Microgrid capacity configuration optimization method and system under abnormal working condition

Abstract

The invention provides a micro-grid capacity configuration optimization method and system under an abnormal working condition, and relates to the technical field of micro-grid capacity configuration.The method comprises the steps that for a target micro-grid, a working condition system is determined by conducting abnormal working condition decomposition, and an action sequence is determined by conducting explicit modeling on an action sequence under the abnormal working condition, the decomposition standard at least comprises a working condition mode and a configuration dimension; according to the working condition system and the action sequence, deploying an initialization port in a power grid central controller, converting an uploaded abnormal working condition configuration task, and determining a task configuration demand; inputting the task configuration requirement into a capacity configurator, executing column constraint optimization generation, and determining a task configuration scheme; and performing capacity configuration control on the target microgrid according to the task configuration scheme. The technical problem that in the prior art, the capacity configuration efficiency of the micro-grid is poor is solved, and the technical effect of improving the capacity configuration efficiency of the micro-grid is achieved.

Description

Technical Field

[0001] This application relates to the field of microgrid capacity configuration technology, and in particular to a method and system for optimizing microgrid capacity configuration under abnormal operating conditions. Background Technology

[0002] With the large-scale integration of distributed energy resources and the development of a high proportion of renewable energy, microgrids have become an important system for improving regional energy resilience and power supply reliability. Especially under abnormal operating conditions such as islanded operation, extreme weather disturbances, and local grid instability, microgrids play a crucial role in ensuring continuous power supply to critical loads. However, abnormal operating conditions are typically characterized by suddenness, rapid evolution, incomplete information, and limited resources, significantly increasing the complexity of microgrid capacity configuration and operation scheduling. Current microgrid configuration methods mostly rely on offline planning models or static capacity matching rules based on normal operating conditions. When dealing with abnormal operating conditions, these methods often struggle to describe the multi-stage characteristics of condition evolution and cannot accurately reflect the timing constraints of equipment actions, energy deficit pressure, and control uncertainties caused by communication failures. Currently, most existing capacity configuration methods use single-layer operating condition descriptions, lacking hierarchical modeling of emergency resilience modes, degraded operation modes, and recovery / reconfiguration modes. They also fail to adequately handle the differences between short-term and continuous configurations, making it difficult to form consistent and coherent control strategies across different operating condition stages.

[0003] In summary, existing technologies suffer from problems such as limited descriptions of operating conditions, lack of explicit modeling of action sequences, absence of feasible iterative constraints in the optimization process, failure to utilize differences in equipment types, and lack of timestamp control at the execution layer. These issues prevent microgrids from achieving rapid, accurate, and executable capacity configuration under abnormal operating conditions, thereby affecting the overall system's operational resilience, scheduling continuity, and recovery capabilities. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for optimizing microgrid capacity configuration under abnormal operating conditions, in order to solve the technical problems in the prior art, such as the inability of microgrids to achieve fast, accurate and executable capacity configuration under abnormal operating conditions due to the single operating condition description, the lack of explicit modeling of action timing, the lack of feasible iterative constraints in the optimization process, the failure to utilize differences in equipment types, and the lack of timestamp control at the execution layer. This results in the overall system's operational resilience, scheduling continuity and recovery capability being affected.

[0005] In view of the above problems, this application provides a method and system for optimizing microgrid capacity configuration under abnormal operating conditions.

[0006] Firstly, this application provides a method for optimizing microgrid capacity configuration under abnormal operating conditions, implemented through a microgrid capacity configuration optimization system under abnormal operating conditions. The method includes: for a target microgrid, determining the operating condition system by decomposing abnormal operating conditions; determining the action sequence by explicitly modeling the action timing under abnormal operating conditions, wherein the decomposition criteria at least include operating condition mode and configuration dimension; based on the operating condition system and action sequence, deploying an initialization port in the grid control center to convert the uploaded abnormal operating condition configuration tasks and determine the task configuration requirements; inputting the task configuration requirements into a capacity configurator, performing column constraint optimization generation, and determining the task configuration scheme; and performing capacity configuration management and control of the target microgrid according to the task configuration scheme.

[0007] Preferably, the microgrid capacity configuration optimization method under abnormal operating conditions further includes: determining a first operating condition layer according to the operating condition mode, wherein the first operating condition layer includes at least an emergency resilience mode, a degraded operation mode, and a recovery and reconfiguration mode; determining a second operating condition layer according to the configuration dimension, wherein the second operating condition layer includes at least a short-term configuration and a continuous configuration, wherein the continuous configuration includes a survival configuration under a continuous energy deficit and a dynamic balance configuration under a continuous incomplete information state; cascading the first operating condition layer and the second operating condition layer to determine the operating condition system.

[0008] Preferably, the microgrid capacity configuration optimization method under abnormal operating conditions further includes: adopting a dynamic programming framework to discretize the multi-stage time axis of the abnormal operating conditions and determine a first planning template; introducing a second logical constraint based on the operating condition actions, wherein the operating condition actions include at least equipment start-up time, power ramp-up rate, and load switching sequence; and deploying the action sequence according to the first planning template and the second logical constraint.

