A Black Start Method and System for Transmission Networks Based on Robust Model Predictive Control
By dividing the black-start recovery zone in the transmission network and constructing a robust model predictive control method, the problems of robust safety and online decision-making timeliness in the black-start recovery of the transmission network under high proportion of wind power access are solved, and efficient frequency-stable recovery and smooth grid connection of multiple zones are achieved.
Patent Information
- Application Number
- CN202611213223.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-11
- Publication Date
- 2026-11-17
AI Technical Summary
In the black start recovery of transmission networks with a high proportion of wind power integration, it is difficult to balance the robustness and safety of the recovery strategy under wind power output fluctuations with the timeliness of online decision-making.
A robust model-based predictive control method is adopted to divide the post-disaster power grid topology model into multiple black-start recovery partitions, construct an uncertainty set, perform dual transformation through a two-layer robust optimization model to generate a recovery sequence, and perform feedback correction in the rolling time domain until the partition recovery is completed and grid connection is achieved under the condition of synchronization.
It improved the recovery efficiency and operational resilience of the power grid after disasters, ensured frequency stability, and enhanced the robustness of recovery strategies and the timeliness of online decision-making under wind power fluctuations.
Smart Images

Figure CN123121789A_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power system restoration and resilient operation technology, specifically relating to a black start method and system for power transmission networks based on robust model predictive control. Background Technology
[0002] With the large-scale integration of renewable energy sources such as wind power and the frequent occurrence of extreme disasters, the safe operation and resilience of the power system face severe challenges. Disasters such as ice storms, typhoons, and earthquakes can easily cause large-scale transmission line breaks, equipment damage, and even major power outages. Rapid, safe, and orderly black-start and restoration of the power grid after a disaster has become an important issue.
[0003] In post-disaster recovery, the grid structure, black-start turbines, and loads to be restored are usually known; the challenge lies in addressing the intermittency and volatility of wind power. Existing research on this topic falls into three categories: deterministic optimization uses predicted values to determine the time series, which is simple but ignores errors and is prone to infeasibility under volatility; stochastic optimization uses a large number of scenarios to characterize probability distributions, but the scenarios are large, computationally burdensome, and dependent on difficult-to-obtain distribution information; robust optimization uses uncertain sets to describe the range of volatility and optimizes the worst-case scenario, ensuring robust feasibility without precise distribution, but static methods are too conservative; and Adaptive Robust Optimization (ARO) requires offline column-and-constraint generation (C&CG) solutions, which are time-consuming and difficult to adapt to online rolling decision-making.
[0004] On the other hand, Model Predictive Control (MPC), with its rolling optimization and feedback correction, can incorporate the latest measurements to refine decisions, making it suitable for highly time-varying recovery. However, it is largely based on deterministic predictions and struggles to directly account for wind power uncertainties. Furthermore, the limited capacity, low inertia, and weak disturbance resistance in the initial recovery phase mean that severe wind power fluctuations can easily lead to excessive frequency change rates, frequency or voltage exceeding limits, interrupting recovery or even causing secondary power outages. In summary, current black-start recovery strategies for transmission networks with a high proportion of wind power integration are relatively difficult to balance robustness and safety with the timeliness of online decision-making under wind power output fluctuations. Summary of the Invention
[0005] This invention provides a black start method and system for power transmission networks based on robust model predictive control. The purpose is to solve the problem that in the current black start recovery of power transmission networks with a high proportion of wind power access, it is relatively difficult to balance the robustness and safety of the recovery strategy under the fluctuation of wind power output with the timeliness of online decision-making.
[0006] To achieve the above objectives, the present invention adopts the following technical solution: This invention provides a black start method for power transmission networks based on robust model predictive control, comprising the following steps: Based on the black start capability of each power source, the spatial distribution of power sources, and the network connectivity, the nodes in the post-disaster power grid topology model are divided into multiple black start recovery zones, each containing one black start unit as the starting power source. For the wind power output fluctuations within each black-start recovery zone, an uncertainty set is constructed to describe the uncertainty of wind power output. The dual transformation of the two-layer robust optimization model is performed to obtain the single-layer deterministic optimization model. The control decision vector at the current rolling moment is obtained by solving the model. The recovery instruction of the first control period is executed. The system state is updated based on the measured data at the next rolling moment and the rolling solution is continued until the partition recovery is completed. After each black boot recovery partition is running stably, when it is detected that each partition meets the network synchronization conditions, the connection line between the partitions is closed to achieve network connection.
