supernode The generator set within; the above logical relationship can be achieved through large Constraints or equivalent linear inequalities are transformed into mixed integer linear constraints.

[0081] This constraint, through supernode network connectivity constraints, further determines whether a supernode can be connected to a supernode with generator support via an online line. Only supernodes belonging to a subnet with generator support have physical meaning in terms of frequency variables and load recovery. This constraint can prevent unreasonable load supply or frequency response in islanded systems without power support.

[0082] Among them, the low-frequency load shedding and high-frequency tripping constraints are to ensure frequency safety during the recovery process. This invention incorporates low-frequency load shedding and high-frequency tripping as safety and stability control actions into the single-scenario full-time-domain recovery optimization model. The low-frequency load shedding and high-frequency tripping constraints include: low-frequency load shedding constraints and high-frequency tripping constraints.

[0083] Among them, the low-frequency load shedding constraints include low-frequency load shedding trigger logic constraints and low-frequency load shedding amount constraints.

[0084] Let the first The low-frequency load reduction threshold is The load reduction ratio is Define a binary variable. Supernode During the period Does it trigger the first Reduce load at low frequencies.

[0085] The low-frequency load shedding triggers the action of the current period according to the frequency of the previous period. The specific constraints of the low-frequency load shedding triggering logic are as follows:

[0086] When low-frequency load shedding is triggered, the supernode The load shedding amount within the range should meet the low-frequency load shedding constraint; the low-frequency load shedding constraint is specifically expressed as follows:

[0087] In the formula, This represents the total load of supernode r during time period t-1; Supernode During the period Low-frequency load reduction power, Supernode During the period Low-frequency load reduction power.

[0088] Among them, the high-frequency switching constraints include high-frequency switching trigger logic constraints and high-frequency switching quantity constraints.

[0089] Let the first The threshold for high-frequency switching is The cutting ratio is Define a binary variable. Supernode During the period Does it trigger the first High-frequency cutting machine.

[0090] The high-frequency switching mechanism triggers the current time period's action according to the frequency of the previous time period. The specific constraints of the high-frequency switching triggering logic are as follows:

[0091] When the high-frequency switching is triggered, the supernode The switching power within the range should meet the high-frequency switching quantity constraint; the high-frequency switching quantity constraint is specifically expressed as follows:

[0092] In the formula, Supernode During the period Total power generation; Supernode During the period High-frequency cutting power; Indicates generator During the period Those who have made meritorious contributions; This represents the set of generators within the supernode r.

[0093] Among them, when At that time, the frequency of the previous time period in the above triggering conditions is taken as the initial time. The frequency state.

[0094] When the AC / DC system includes a DC grid, DC constraints are enabled, where DC constraints are the coupling constraints between the DC network and the converter.

[0095] Specifically, the coupling constraints between the DC network and the converter include: After linearizing the DC node power equation, the equation becomes:

[0096] In the formula, Indicates converter During the period DC-side power, This represents the elements of the DC network admittance matrix (i.e., the mutual conductance or mutual susceptance between two DC buses). Indicates that DC bus n is in the time period voltage, This represents the reference voltage of DC bus n; Represents the DC bus assembly. Indicates connection to DC bus A collection of converters.

[0097] The AC and DC power sides of the converter satisfy the coupling relationship; when the converter is a rectifier:

[0098] When the converter is an inverter:

[0099] In the formula, Indicates converter During the period DC-side power (with the positive direction of power injected from the converter into the DC network). Indicates converter During the period AC side power (with the power injected from the converter into the AC system as the positive direction); This indicates the converter efficiency.

[0100] Step S5: By solving the single-scenario full-time domain recovery optimization model, the optimal recovery path and operation scheme in the complete time domain are obtained.

[0101] It should be noted that the recovery decision needs to weigh the improvement of load recovery level against the suppression of frequency deviation, reduction of operational costs and safety control costs. Therefore, in order to select the optimal solution that takes into account multiple recovery objectives in the feasible region, it is necessary to construct a comprehensive objective function that includes load loss, frequency deviation, recovery action cost and stability control cost, and call a mixed integer linear programming solver to solve it.

[0102] Specifically, a comprehensive objective function for a single-scenario, full-time-domain recovery optimization model is constructed. The comprehensive objective function is obtained by multiple weighted combinations of load loss cost, frequency deviation cost, recovery action cost, low-frequency load shedding cost, and high-frequency tripping cost.