[0009] Preferably, the microgrid capacity configuration optimization method under abnormal operating conditions further includes: deploying a first initialization threshold according to the operating condition system, and deploying a second initialization threshold according to the action sequence; cascading the first initialization threshold and the second initialization threshold, and performing logic training until convergence to form the initialization port and embedding it in the power grid central control.

[0010] Preferably, the microgrid capacity configuration optimization method under abnormal operating conditions further includes: performing modal matching and task parameter writing at the first operating condition layer for the abnormal operating condition configuration task according to the first initialization threshold, performing configuration matching and configuration guidance writing at the second operating condition layer to determine the first task system; importing the first task system into the second initialization threshold, performing time axis discretization processing and timing logic constraints on the first task system to generate the task configuration requirements.

[0011] Preferably, the microgrid capacity configuration optimization method under abnormal operating conditions further includes: dividing the target microgrid into fast response devices and adaptive response devices; using the fast response devices and adaptive response devices as the configuration environment, and forming a loop with the main problem module, sub-problem module, and post-processing module to build a capacity configurator.

[0012] Preferably, the microgrid capacity configuration optimization method under abnormal operating conditions further includes: generating a master problem based on the task configuration requirements, wherein the master problem initially only includes cost constraints and has an upper bound; generating a first candidate capacity configuration scheme and a first optimistic cost by solving the master problem; generating sub-problems, substituting the first candidate capacity configuration scheme into the sub-problems to determine the target value, wherein the sub-problems are operational feasibility check problems, and the sub-problems aim to minimize the violation of operational constraints; and generating the task configuration scheme by analyzing the target value and performing iterative solving.

[0013] Preferably, the microgrid capacity configuration optimization method under abnormal operating conditions further includes: if the target value is zero, marking the first candidate capacity configuration scheme as feasible and updating the upper bound; if the target value is non-zero, generating a first linear constraint and adding the first linear constraint to the main problem.

[0014] Preferably, the microgrid capacity configuration optimization method under abnormal operating conditions further includes: performing multi-threaded decomposition of the task configuration scheme based on the execution of response devices to determine multiple sub-schemes; and imposing execution timestamp constraints on the multiple sub-schemes to perform multi-threaded targeted distribution and capacity configuration management.

[0015] Secondly, this application also provides a microgrid capacity configuration optimization system under abnormal operating conditions, used to execute a microgrid capacity configuration optimization method under abnormal operating conditions as described in the first aspect, comprising: an action sequence determination module, used to determine the operating condition system by performing abnormal operating condition decomposition for a target microgrid, and to determine the action sequence by explicitly modeling the action timing under abnormal operating conditions, wherein the decomposition criteria at least include operating condition mode and configuration dimension; a task configuration requirement determination module, used to deploy an initialization port in the power grid central control based on the operating condition system and action sequence, and to convert the uploaded abnormal operating condition configuration task to determine the task configuration requirement; a task configuration scheme determination module, used to input the task configuration requirement into a capacity configurator, perform column constraint optimization generation, and determine the task configuration scheme; and a capacity configuration management module, used to perform capacity configuration management of the target microgrid according to the task configuration scheme.

[0016] The technical solution provided in this application has at least the following technical effects or advantages: by achieving the technical goal of rapidly solving and dynamically optimizing the feasible capacity under multiple working conditions, multiple devices and multiple time constraints, it achieves the technical effects of explicitly characterizing complex working conditions, making full use of the differences in equipment characteristics, ensuring that the action sequence is executable and has timestamp consistency.

[0017] The above description is merely an overview of the technical solution of this application. To enable a clearer understanding of the technical means of this application and to facilitate its implementation according to the description, and to make the above and other objects, features, and advantages of this application more apparent, specific embodiments of this application are described below. It should be understood that the content described in this section is not intended to identify key or important features of the embodiments of this application, nor is it intended to limit the scope of this application. Other features of this application will become readily apparent through the following description. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this application 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 merely exemplary. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a microgrid capacity configuration optimization method under abnormal operating conditions according to this application.

[0020] Figure 2 This is a schematic diagram of a microgrid capacity configuration optimization system under abnormal operating conditions according to this application.

[0021] Figure labeling: Action sequence determination module 1, task configuration requirement determination module 2, task configuration scheme determination module 3, capacity configuration control module 4. Detailed Implementation

[0022] This application provides a method and system for optimizing microgrid capacity configuration under abnormal operating conditions. It addresses existing technologies that suffer from problems such as limited operating condition descriptions, lack of explicit modeling of action sequences, absence of feasible iterative constraints in the optimization process, underutilization of equipment type differences, and lack of timestamp control at the execution layer. These issues prevent microgrids from achieving rapid, accurate, and executable capacity configuration under abnormal operating conditions, thus affecting the overall system's operational resilience, scheduling continuity, and recovery capabilities. The application achieves the technical goal of rapidly solving for and dynamically optimizing feasible capacity configuration under multiple operating conditions, multiple devices, and multiple time-series constraints. It achieves the technical effects of explicitly characterizing complex operating conditions, fully utilizing the differences in equipment characteristics, ensuring the executableness of action sequences, and maintaining timestamp consistency.