[0007] In some implementations, the construction of the post-disaster power grid topology model includes: obtaining the power grid topology information, available power source information, and load information to be restored after extreme disasters; the construction of the two-layer robust optimization model includes: being constructed in the rolling time domain based on an uncertainty set, with the goal of minimizing the weighted outage loss under the worst wind power scenario.
[0008] In some implementations, the first is defined The available power generation capacity of each partition under robust conditions is: ; A partition can be used as an initial black start recovery partition when its power generation capacity meets the following conditions: ; in, For the first A set of partition nodes, This refers to the node set of a conventional thermal power unit. For wind turbine nodes; For the set of load nodes, For nodes The maximum active power output of the unit. For wind turbines Minimum available output under an uncertain set This refers to the critical load ratio factor. For load demand; This is the starting power for plant use.
[0009] In some implementations, within a rolling time-domain optimization framework, a system state equation describing the dynamic evolution of the post-disaster recovery process is established: ; in, For the rolling moment The system recovery state vector includes the unit startup state, the energized state of the bus and lines, the active power output of the unit, and the load recovery amount, etc. For the rolling moment The control decision vector; For the rolling moment The measured wind power output disturbance vector; A , B , C These are the state transition matrix, control input matrix, and disturbance input matrix, respectively.
[0010] In some implementations, the uncertainty set is a budgetary polyhedral uncertainty set, denoted as: ; in, Predict the available output of wind turbine units; This refers to the fluctuation amplitude. , These represent positive and negative fluctuations, respectively. For the set of recovery periods; For uncertain budget parameters.
[0011] In some implementations, the two-layer robust optimization model is as follows: ; Among them, the decision vector , The unit is in startup status; Bus / line charging status; To provide power to the generator unit; This is the amount of load recovery; These are the weighting coefficients; Power outage; To constrain the operation safety of the generator set and network.
[0012] Furthermore, an upper bound variable is introduced into the objective function of the two-layer robust optimization model. The robust equivalence is: ; in, , which is the upper bound variable of the objective function; To address the uncertain budget for wind power The corresponding dual variable; To constrain wind power fluctuations in different time periods The corresponding dual variable; For wind power output in the objective function The corresponding coefficient; To predict node speeds in wind power scenarios Time period Power outage.
[0013] In some implementations, for any operational constraints containing uncertainties Its equation is transformed into: ; in: , Constraints Decision variables With uncertainty The coefficient vector; To constrain The right-hand side constant; The vector of uncertainties in wind power output ( ); Constraint numbering; To constrain Corresponding wind turbine Time period The uncertainty coefficient; To be in accordance with constraints Uncertain budget The corresponding dual variable; To be in accordance with constraints The dual variables corresponding to the fluctuation constraints in each time period.
[0014] In some implementations, during the initial stage of rolling recovery, the system satisfies the maximum rate of change constraint to ensure frequency stability: ; in: This is the system's rated power; This represents a power step change. The inertia constant of the unit; Based on capacity; The maximum permissible rate of frequency change; The conditions for grid connection and synchronization are: ; in, , For partitioning , The system frequency; , For partitioning , The voltage amplitude; , For partitioning , The phase angle; , , These are the allowable frequency, voltage, and phase angle deviation thresholds, respectively.
[0015] This invention also provides a black-start system for a power transmission network based on robust model predictive control (BMR), used to implement the aforementioned black-start method for a power transmission network based on robust model predictive control (BMR), including a data acquisition and modeling module, a black-start partitioning module, a robust optimization and rolling solution module, and a grid synchronization module; wherein: The data acquisition and modeling module is used to divide the nodes in the post-disaster power grid topology model into multiple black start recovery partitions based on the black start capability of each power source, the spatial distribution of power sources, and the network connectivity. Each partition contains one black start unit as the starting power source. The black start partitioning module is used to: construct an uncertainty set to describe the uncertainty of wind power output for wind power output fluctuations within each black start recovery partition; The robust optimization and rolling solution module is used to: perform dual transformation on the two-layer robust optimization model to obtain a single-layer deterministic optimization model; solve for the control decision vector at the current rolling moment; execute the recovery instruction of the first control period; and continue rolling solution after updating the system state based on the measured data at the next rolling moment until the partition recovery is completed. The grid synchronization module is used to: after each black boot recovery partition is running stably, when it is detected that the grid synchronization conditions between each partition are met, close the connection line between the partitions to achieve grid connection.