[0103] The comprehensive objective function is specifically expressed by the following formula:

[0104] In the formula, Indicates the cost of load loss. Indicates the cost of frequency deviation. Indicates the cost of resuming the action. This indicates the cost of low-frequency load reduction. This indicates the cost of high-frequency switching. The preset weight for load loss cost, The preset weights for the frequency deviation cost, To restore the preset weights of the action cost, Preset weights for the cost of low-frequency load reduction The preset weights for the high-frequency switching cost; J represents the comprehensive objective function value. This represents the function that takes the minimum value. In this embodiment... , , , , In this embodiment, for , , , and No specific limitations are set; the implementer can decide based on the specific circumstances.

[0105] Specifically, the cost of load loss is expressed as follows:

[0106] In the formula, Indicates the time interval for the recovery decision (i.e., the duration of a time period); Indicates load During the period The actual power value, Indicates load The rated power requirement; T represents the load set; T represents the time period set, i.e., the time domain set. , This represents the total number of steps taken during recovery.

[0107] The frequency deviation cost is specifically expressed as follows:

[0108] In the formula, Supernode During the period The frequency variable is denoted by R, which represents the set of all supernodes. It is the absolute value symbol.

[0109] The cost of the recovery action is specifically expressed as follows:

[0110] In the formula, Indicates generator During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery is performed); Indicates load During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery is performed); Indicates AC line During the period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery is performed); This indicates that converter c is in time period The recovery action variable (1 indicates recovery is performed, 0 indicates no recovery is performed); Represents a set of generators. Represents the load set. Indicates a set of communication lines. This represents a set of converters.

[0111] The costs of low-frequency load shedding and high-frequency tripping are specifically expressed as follows:

[0112] In the formula, Supernode During the period Low-frequency load reduction power, Supernode During the period High-frequency cutting power.

[0113] By using the above objective function, the present invention can improve the load recovery level while avoiding over-reliance on protection actions and suppressing frequency deviations during the recovery process.

[0114] The MILP solver (such as CPLEX, Gurobi, SCIP, etc.) is called to solve the single-scenario full-time domain recovery optimization model in one go. After the solution is successful, the result is output. The optimal recovery path and operation scheme in the full time domain are obtained from the output result.

[0115] The output results include: the recovery sequence, online status, power setting, frequency response, and safety control measures of the equipment in each time period, forming a complete recovery plan.

[0116] The optimal recovery path refers to which equipment (generators, loads, lines, converters) needs to perform recovery operations in each time period and their order, i.e., the complete equipment recovery sequence. The operation plan refers to the set values ​​of system operation parameters such as generator active power output, load power supply, line active power flow, converter AC / DC power, bus energization status, overload frequency, low-frequency load shedding, and high-frequency generator tripping in each time period.

[0117] MILP (Mixed Integer Linear Programming) is a solution for mixed-integer linear programming; SCIP (Solving Constraint Integer Programs) is a solution for constrained integer programming. CPLEX and Gurobi are product names.

[0118] The following specific example illustrates the implementation process of this solution: To further illustrate the implementation process of the AC / DC system fault-based full-time domain recovery decision method described in this invention, the IEEE 39-bus system is used as an example below. Figure 2 The network structure of the example system is given, including equipment such as AC buses, generators, loads, and AC lines. Based on Figure 2 The system shown first obtains the system network topology, device parameters, initial operating status and initial fault set according to step S1, and forms an available topology map after the fault based on the initial fault set.

[0119] In this embodiment, the initial fault conditions are shown in Table 1. There are 7 initial generator faults, 16 initial load faults, 42 initial line faults, 0 initial converter faults, and 32 initial power outage buses. In Table 1, all numbers are internal numbers for the example; "—" indicates no corresponding initial fault device. Since the converter did not experience an initial fault, this embodiment focuses on examining the impact of coordinated restoration of generators, loads, and lines on the system's power supply range, load restoration level, network energization status, and frequency security status.

[0120] Under the aforementioned fault conditions, the initial power supply range of the system is limited by both line disconnection and bus de-energization. For loads connected to the de-energized bus, their power supply is limited to zero; for generators and converters connected to the de-energized area, their power output or power exchange is also constrained by the energized state of the bus. This approach ensures that both the initial state and the subsequent recovery process meet basic electrical feasibility requirements, avoiding the unreasonable situation where the de-energized bus continues to supply power to loads or exchange power with equipment.