[0023] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. It should be understood that this application is not limited to the exemplary embodiments described herein. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application. It should also be noted that, for ease of description, only the parts related to this application are shown in the accompanying drawings, not all of them.

[0024] Example 1, please refer to the appendix. Figure 1 This application provides a microgrid capacity configuration optimization method under abnormal operating conditions, applied to a microgrid capacity configuration optimization system under abnormal operating conditions, specifically including the following steps: S1: For the target microgrid, the operating condition system is determined by decomposing abnormal operating conditions, and the action sequence is determined by explicitly modeling the action timing under abnormal operating conditions. The decomposition criteria include at least the operating condition mode and configuration dimension.

[0025] Specifically, for the target microgrid, the operating condition system is determined by decomposing abnormal operating conditions. This means first selecting a specific microgrid as the analysis object and classifying it according to its possible operating scenarios under abnormal conditions. Abnormal operating conditions can include equipment failure, external grid disconnection, or power outages caused by extreme weather. Decomposition involves classifying complex scenarios according to different impact ranges and severity to form a structured operating condition system to guide the formulation of subsequent strategies. Next, the action sequence is determined by explicitly modeling the action timing under abnormal operating conditions. This means representing the action sequence, start-up time, and response path of various devices and loads in the microgrid under abnormal conditions using a clear mathematical or logical model. The action timing covers the complete operation chain from the occurrence of the fault to system stability, enabling the optimization algorithm to handle constraints. Furthermore, the decomposition standard includes at least operating condition modes and configuration dimensions, which means that the division of the operating condition system depends on two basic dimensions. The operating condition mode is used to describe the macro state, such as emergency maintenance, survival operation, or recovery and reconfiguration, while the configuration dimension corresponds to the capacity resource configuration requirements of the microgrid under different states, such as short-term temporary configuration or long-term continuous configuration. Together, they constitute the basic framework of operating condition decomposition.

[0026] S2: Based on the operating condition system and action sequence, deploy an initialization port in the power grid central control system to convert the uploaded abnormal operating condition configuration tasks and determine the task configuration requirements.

[0027] Specifically, in the process of basing the operating condition system and action sequence on the operating condition system, by associating the operating condition system with the action sequence, the system can automatically match the corresponding action flow after identifying abnormal signals. For example, when a low voltage condition is detected, compensation instructions, node switching, and log recording are executed sequentially according to a preset order, thus laying the foundation for subsequent task configuration. Furthermore, the system can distinguish between complex operating conditions through feature parameter comparison. For instance, when load fluctuations exceed a certain threshold and the duration exceeds the digital order of seconds, a higher priority action sequence can be automatically selected.

[0028] When deploying the initialization port in the power grid central control system, the central control system refers to the core system platform used for centralized dispatch and control, while the initialization port refers to the interface module used to load basic resources, register service modules, and perform key parameter resets. The deployment process includes loading communication protocols, service channels, and security verification mechanisms to ensure that subsequent configuration tasks can be executed in a unified environment.

[0029] When converting uploaded abnormal operating condition configuration tasks, abnormal operating conditions refer to states that deviate from the normal operating thresholds of the power grid, including equipment failures, parameter drift, short-term shocks, or continuous abnormal trends. Configuration tasks are strategic adjustment instructions generated by the user or system based on these operating conditions. The task conversion process includes structured parsing of the task content, reconstruction of action requirements, and parameter completion of execution conditions. When determining task configuration requirements, these requirements refer to the resource conditions, action parameters, and execution range information necessary to ensure the task can be correctly executed in the power grid system. The process of determining these requirements includes extracting key action parameters, verifying equipment operating constraints, and determining execution priorities.

[0030] S3: Input the task configuration requirements into the capacity configurator, perform column constraint optimization to generate and determine the task configuration scheme.

[0031] Specifically, inputting task configuration requirements into the capacity configurator means sending the task configuration requirements, which have been processed by the initialization port and already contain operating condition level information, action timing constraints, energy demand characteristics, and equipment status parameters, as input data into the capacity configurator. The capacity configurator is an iterative optimization solution environment built based on the main problem, sub-problems, and post-processing modules, which can perform capacity planning under multiple equipment types, multiple time scales, and multiple constraints.

[0032] Execute column constraint optimization generation, which means that the capacity configurator uses the column constraint optimization method to solve the problem. Among them, column constraint optimization belongs to a class of decomposition optimization techniques that alternately run the master problem and the sub-problem. By continuously generating new constraints to exclude infeasible solutions and shrink the search space, the solution process gradually approaches the feasible optimal solution. After multiple rounds of master problem solving, sub-problem checking, and linear constraint supplementation, when the candidate solution meets all operating constraints and there is no further violation, the capacity configurator outputs the final task configuration plan, determines the task configuration plan, that is, the equipment capacity configuration and scheduling combination that can be achieved, has better economy, and is executable under the abnormal operating conditions of the target microgrid.