[0016] Compared with existing technologies, the black start method and system for power transmission networks based on robust model predictive control of the present invention has the following advantages: This invention constructs a post-disaster power grid topology model by integrating black-start partitioning strategy, rolling recovery state equation modeling, budgetary uncertainty set modeling, and robust model predictive control. It generates a recovery sequence and performs feedback correction. By combining robust equivalence transformation based on duality theory, the min-max two-level model is equivalently transformed into a single-level deterministic problem that can be directly solved, replacing the existing two-stage iterative solution and improving the timeliness of decision-making to a certain extent. Furthermore, this invention ensures the relative frequency stability of the system under wind power fluctuations in the early stages of recovery through rolling time-domain feedback correction and RoCoF constraints. Then, by using grid connection criteria to achieve smooth merging of multiple partitions, it effectively improves the recovery efficiency and operational resilience of the post-disaster transmission network, demonstrating certain engineering practical value. Attached Figure Description
[0017] The accompanying drawings are provided to further understand the invention and constitute a part of this invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.
[0018] Figure 1 This is a flowchart illustrating an embodiment of a black start method for a power transmission network based on robust model predictive control according to the present invention. Figure 2 This is a schematic diagram of the rolling optimization mechanism in an embodiment of a black-start system for a power transmission network based on robust model predictive control according to the present invention; Figure 3 This is a schematic diagram of the IEEE 39-node system topology in an embodiment of a black-start system for a power grid based on robust model predictive control according to the present invention. Figure 4 This is a schematic diagram of the system load recovery curve in an embodiment of a black start system for a power transmission network based on robust model predictive control according to the present invention. Detailed Implementation
[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.
[0020] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.
[0021] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.
[0022] It should be noted that the apparatus and methods disclosed in the embodiments herein can also be implemented in other ways. The apparatus embodiments described above are merely illustrative; for example, the flowcharts and block diagrams in the accompanying drawings show the architecture, functionality, and operation of possible implementations of apparatus, methods, and computer program products according to various embodiments herein. In this regard, each block in a flowchart or block diagram may represent a module, program, or part of code containing one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than those marked in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram and / or flowchart, and combinations of blocks in block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system to perform the specified function or action, or can be implemented using a combination of dedicated hardware and computer instructions.
[0023] In addition, the functional modules in the various embodiments of this article can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.
[0024] How to provide a black start method for power transmission networks, using black start units as the starting power source for zones, establishing recovery state equations in the rolling time domain, constructing a two-layer robust optimization model to handle wind power fluctuations, and achieving safe and gradual recovery of the power grid after disasters through rolling optimization and feedback correction.
[0025] Based on this, the present invention provides a black start method for power transmission networks based on robust model predictive control, comprising the following steps: Based on the black start capability of each power source, the spatial distribution of power sources, and the network connectivity, the nodes in the post-disaster power grid topology model are divided into multiple black start recovery zones, each containing one black start unit as the starting power source. For the wind power output fluctuations within each black-start recovery zone, an uncertainty set is constructed to describe the uncertainty of wind power output. The dual transformation of the two-layer robust optimization model is performed to obtain the single-layer deterministic optimization model. The control decision vector at the current rolling moment is obtained by solving the model. The recovery instruction of the first control period is executed. The system state is updated based on the measured data at the next rolling moment and the rolling solution is continued until the partition recovery is completed. After each black boot recovery partition is running stably, when it is detected that each partition meets the network synchronization conditions, the connection line between the partitions is closed to achieve network connection.
[0026] This invention addresses the issue of large-scale power outages in transmission networks caused by extreme disasters. It constructs a post-disaster topology model based on a given grid structure, available power sources, and loads to be restored. The model then divides the network into non-overlapping black-start recovery zones, each powered by a self-starting black-start unit. To address the stochastic fluctuations in wind power output after a disaster, this invention employs a budget-type polyhedral uncertainty set for interval modeling. Robust model predictive control is used in the rolling time domain to establish a system state equation reflecting the dynamic evolution of the recovery process. A two-layer robust optimization model is constructed with the objective of minimizing the weighted outage loss under the worst-case wind power scenario. The black-start recovery zone undergoes a robust equivalence transformation based on duality theory, converting the inner worst-case scenario problem, which considers wind power uncertainty, into deterministic constraints. This transforms the two-layer model into a single-layer deterministic mixed-integer linear programming problem for direct solution. At each rolling time step, only the recovery command for the first control period is executed, with feedback correction and state updates based on measured data. Finally, under the condition of frequency, voltage and phase angle synchronization, the closed tie line realizes the gradual grid connection and restoration of multiple zones; thus, it can effectively cope with the uncertainty of wind power output, improve the recovery efficiency and operational resilience of the transmission network after disaster, and has certain practical significance.