[0121] In this embodiment, the recovery step size is 2 minutes, and the recovery process covers 0 minutes to 12 minutes, including a total of 6 recovery decision periods. The optimization model uniformly determines the recovery action for each period across the entire time domain. Table 2 shows the number of newly added recovery devices in each recovery step. As shown in Table 2, in step 1, 1 generator, 2 loads, and 6 lines are restored, for a total of 9 restored devices; in step 2, 2 generators, 5 loads, and 10 lines are restored, for a total of 17 restored devices; in step 3, 1 generator, 5 loads, and 9 lines are restored, for a total of 15 restored devices; in step 4, no generators are restored, but 2 loads and 8 lines are restored, for a total of 10 restored devices; in step 5, 2 generators, 2 loads, and 6 lines are restored, for a total of 10 restored devices; in step 6, 1 generator and 3 lines are restored, but no loads are restored, for a total of 4 restored devices. In the entire recovery process, a total of 7 generators, 16 loads, and 42 lines are restored, consistent with the initial number of faulty devices in Table 1. Since no initial converter fault was set in this embodiment and no converter recovery action occurred during the recovery process, Table 2 only lists the number of generators, loads and lines that were restored.

[0122] Tables 1 and 2 are shown below in detail: Table 1 Initial Fault Status Table

[0123] Table 2. Statistics on the Gradual Restoration of Equipment

[0124] Figure 3 The changes in load supply rate and load loss rate during the recovery process are presented. At 0 min, the system load supply rate was 32.47%; at 2 min, 4 min, and 6 min, the load supply rate increased to 41.14%, 59.23%, and 82.19%, respectively, while the corresponding load loss rates decreased to 58.86%, 40.77%, and 17.81%; at 8 min, the load supply rate reached 99.73%, and the load loss rate decreased to 0.27%, with 16.7 MW of load still not restored; at 10 min, the load supply rate reached 100%, and remained at full supply for 12 min. These results demonstrate that the present invention can rapidly restore load power and continue to restore the remaining network and equipment in subsequent periods.

[0125] Further analysis of Table 2 reveals that the improvement in load power supply rate is closely related to the coordinated restoration of generators, loads, and lines. The first three steps restored a total of 4 generators, 12 loads, and 25 lines, continuously improving the system's power supply capacity. The fourth step restored 2 loads and 8 lines, bringing the load power supply rate to 99.73% after 8 minutes. The fifth step restored 2 generators, the remaining 2 load devices, and 6 lines, bringing the load power supply rate to 100% after 10 minutes. The sixth step continued by restoring the remaining 1 generator and 3 lines to complete the restoration of all equipment and the lost busbar. This demonstrates that the model can prioritize restoring load power and subsequently complete the restoration of the remaining network and equipment.

[0126] Figure 4 The changes in the number of de-energized buses during the restoration process are presented. At 0 min, the number of de-energized buses was 32; at 2 min, 4 min, 6 min, 8 min, and 10 min, it decreased to 26, 18, 10, 6, and 1, respectively; at 12 min, the number of de-energized buses dropped to 0. As generators and lines were gradually restored, the energized area of ​​the system continued to expand. Figure 3 In comparison, at 8 minutes, the load power supply rate reached 99.73%, but 6 buses were still out of power; at 10 minutes, the load power supply rate reached 100%, but 1 bus was still out of power; and it was not until 12 minutes that all buses were restored. This result shows that the model can prioritize restoring load power supply and continue to restore the remaining buses and network equipment in subsequent time periods.

[0127] From the perspective of recovery mechanism, Figure 4 The continuous decrease in the number of power-loss buses was mainly due to the combined effect of generator restoration and line restoration, as shown in Table 2. Generator restoration increased the number of buses and subnets that could provide power support, while line restoration changed the network connectivity structure after the fault, allowing buses that were originally disconnected from the generator support area to gradually reconnect to the energized network. Simultaneously, load restoration gradually restored power to the loads in the energized area. All loads were restored within the first five restoration periods, but one bus remained power-loss-free for 10 minutes. The sixth restoration period completed the final bus's energization by restoring the remaining generators and lines. This process demonstrates the inherent connection between "power support restoration—energized connectivity determination—line restoration—load restoration" in the method of this invention.