[0033] S4: According to the task configuration plan, perform capacity configuration management and control on the target microgrid.

[0034] Specifically, after obtaining the task configuration plan obtained through optimized solution, according to the capacity allocation strategy, equipment action timing, and power scheduling instructions defined therein, specific capacity management and control operations are implemented on various energy storage units, power generation units, load units, and energy routing devices inside the target microgrid. Among them, capacity configuration management and control includes the execution of the allocated capacity, the implementation of the equipment start-up and shutdown sequence, the scheduling of the power ramp-up and curtailment process, and the linkage management of various response devices, so that the actual operating state is consistent with the optimized plan. For example, when the plan requires increasing the output of the fast energy storage by 10% at the 5th minute and adjusting the output of the standby generator to 60% of the rated power at the 20th minute, the capacity configuration management and control is responsible for sending these instructions to the relevant equipment on time and monitoring their execution to ensure a stable transition of the entire microgrid under abnormal operating conditions.

[0035] Furthermore, this application also includes: determining the first operating condition layer according to the operating condition mode, where the first operating condition layer at least includes the emergency resilience mode, the degraded operation mode, and the recovery and reconstruction mode; determining the second operating condition layer according to the configuration dimension, where the second operating condition layer at least includes short-term configuration and continuous configuration, and the continuous configuration includes survival configuration under continuous energy deficit and dynamic balance configuration under continuous incomplete information state; cascading the first operating condition layer and the second operating condition layer to determine the operating condition system.

[0036] Specifically, determining the first operating condition layer based on operating condition modes refers to classifying abnormal operating situations of a microgrid according to macroscopic operating states. Operating condition modes characterize the changing trends of the overall operating capacity of the microgrid and the focus of its operating objectives. For example, the emergency resilience mode describes the emergency response state taken to maintain continuous power supply to core loads when encountering sudden shocks or critical equipment disconnection; the degraded operation mode describes the operating state where service levels must be reduced to maintain overall stability due to decreased energy supply capacity or changes in load demand; and the recovery and reconfiguration mode describes the process of gradually restoring normal operating capacity after fault mitigation and achieving structural reconfiguration through the reconfiguration of equipment and energy flow. This classification based on operating condition modes allows for the directional differentiation of abnormal situations.

[0037] Subsequently, based on the configuration dimension, the second operating condition layer is determined. This means that in addition to distinguishing different operating conditions from a macroscopic perspective, it is also necessary to further refine the operating status from the perspective of resource allocation. The configuration dimension is used to describe the capacity resource allocation method of the microgrid under different time scales and information integrity conditions. For example, short-term configuration is used for emergency situations that require a very short response time, such as load surges or local power impacts, relying on fast-response equipment for transient adjustment; continuous configuration is used for resource supply and demand rebalancing over a longer period of time. Continuous configuration further includes survival configuration under continuous energy deficit and dynamic balancing configuration under continuous incomplete information conditions. Survival configuration is used to describe the operating mode that maintains the survival power supply of critical loads under long-term energy shortage conditions, while dynamic balancing configuration is used to describe the way to maintain local and overall balance among multiple units under dynamic changes through distributed rules or local autonomous strategies when communication or monitoring information is incomplete.

[0038] Next, the first and second operating condition layers are cascaded to determine the operating condition system. This involves combining operating condition modes and configuration dimensions according to hierarchical relationships, ensuring that each macro-level operating condition mode can be matched with a corresponding configuration strategy, thus forming a complete operating condition system covering multiple abnormal situations. Through cascading, an operating condition structure with two-dimensional interpretation capabilities can be formed, enabling rapid identification of locations and invocation of corresponding optimization strategies under different modes and configuration requirements. For example, in situations where continuous rain leads to insufficient power generation, critical energy storage units unexpectedly lose 50% of their capacity, and the load experiences an unpredictable 30% pulse increase, the operating condition mode is degraded operation or even tends towards emergency resilience, while the configuration dimension corresponds to survival configuration under continuous energy deficit, thus entering this specific combined operating condition to guide resource allocation. As another example, when a partial communication network outage causes multiple control units to be disconnected, the operating condition mode may correspond to degraded operation or the initial stage of recovery and reconstruction, while the configuration dimension corresponds to dynamic balancing configuration under continuous incomplete information. In this case, an autonomous distributed collaborative strategy should be adopted to maintain overall stability.

[0039] Furthermore, this application also includes: employing a dynamic programming framework to discretize the multi-stage time axis for continuous abnormal operating conditions and determine a first planning template; introducing a second logical constraint based on operating condition actions, wherein the operating condition actions include at least equipment startup time, power ramp-up rate, and load switching sequence; and deploying the action sequence according to the first planning template and the second logical constraint.