[0027] Specifically, this invention aims to improve the uncertainty challenges and online solution timeliness issues faced by the black start of the transmission network under high-proportion wind power integration. By constructing a post-disaster coverage model, generating a recovery sequence, and performing feedback correction, the min-max two-layer model is equivalently transformed into a single-layer deterministic problem for direct solution using robust equivalence transformation based on duality theory. In conjunction with feedback correction in the rolling time domain and the maximum frequency change rate constraint, the frequency safety of the recovery process is ensured. Furthermore, the smooth merging of multiple zones is achieved through grid connection criteria, effectively improving the recovery efficiency and operational resilience of the transmission network after a disaster.
[0028] In some operating conditions, such as Figure 1 As shown, the present invention provides a black start method for power transmission networks based on robust model predictive control, which is carried out according to the following steps: Step 1: Based on the known grid structure, available power sources, and loads to be restored after a large-scale power outage caused by an extreme natural disaster, establish a post-disaster power grid topology model. Divide the system into several non-overlapping recovery zones, each containing one black-start generator unit, according to generator capacity and network connectivity. The power grid topology is represented as follows: ; in: ; in: The set of all nodes in the system; This refers to the set of nodes where the black-start generator sets are located. This refers to the node set of a conventional thermal power unit. For wind turbine nodes; For the set of load nodes; This is a set of routes.
[0029] Based on the unit's black start capability, power supply spatial distribution, and network connectivity, the system is divided into several non-overlapping black start recovery partitions, each represented as: ; in: ; in: For the set of partition numbers; For the first A set of partition nodes, each partition having exactly one black boot machine group as the boot source; define the first... The available power generation capacity of each partition under robust conditions is: ; in: For nodes The maximum active power output of the unit; For wind turbines Minimum available output under uncertainty. A partition is deemed suitable as an initial black-start recovery partition when its power generation capacity meets the following conditions: ; in: This refers to the critical load ratio factor. For load demand; This is the starting power for plant use.
[0030] like Figure 2 As shown, step 2, based on the rolling time-domain mechanism, establishes a system state equation reflecting the dynamic evolution of the post-disaster recovery process, describing the transition relationship and irreversibility of the recovery state between adjacent rolling moments, and providing dynamic constraints for subsequent rolling optimization; the system state equation is: ; in: For the rolling moment The system recovery state vector includes the unit startup state, the energized state of the bus and lines, the active power output of the unit, and the load recovery amount, etc. For the rolling moment The control decision vector; For the rolling moment The measured wind power output disturbance vector; A , B , C These are the state transition matrix, control input matrix, and disturbance input matrix, respectively.
[0031] The system recovery state vector is updated hourly based on control decisions and measured wind power output. Considering the irreversibility of the post-disaster recovery process, the started units, charged buses, and lines should remain energized in subsequent time steps, satisfying the following monotonicity constraint: ; in, For the unit At any moment The startup state variables, For line (or busbar) At any moment The energized state variables are both 0 or 1, representing outage and energized states respectively. At each rolling time, only the recovery command for the first control period is executed, and the system state is updated with the measured wind power output before rolling forward to solve the problem.
[0032] Step 3: Based on the random fluctuation characteristics of post-disaster wind power output, a budget-type polyhedral uncertainty set is used to perform interval modeling of the predicted wind power output and characterize its uncertainty boundary.
[0033] The set of uncertain wind power output is: ; in, Predict the available output of wind turbine units; This refers to the fluctuation amplitude. , These represent positive and negative fluctuations, respectively. For the set of recovery periods; For uncertain budget parameters.
[0034] Step 4: Based on the rolling time domain mechanism, with the goal of minimizing the weighted outage loss under the worst wind power scenario, a two-layer robust optimization recovery model is constructed to comprehensively optimize unit startup, line charging and load restoration.