[0128] Figure 5The changes in the absolute value of the maximum frequency difference during the recovery process are presented. At 0 min, the absolute value of the maximum frequency difference is approximately 0.0078 pu, indicating that there is a certain frequency deviation in the system after the initial fault. At 2 min, the absolute value of the maximum frequency difference rapidly decreases to near 0 and remains close to 0 during the subsequent recovery process from 2 min to 12 min. This result demonstrates that the full-time domain recovery optimization model established in this invention, while coordinating the recovery of generators, loads, and lines, can avoid new significant frequency deviations caused by subsequent recovery actions through frequency consistency constraints, generator support subgrid discrimination, and low-frequency load shedding and high-frequency generator tripping constraints, and limits the frequency deviation during the recovery process to an acceptable range.

[0129] Combination Figures 3 to 5 The recovery process can be divided into three stages. The first stage, from 0 to 2 minutes, involves the initial restoration of generators, loads, and lines, leading to an increase in load power supply and a decrease in the absolute value of the maximum frequency difference to near zero. The second stage, from 2 to 8 minutes, sees the load power supply rate increase from 41.14% to 99.73%, and the number of out-of-power buses decrease from 26 to 6, indicating that the system has essentially restored load power supply. The third stage, from 8 to 12 minutes, involves the restoration of the remaining 2 load devices, 2 generators, and 6 lines during the fifth recovery period, bringing the load power supply rate to 100% by 10 minutes. The sixth recovery period restores the remaining 1 generator and 3 lines, completing the restoration of all faulty equipment and out-of-power buses by 12 minutes. This process demonstrates that the present invention can prioritize the restoration of load power supply and subsequently complete the restoration of the remaining network and equipment.

[0130] The results of this embodiment also show that, in the case of initial faults in generators, loads, and lines, where the converter does not participate in the recovery action, the system recovery effect depends on the power support provided by generator recovery, the network connectivity improved by line recovery, and the power supply level improved by load recovery. The model can uniformly arrange the recovery actions of 7 generators, 16 loads, and 42 lines throughout the entire time domain, and achieve full load supply before the entire network recovery while maintaining frequency safety. If there is an initial fault in the converter in other embodiments, the model of this invention can also incorporate converter recovery into the unified optimization framework throughout the entire time domain through converter state transition constraints, DC node equations, and converter power coupling constraints. Therefore, this invention is not only applicable to the AC system recovery scenario shown in this embodiment, but also to the AC / DC hybrid system recovery scenario containing converters and DC networks.

[0131] In summary, this embodiment demonstrates that the present invention can uniformly optimize the status of generators, loads, lines, and converters within the complete recovery time domain under known initial fault conditions. All 7 initially faulted generators, 16 initially faulted loads, and 42 initially faulted lines were restored within 6 recovery periods; the load power supply rate increased from 32.47% to 99.73% at 8 minutes, and reached 100% at 10 minutes; all 32 initially de-energized buses were restored to energization within 12 minutes; the initial maximum frequency difference absolute value was approximately 0.0078 pu, and decreased to near zero within the first recovery period. These results verify the effectiveness of the method in equipment recovery, load recovery, topology recovery, and frequency safety control.

[0132] This concludes the embodiment.

[0133] A second aspect of the present invention is to provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement a full-time-domain recovery decision method after an AC / DC system fault.

[0134] A third aspect of the present invention is to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements a full-time-domain recovery decision-making method for AC / DC system faults.

[0135] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, optical storage, etc.) containing computer-usable program code.

[0136] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, systems, and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0137] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0138] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0139] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the present invention.

Claims

1. A full-time-domain recovery decision-making method for AC / DC system after a fault, characterized in that, include: Obtain the network topology, equipment parameters, initial operating status, and initial fault set of the AC / DC system; The network topology is processed by the initial fault set to obtain an available topology graph; Determine available generators; based on the available topology and available generators, identify and obtain the initial set of energized buses and the initial set of de-energized buses; then perform state closure processing on the equipment connected to the de-energized buses, setting the power variables of the equipment to zero or constraining them to the energized state of the buses; Obtain the fixed online AC lines of the AC power grid in the AC / DC system within the recovery time domain, establish connectivity through the fixed online AC lines, aggregate the buses in the AC power grid into AC supernodes; establish a mapping from the buses to the supernodes in the AC power grid, and use the recoverable AC lines in the AC power grid as recoverable edges between supernodes to construct an AC supernode network; The available topology, initial operating state, initial set of energized buses, and initial set of de-energized buses are used as initial boundary conditions. A single-scenario full-time domain recovery optimization model is constructed using the initial boundary conditions, equipment parameters, and AC supernode network. By solving the single-scenario full-time domain recovery optimization model, the optimal recovery path and operation scheme within the complete time domain are obtained; When solving the single-scenario full-time domain recovery optimization model, if there is no DC grid in the AC / DC system, the solution is directly obtained by constructing AC constraints; if there is a DC grid in the AC / DC system, both AC constraints and DC constraints are constructed and solved together.