[0040] Specifically, a dynamic programming framework is adopted to discretize the multi-stage timeline of abnormal operating conditions and determine the first planning template. This means that when constructing a control model for a microgrid under abnormal conditions, the continuous abnormal operating conditions are described as a discrete time series composed of multiple time stages, so that the state has a computable and derivable transition relationship at each stage. The dynamic programming framework is used to handle cross-stage decision dependencies, ensuring that the control decision at each stage is not only affected by the current state but also related to the target optimization results of future stages. The multi-stage timeline discretization means breaking down the entire abnormal process into multiple analyzable time segments according to fixed time steps or event triggering methods, giving the control strategy a clear structure in the time dimension. The first planning template is thus constructed, serving as the basic structure for describing action planning under abnormal operating conditions, and guiding the deployment and constraint solving of subsequent action logic.

[0041] Next, a second logical constraint based on operating condition actions is introduced. These actions include at least equipment startup time, power ramp-up rate, and load switching sequence. This supplements the first planning template with constraints reflecting physical operating characteristics, ensuring the planning model not only has a discrete structure in the time dimension but also conforms to the actual operating patterns of equipment and loads. Specifically, equipment startup time represents the minimum startup time required for different types of equipment to transition from a shutdown state to an operating state; for example, a certain energy storage inverter may require several seconds to fully engage power. Power ramp-up rate describes the maximum change in output per unit time to prevent equipment damage or system instability due to excessively rapid changes. Load switching sequence describes the order in which different loads are prioritized for activation or deactivation under abnormal conditions; for example, prioritizing critical loads such as hospitals while delaying the restoration of general residential loads. Through these logical constraints, the planning model ensures that the action sequence meets safety and feasibility requirements.

[0042] Finally, based on the first planning template and the second logical constraints, the action sequence is deployed. This involves combining the discrete time structure with action logic rules, solving for or constructing an executable time-series action plan, so that equipment startup, power adjustment, and load operation can be executed in the correct order and under constraints at each discrete time stage, thereby generating a complete abnormal operating condition response sequence. For example, when affected by a sudden load increase in the first stage, the action sequence may arrange for rapid-response energy storage to increase output by limiting the ramp rate; in the second stage, a backup generator may be started to stabilize power supply; as time progresses, the actions in each stage strictly follow the planning template and logical constraints, enabling the entire sequence to complete the transition from emergency to stability.

[0043] Furthermore, this application also includes: deploying a first initialization threshold according to the operating condition system, and deploying a second initialization threshold according to the action sequence; cascading the first initialization threshold and the second initialization threshold, and performing logic training until convergence to form the initialization port and embedding it in the power grid central control.

[0044] Specifically, deploying the first initialization threshold according to the operating condition system refers to transforming the triggering conditions, classification boundaries, and operational criteria corresponding to various operating condition modes and configuration dimensions in the operating condition system into a set of quantifiable and calculable judgment thresholds after the operating condition system is constructed. This is used to quickly identify the current operating condition category during operation. The operating condition system contains a two-layer structure composed of operating condition modes and configuration dimensions. Therefore, the first initialization threshold is used to realize the real-time discrimination of the two-layer structure. For example, in the emergency resilience mode, when the critical load supply capacity decreases by more than a certain set proportion, or when the remaining energy storage capacity is lower than a certain threshold under the continuous energy deficit state, the identification logic of the corresponding operating condition can be triggered.

[0045] Subsequently, a second initialization threshold is deployed based on the action sequence. This refers to parameterizing the action triggering conditions, time constraints, and action feasibility boundaries involved in the sequence after the action sequence is established, forming another type of judgment threshold. This threshold is used to constrain the logical legality when performing startup, switching, or adjustment operations. For example, if the startup time of a device is required to be no less than a certain number of seconds, a minimum time threshold for the startup action can be set; if the power ramp-up rate is limited, a maximum boundary for power change per unit time can be set; if the load switching sequence must be strictly followed, a threshold can be used to ensure that the sequence is not violated. The second initialization threshold mainly serves the constraint check during the action execution phase, ensuring the executability of the action sequence in actual operation.

[0046] Next, the cascaded first and second initialization thresholds, through logical training until convergence, form an initialization port and are embedded in the power grid central control system. This means integrating the two threshold sets according to logical relationships, enabling the operating condition identification logic and action execution logic to operate jointly, thereby forming an initialization port with adaptive discrimination and action verification capabilities. Logical training until convergence refers to optimizing the threshold combination through iterative learning based on actual operating data, simulation experimental data, or expert rules, so that the port can output stable and consistent initialization judgment results under various operating conditions. For example, when continuous rain leads to a complex situation of energy deficit and energy storage withdrawal, the first initialization threshold identification system is in a continuous deficit mode; simultaneously, the second initialization threshold verifies whether the standby equipment startup time in the action sequence meets the requirements.