[0035] The decision vector contains all decision variables in the prediction time domain: ; The min-max robust optimization recovery model is: ; in, The unit is in startup status; Bus / line charging status; To provide power to the generator unit; This is the amount of load recovery; These are the weighting coefficients; Power outage; To ensure the safety of unit and network operations, the system executes only the recovery command for the first time period at any given moment before rolling forward.
[0036] Step 5: Based on real-time measurement data, feedback correction is performed on the wind power output deviation, and the maximum frequency change rate constraint is introduced to realize the dynamic update of the recovery command taking into account real-time fluctuations, so as to ensure frequency stability.
[0037] The formula for correcting wind power predicted output is: ; in: For the current scrolling moment The measured output force; This is due to output deviation; This is the feedback correction coefficient.
[0038] Meanwhile, during the initial stage of rolling recovery, the system satisfies the maximum rate of change constraint to ensure frequency stability: ; in: This is the system's rated power; This represents a power step change. The inertia constant of the unit; Based on capacity; This represents the maximum permissible rate of frequency change.
[0039] Step 6: Based on duality theory, perform dual transformation on the inner worst-case scenario problem of the min-max bilayer robust model, derive its robust equivalence, and convert the original bilayer model into a single-layer deterministic mixed-integer linear programming problem to meet the requirement of online fast solution at rolling time.
[0040] Introduce an upper bound variable to the objective function Its robust equivalence is transformed into: ; in: , which is the upper bound variable of the objective function; To address the uncertain budget for wind power The corresponding dual variable; To constrain wind power fluctuations in different time periods The corresponding dual variable; For wind power output in the objective function The corresponding coefficient; To predict node speeds in wind power scenarios Time period Power outage.
[0041] For any operational constraints containing uncertainties Its equation is transformed into: ; in: , Constraints Decision variables With uncertainty The coefficient vector; To constrain The right-hand side constant; The vector of uncertainties in wind power output ( ); Constraint numbering; To constrain Corresponding wind turbine Time period The uncertainty coefficient; To be in accordance with constraints Uncertain budget The corresponding dual variable; To be in accordance with constraints The dual variables corresponding to the fluctuation constraints in each time period.
[0042] By combining the above robust equivalence equations, a single-layer deterministic equivalent model is constructed and solved directly at each rolling time step.
[0043] Step 7: Based on the stable operating status of each zone, monitor the frequency, voltage, and phase angle grid connection conditions. When the synchronization conditions are met, close the inter-zone tie lines and gradually expand the recovery range until the entire network is restored. During the parallel recovery phase, the inter-zone tie lines remain disconnected to ensure electrical isolation. ; in, This is a collection of inter-regional connection lines.
[0044] Once multiple partitions are running stably, network reconnection is achieved through synchronization criteria. The network reconnection conditions are: ; in: , For partitioning , The system frequency; , For partitioning , The voltage amplitude; , For partitioning , The phase angle; , , These are the allowable frequency, voltage, and phase angle deviation thresholds, respectively.
[0045] After the above-mentioned frequency, voltage and phase angle synchronization conditions are met, the relevant tie lines are closed, and the entire power grid system is gradually restored.
[0046] This invention also provides a black-start system for power transmission networks based on robust model predictive control, including a data acquisition and modeling module, a black-start partitioning module, a robust optimization and rolling solution module, and a grid synchronization module; wherein: The data acquisition and modeling module is used to divide the nodes in the post-disaster power grid topology model into multiple black start recovery partitions based on the black start capability of each power source, the spatial distribution of power sources, and the network connectivity. Each partition contains one black start unit as the starting power source. The black start partitioning module is used to: construct an uncertainty set to describe the uncertainty of wind power output for wind power output fluctuations within each black start recovery partition; The robust optimization and rolling solution module is used to: perform dual transformation on the two-layer robust optimization model to obtain a single-layer deterministic optimization model; solve for the control decision vector at the current rolling moment; execute the recovery instruction of the first control period; and continue rolling solution after updating the system state based on the measured data at the next rolling moment until the partition recovery is completed. The grid synchronization module is used to: after each black boot recovery partition is running stably, when it is detected that the grid synchronization conditions between each partition are met, close the connection line between the partitions to achieve grid connection.
[0047] The following experiments will verify the effectiveness of the black start method and system for power transmission networks based on robust model predictive control of the present invention.