2. The AC / DC system fault full-time domain recovery decision method according to claim 1, characterized in that, The process of obtaining the available topology map is as follows: The initial fault set is removed from the network topology of the AC / DC system to form the initial system topology after the fault, which is denoted as the available topology graph.

3. The full-time-domain recovery decision-making method for AC / DC system after a fault, as described in claim 1, is characterized in that... The process of obtaining the initial set of energized buses and the initial set of de-energized buses is as follows: Using the bus where the available generator is located as the power source, a connectivity search is performed in the available topology map to determine whether each bus is connected to any available generator, and the connectivity determination results of all buses are obtained. Based on the connectivity determination results, the buses connected to available generators are assigned to the initial set of energized buses, and the buses that cannot be connected are assigned to the initial set of de-energized buses. The specific process of busbar division is as follows: In the connectivity determination result, it is determined whether each bus has a connected path with at least one available generator in the available topology graph. If so, the bus is in a live state; otherwise, the bus is in a de-energized state. All buses in the live state are assigned to the initial live bus set, and all buses in the de-energized state are assigned to the initial de-energized bus set.

4. The full-time-domain recovery decision-making method for AC / DC system after a fault, as described in claim 1, is characterized in that... The process of obtaining the communication supernode is as follows: Using fixed online AC lines as connecting edges, a connectivity analysis is performed on all AC buses, and interconnected buses are divided into the same connected component; each connected component is defined as an AC supernode.

5. The full-time-domain recovery decision-making method for AC / DC system after a fault, as described in claim 1, is characterized in that... The AC constraints include: equipment status and recovery action logic constraints, bus energized status and equipment power constraints, AC active power balance constraints, AC line power flow constraints, frequency consistency and generation support subgrid constraints, and low-frequency load shedding and high-frequency generator tripping constraints.

6. The AC / DC system fault full-time domain recovery decision method according to claim 1, characterized in that, The DC constraint is the coupling constraint between the DC network and the converter.

7. The full-time-domain recovery decision-making method for AC / DC system after a fault, as described in claim 1, is characterized in that... The single-scenario full-time domain recovery optimization model includes a comprehensive objective function; the comprehensive objective function consists of load loss cost, frequency deviation cost, recovery action cost, low-frequency load shedding cost, and high-frequency tripping cost; Specifically, the cost of load loss is expressed as follows: In the formula, Indicates the time interval for the decision to resume; Indicates load During the period The actual power value, Indicates load The rated power requirement; T represents the load set; T represents the time period set. The frequency deviation cost is specifically expressed as follows: In the formula, Supernode During the period The frequency variable is denoted by R, which represents the set of all supernodes. It is the absolute value symbol; The cost of the recovery action is specifically expressed as follows: In the formula, Indicates generator During the period The recovery action variable; Indicates load During the period The recovery action variable; Indicates AC line During the period The recovery action variable; This indicates that converter c is in time period The recovery action variable; Represents a set of generators. Represents the load set. Indicates a set of communication lines. Represents a set of converters; The costs of low-frequency load shedding and high-frequency tripping are specifically expressed as follows: In the formula, Supernode During the period Low-frequency load reduction power, Supernode During the period High-frequency cutting power; in, Indicates the cost of load loss. Indicates the cost of frequency deviation. Indicates the cost of resuming the action. This indicates the cost of low-frequency load reduction. This indicates the cost of high-frequency switching.

8. The AC / DC system fault full-time domain recovery decision method according to claim 7, characterized in that, The comprehensive objective function is specifically expressed by the following formula: In the formula, The preset weight for load loss cost, The preset weights for the frequency deviation cost, To restore the preset weights of the action cost, Preset weights for the cost of low-frequency load reduction Preset weights for the cost of high-frequency switching; J represents the value of the comprehensive objective function. This represents the function that takes the minimum value.

9. An electronic device, characterized in that, The system includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the full-time-domain recovery decision method for AC / DC system faults as described in any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the full-time-domain recovery decision method for AC / DC system faults as described in any one of claims 1-8.

Images