[0047] Furthermore, this application also includes: performing modal matching and task parameter writing at the first working condition layer on the abnormal working condition configuration task according to the first initialization threshold, performing configuration matching and configuration guidance writing at the second working condition layer to determine the first task system; importing the first task system into the second initialization threshold, performing time axis discretization processing and timing logic constraints on the first task system to generate the task configuration requirements.

[0048] Specifically, based on the first initialization threshold, the modal matching and task parameter writing for abnormal operating condition configuration tasks at the first operating condition layer refers to, upon receiving an abnormal operating condition configuration task, using the operating condition identification threshold included in the first initialization threshold to perform a matching and judgment at the operating condition mode level for the abnormal situation described by the task, thereby determining whether the task corresponds to an emergency resilience mode, a degraded operation mode, or a recovery and reconstruction mode. Modal matching compares indicators such as load changes, energy capacity degradation, and equipment failure characteristics presented in the task with the mode boundaries set at the operating condition layer to determine the system's response direction. Task parameter writing, after identifying the corresponding mode, involves writing key indicators involved in the task, such as the energy deficit ratio, the type of affected load, and the equipment withdrawal capacity, into the task structure of the corresponding mode, giving the structure complete operating condition input information. This enables the task to be correctly classified according to the operating condition mode and to record necessary parameters.

[0049] Next, the second-level configuration matching and configuration-guided writing involves further refining the configuration-level matching of tasks after completing the pattern-level matching, based on the judgment thresholds for configuration dimensions in the first initialization threshold. This determines the specific configuration category of the task, such as short-term configuration, survival configuration under continuous energy deficit, or dynamic balancing configuration under continuous incomplete information. Configuration matching is used to judge based on the task's duration, the completeness of available information, and resource changes. Configuration-guided writing involves writing the policy direction required by the matched configuration category into the task structure, including priority load protection strategies, call weights for fast-response devices, and self-coordination rules for autonomous units. This gives the task structure a clear direction for subsequent optimization, ultimately forming the first task system. The first task system is a structured description of tasks obtained by integrating pattern recognition results and configuration dimension interpretation.

[0050] Subsequently, the first task system is imported into the second initialization threshold. Time-axis discretization and temporal logic constraints are then applied to the first task system to generate task configuration requirements. This involves inputting the established first task system into the second initialization threshold, subjecting the task system to constraint verification related to action timing. Time-axis discretization involves splitting the task content into different time stages according to the time segmentation rules of the action sequence, enabling the task to clearly define the actions and resource status to be executed in each time period. Temporal logic constraints are used to check the feasibility of the split actions, such as whether the equipment startup time matches the sequence, whether power ramp-up meets the limits, and whether load switching conforms to the sequence rules. This transforms the original task into task configuration requirements that can be directly invoked by the capacity configurator. In other words, the task possesses a time structure, action structure, and logical integrity, and can support subsequent column constraint optimization solutions.

[0051] Furthermore, this application also includes: dividing the target microgrid into fast response devices and adaptive response devices; using the fast response devices and adaptive response devices as the configuration environment, and forming a loop with a main problem module, a sub-problem module, and a post-processing module to build a capacity configurator.

[0052] Specifically, when classifying fast-response and adaptive-response devices within a target microgrid, the target microgrid refers to a distributed power system with autonomous control, power load coordination, and energy management capabilities, operating on a local distribution area basis. Fast-response devices are electrical equipment capable of adjusting power output within seconds to minutes to cope with frequency fluctuations, voltage disturbances, or sudden load changes. For example, when a sudden load increase is detected, a supercapacitor can release energy within seconds, a high-speed flywheel can provide transient inertial support, and a converter with inertial response can quickly adjust its output power through control algorithms. Adaptive-response devices, on the other hand, primarily provide stable energy supply or buffer energy gaps within hours to days. For example, long-term energy storage can continuously replenish the system's energy for several hours, and standby generators can start outputting power when the grid enters a prolonged period of low energy levels. By classifying these two types of equipment, a clear functional division of labor can be established for the microgrid's short-term and medium-to-long-term energy regulation.

[0053] When using fast-response and adaptive-response devices as the configuration environment, and building a capacity configurator with a main problem module, sub-problem modules, and post-processing modules forming a loop, the configuration environment typically refers to the set of device capacity boundaries, response characteristics, scheduling constraints, and operating strategies upon which the capacity configuration calculation depends. In this environment, the main problem module models the overall capacity planning, including load forecasting, device capacity boundary constraints, and multi-timescale coordination strategies. The sub-problem modules solve the sub-objectives derived from the decomposition of the main problem, such as performing second-level power regulation optimization for fast-response devices and hour-level energy storage scheduling optimization for adaptive-response devices. The post-processing module performs consistency checks, parameter corrections, and strategy fusion between the results of the two modules to ensure that the final configuration result meets the system's global feasibility and operational safety requirements. For example, when the main problem module suggests increasing the fast-response capacity by a digital order of magnitude, the sub-problem modules will further evaluate whether this capacity matches the hourly energy supply strategy of the energy storage devices; if not, the post-processing module will make corrections. By constructing a cyclic loop, the capacity allocator can iterate between solving the main problem and solving the subproblems, so that the final capacity allocation result achieves a balance between local optimum and overall feasibility at different time scales.