[0048] like Figure 3 As shown, a simulation analysis is performed using an improved IEEE 39-node system as an example. The system topology is as follows. Figure 3 : Figure 3 The nodes are numbered 1-39, with G representing conventional generator sets and R representing wind power and other new energy generator sets, to simulate a post-disaster recovery scenario with a high proportion of wind power access. Based on the black start capability of each generator set, the spatial distribution of power sources and network connectivity, the entire network is divided into three non-overlapping black start recovery zones. Zones 1, 2 and 3 use the black start generator sets at nodes 34, 31 and 30, respectively, as the start-up sources. The cross-zone interconnection lines between each zone remain disconnected during the parallel recovery phase to ensure electrical isolation.
[0049] like Figure 4 As shown, the method of this invention was used to perform a black boot recovery on the above system, and the resulting load recovery curves for each partition and the entire network are as follows. Figure 4 : Figure 4The horizontal axis represents recovery time (min), and the vertical axis represents the recovered load (MW). The three solid lines correspond to the recovery processes of zone 1, zone 2, and zone 3, respectively, while the black dashed line represents the total system recovery. Figure 4 It can be seen that under the rolling time-domain mechanism, the restored load of each zone is monotonically non-decreasing as the recovery process progresses, reflecting the irreversibility of the post-disaster recovery process. The three zones independently and smoothly advance load recovery during the parallel recovery phase. Zone 1, Zone 2, and Zone 3 ultimately restored loads of 2317MW, 2384MW, and 1553MW, respectively. After the frequency, voltage, and phase angle synchronization criteria were met, the inter-zone tie line was closed. The total system recovery volume steadily increased to 6200MW within 400 minutes and tended to stabilize, and the load of the entire network was basically restored.
[0050] The results show that the black start method for power transmission networks based on robust model predictive control of the present invention can still ensure frequency safety in the early stage of recovery and achieve efficient and stable grid connection of each zone under the premise of strictly taking into account the uncertainty of wind power output. The verification shows that the present invention has good recovery efficiency and operational resilience under wind power fluctuations.
[0051] In summary, this invention ensures that the recovery strategy meets safety constraints within any fluctuation range of wind power output, improving the vulnerability of deterministic methods to infeasibility under wind power fluctuations. By leveraging a robust equivalence transformation based on duality theory, the two-layer model is equivalently transformed into a single-layer deterministic mixed-integer linear programming problem for direct solution. Combined with a rolling time-domain mechanism, it meets the timeliness requirements for rapid online decision-making. Simultaneously, the introduction of a maximum frequency change rate constraint effectively suppresses the risk of frequency exceeding limits caused by wind power fluctuations. Through partitioned parallel recovery and synchronous grid connection strategies, it achieves stable recovery of the entire network, improving the recovery efficiency and operational resilience of large-scale transmission networks under extreme disasters.
[0052] Finally, it should be noted that the above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Anyone skilled in the art can readily implement the present invention according to the description and above. Any modifications, alterations, or equivalent variations made using the technical content disclosed above are equivalent embodiments of the present invention. Furthermore, any modifications, alterations, or variations made to the above embodiments based on the essential technology of the present invention are still within the protection scope of the present invention.
Claims
1. A black start method for power transmission networks based on robust model predictive control, characterized in that, Includes the following steps: Based on the black start capability of each power source, the spatial distribution of power sources, and the network connectivity, the nodes in the post-disaster power grid topology model are divided into multiple black start recovery zones, each containing one black start unit as the starting power source. For the wind power output fluctuations within each black-start recovery zone, an uncertainty set is constructed to describe the uncertainty of wind power output. The dual transformation of the two-layer robust optimization model is performed to obtain the single-layer deterministic optimization model. The control decision vector at the current rolling moment is obtained by solving the model. The recovery instruction of the first control period is executed. The system state is updated based on the measured data at the next rolling moment and the rolling solution is continued until the partition recovery is completed. After each black boot recovery partition is running stably, when it is detected that each partition meets the network synchronization conditions, the connection line between the partitions is closed to achieve network connection.
2. The black start method for power transmission networks based on robust model predictive control according to claim 1, characterized in that, The construction of the post-disaster power grid topology model includes: obtaining power grid topology information, available power source information, and load information to be restored after extreme disasters; the construction of the two-layer robust optimization model includes: being constructed in the rolling time domain based on an uncertainty set, with the objective of minimizing the weighted outage loss under the worst wind power scenario.