[0054] Furthermore, this application also includes: generating a master problem based on the task configuration requirements, wherein the master problem initially only contains cost constraints and has an upper bound; generating a first candidate capacity configuration scheme and a first optimistic cost by solving the master problem; generating a sub-problem, substituting the first candidate capacity configuration scheme into the sub-problem to determine a target value, wherein the sub-problem is an operational feasibility check problem, and the sub-problem aims to minimize the violation of operational constraints; and generating the task configuration scheme by analyzing the target value and performing iterative solving.

[0055] Specifically, when generating the master problem based on task configuration requirements, the master problem is a mathematical programming model that serves as the global decision-making layer in the optimization framework. Initially, it only contains cost constraints, indicating that in the planning phase, the main optimization objectives are system investment cost, operating cost, or scheduling cost. By setting an upper bound, the overall cost is limited to a certain maximum allowable value, thereby ensuring that the planning results meet the feasible range in terms of economics.

[0056] When generating the first candidate capacity configuration scheme and the first optimistic cost by solving the main problem, the solution process uses a linear programming or mixed-integer programming solver. The resulting first candidate capacity configuration scheme is the optimal capacity combination that satisfies the initial cost constraints, including the capacity allocation ratio, power level, and operating mode of fast-response and adaptive-response equipment. The first optimistic cost represents the lowest possible cost value under the current planning conditions without considering operational constraints. For example, when the planning model relies solely on cost constraints, the solution may tend to reduce the investment in high-cost equipment to obtain a lower optimistic cost in the tens of thousands of yuan.

[0057] When generating subproblems and substituting the first candidate capacity configuration scheme into the subproblems to determine the target value, the subproblems serve as operational feasibility checks. Their core function is to verify whether the candidate capacity configuration scheme can meet various operational constraints during actual operation. These constraints include power balance constraints, equipment start-up and shutdown constraints, response time constraints, and network security constraints. The subproblems aim to minimize the violation of operational constraints. Violations are typically represented by penalty variables. When a candidate scheme fails to meet the operational conditions, the penalty variable gradually increases. For example, when the second-level output of a fast-response device is insufficient to cope with a sudden load surge, the violation may increase by several units, indicating the degree of infeasibility of the candidate scheme.

[0058] When generating task configuration schemes by analyzing target values ​​and performing iterative solutions, the target value reflects the feasibility level of candidate schemes. When the target value is greater than zero, it indicates that there are still operational constraints that are violated, and the capacity configuration needs to be adjusted through iterative solutions between the main problem and sub-problems. For example, when the violation is in numerical units, it can be reduced by increasing the rapid response capacity or improving the adaptive power supply capability. As the number of iterations increases from several to dozens, the target value gradually converges to zero or close to zero, meaning that the operational constraints are fully satisfied. The task configuration scheme obtained at this point is the final capacity configuration result that combines cost feasibility and operational feasibility.

[0059] Furthermore, this application also includes: if the target value is zero, marking the first candidate capacity configuration scheme as feasible and updating the upper bound; if the target value is non-zero, generating a first linear constraint and adding the first linear constraint to the main problem.

[0060] Specifically, if the objective value is zero, the first candidate capacity configuration scheme is marked as feasible, and the upper bound is updated. This means that after substituting the first candidate capacity configuration scheme into the operational feasibility check subproblem, if the objective function value of the subproblem is zero, it indicates that the candidate scheme has no violations under all considered operational constraints. Therefore, the candidate scheme can be identified as an operationally feasible solution. The objective value is used to characterize the aggregate index of violations, the feasible mark is used to distinguish between directly usable configuration schemes and schemes that need to be modified in the subsequent solution process, and the upper bound is the upper limit parameter used to limit cost or other indicators in the main problem. Updating the upper bound means adjusting the upper limit to reflect the new achievable level when a better or more realistic feasible scheme is found. For example, when the initial cost upper bound is 100,000 yuan and the first candidate scheme achieves a cost of 80,000 yuan under the operational constraints, the upper bound can be reduced to 80,000 yuan to narrow the subsequent search space, thereby guiding the main problem to converge in a more economical and feasible direction.

[0061] Furthermore, if the objective value is non-zero, a first linear constraint is generated and added to the main problem. This first linear constraint indicates that when the objective value returned by the subproblem is greater than zero, the candidate solution violates the operational constraints. Additional linear constraints need to be introduced into the main problem to eliminate solution space directions that lead to infeasibility, thereby improving the planning results. The first linear constraint is often constructed based on a feasibility cut obtained from the subproblem analysis or a linear inequality based on dual variables. This constraint is used to indicate an unacceptable operational defect in the current configuration direction. For example, if a certain configuration prevents a fast-response device from providing at least 20% of its rated power instantaneously under second-level impacts, then a linear constraint can eliminate all candidate solutions with similar characteristics in that direction. The process of generating the first linear constraint involves identifying key variables or combinations from the subproblems that lead to violations, and then expressing them in linear form as additional constraints to the main problem. This allows the main problem to avoid known infeasible regions in subsequent solutions, thereby gradually approaching an operationally feasible and cost-acceptable solution.