3. The black start method for power transmission networks based on robust model predictive control according to claim 1, characterized in that, Definition of the first The available power generation capacity of each partition under robust conditions is: ; A partition can be used as an initial black start recovery partition when its power generation capacity meets the following conditions: ; in, For the first A set of partition nodes, This refers to the node set of a conventional thermal power unit. For wind turbine nodes; For the set of load nodes, For nodes The maximum active power output of the unit. For wind turbines Minimum available output under an uncertain set This refers to the critical load ratio factor. For load demand; This is the starting power for plant use.
4. The black start method for power transmission networks based on robust model predictive control according to claim 1, characterized in that, Within the rolling time-domain optimization framework, a system state equation describing the dynamic evolution of the post-disaster recovery process is established: ; in, For the rolling moment The system recovery state vector includes the unit startup state, the energized state of the bus and lines, the active power output of the unit, and the load recovery amount, etc. For the rolling moment The control decision vector; For the rolling moment The measured wind power output disturbance vector; A , B , C These are the state transition matrix, control input matrix, and disturbance input matrix, respectively.
5. The black start method for power transmission networks based on robust model predictive control according to claim 1, characterized in that, The uncertainty set is a budgetary polyhedral uncertainty set, represented as: ; in, Predict the available output of wind turbine units; This refers to the fluctuation amplitude. , These represent positive and negative fluctuations, respectively. For the set of recovery periods; For uncertain budget parameters.
6. The black start method for power transmission networks based on robust model predictive control according to claim 1, characterized in that, The two-layer robust optimization model is as follows: ; Among them, the decision vector , The unit is in startup status; Bus / line charging status; To provide power to the generator unit; This is the amount of load recovery; These are the weighting coefficients; Power outage; To constrain the operation safety of the generator set and network.
7. The black start method for power transmission networks based on robust model predictive control according to claim 6, characterized in that, An upper bound variable is introduced into the objective function of the two-layer robust optimization model. The robust equivalence is: ; in, , which is the upper bound variable of the objective function; To address the uncertain budget for wind power The corresponding dual variable; To constrain wind power fluctuations in different time periods The corresponding dual variable; For wind power output in the objective function The corresponding coefficient; To predict node speeds in wind power scenarios Time period Power outage.
8. The black start method for power transmission networks based on robust model predictive control according to claim 1, characterized in that, For any operational constraints containing uncertainties The equation is transformed into: ; in: , Constraints Decision variables With uncertainty The coefficient vector; To constrain The right-hand side constant; The vector of uncertainties in wind power output ( ); Constraint numbering; To constrain Corresponding wind turbine Time period The uncertainty coefficient; To be in accordance with constraints Uncertain budget The corresponding dual variable; To be in accordance with constraints The dual variables corresponding to the fluctuation constraints in each time period.
9. The black start method for power transmission networks based on robust model predictive control according to claim 1, characterized in that, In the initial stage of rolling recovery, the system satisfies the maximum rate of change constraint to ensure frequency stability: ; in: This is the system's rated power; This represents a power step change. The inertia constant of the unit; Based on capacity; The maximum permissible rate of frequency change; The conditions for grid connection and synchronization are: ; in, , For partitioning , The system frequency; , For partitioning , The voltage amplitude; , For partitioning , The phase angle; , , These are the allowable frequency, voltage, and phase angle deviation thresholds, respectively.
10. A black-start system for a power transmission network based on robust model predictive control, used to implement the black-start method for a power transmission network based on robust model predictive control as described in any one of claims 1-9, characterized in that, It includes modules for data acquisition and modeling, black-start partitioning, robust optimization and rolling solution, and grid synchronization; among which: The data acquisition and modeling module is used to divide the nodes in the post-disaster power grid topology model into multiple black start recovery partitions based on the black start capability of each power source, the spatial distribution of power sources, and the network connectivity. Each partition contains one black start unit as the starting power source. The black start partitioning module is used to: construct an uncertainty set to describe the uncertainty of wind power output for wind power output fluctuations within each black start recovery partition; The robust optimization and rolling solution module is used to: perform dual transformation on the two-layer robust optimization model to obtain a single-layer deterministic optimization model; solve for the control decision vector at the current rolling moment; execute the recovery instruction of the first control period; and continue rolling solution after updating the system state based on the measured data at the next rolling moment until the partition recovery is completed. The grid synchronization module is used to: after each black boot recovery partition is running stably, when it is detected that the grid synchronization conditions between each partition are met, close the connection line between the partitions to achieve grid connection.