[0062] Furthermore, this application also includes: performing multi-threaded decomposition of the task configuration scheme based on the execution of the response device to determine multiple sub-schemes; and imposing execution timestamp constraints on the multiple sub-schemes to perform multi-threaded targeted distribution and capacity configuration management.

[0063] Specifically, multi-threaded decomposition of the task configuration scheme based on the execution of response devices refers to dividing the overall task configuration scheme into several independent execution sub-schemes according to the physical characteristics and operational responsibilities of different types of response devices after obtaining the overall scheme. Response devices include fast-response devices and adaptive-response devices. Fast-response devices are used for disturbance suppression at the second to minute level, such as supercapacitors, high-speed flywheels, or converters with inertial injection capabilities. Adaptive-response devices are used for power compensation and energy supply at the hour to day level, such as long-term energy storage or backup generators. Multi-threaded decomposition means that the computational task is broken down into parallel components according to device type, runtime sequence, or control path, so that each type of device receives a locally scheduled sub-task matching its capabilities, thereby improving deployment efficiency and execution reliability. For example, when it is necessary to quickly release 20% of the energy storage power and simultaneously adjust the discharge strategy of long-term energy storage at the same time, the multi-threaded mechanism allows these two operations to be performed in parallel without blocking each other.

[0064] Applying execution timestamp constraints to multiple sub-schemes means adding explicit timestamps to each sub-scheme to ensure that its start time, response time, and end time meet the overall action sequence requirements of the task configuration scheme. Timestamp constraints are used to avoid execution conflicts between different devices; for example, the second-level start-up of a fast-response device must precede the hour-level reconfiguration of an adaptive device. Subsequently, multi-threaded targeted deployment and capacity configuration management refer to sending the timestamped sub-schemes, according to device type or logical affiliation, to the corresponding device control unit or control node through multi-threading. These nodes then execute specific operations according to the predetermined sequence, capacity allocation parameters, and action logic, thereby achieving actual configuration control of the microgrid capacity. For example, if an action requires rapid energy storage compensation to be completed at the 10th minute and the backup generator to be started at the 30th minute, the timestamp constraint ensures that the two actions are executed sequentially, while multi-threaded deployment ensures that they can run in parallel in different control loops, avoiding delays that could lead to system instability.

[0065] In summary, the microgrid capacity configuration optimization method under abnormal operating conditions provided in this application has the following technical effects: by achieving the technical goal of rapidly solving and dynamically optimizing the feasible capacity under multiple operating conditions, multiple devices, and multiple time-series constraints, it achieves the technical effects of explicitly characterizing complex operating conditions, making full use of the differences in device characteristics, ensuring that the action sequence is executable and has timestamp consistency.

[0066] Example 2: Based on the same inventive concept as the microgrid capacity configuration optimization method under abnormal operating conditions in the foregoing examples, this application also provides a microgrid capacity configuration optimization system under abnormal operating conditions. Please refer to the appendix. Figure 2The system includes: an action sequence determination module 1, used to determine the operating condition system for a target microgrid by decomposing abnormal operating conditions, and to determine the action sequence by explicitly modeling the action timing under abnormal operating conditions, wherein the decomposition criteria include at least the operating condition mode and configuration dimension; a task configuration requirement determination module 2, used to deploy an initialization port in the power grid central control based on the operating condition system and action sequence, and to convert the uploaded abnormal operating condition configuration tasks to determine the task configuration requirements; a task configuration scheme determination module 3, used to input the task configuration requirements into the capacity configurator, perform column constraint optimization generation, and determine the task configuration scheme; and a capacity configuration management module 4, used to perform capacity configuration management on the target microgrid according to the task configuration scheme.

[0067] Furthermore, the microgrid capacity configuration optimization system under abnormal operating conditions is also used for: determining a first operating condition layer according to the operating condition mode, wherein the first operating condition layer includes at least an emergency resilience mode, a degraded operation mode, and a recovery and reconfiguration mode; determining a second operating condition layer according to the configuration dimension, wherein the second operating condition layer includes at least a short-term configuration and a continuous configuration, wherein the continuous configuration includes a survival configuration under a continuous energy deficit and a dynamic balance configuration under a continuous incomplete information state; and cascading the first operating condition layer and the second operating condition layer to determine the operating condition system.

[0068] Furthermore, the microgrid capacity configuration optimization system under abnormal operating conditions is also used to: adopt a dynamic programming framework to discretize the multi-stage time axis of the abnormal operating conditions and determine a first planning template; introduce a second logical constraint based on the operating condition actions, wherein the operating condition actions include at least equipment start-up time, power ramp-up rate, and load switching sequence; and deploy the action sequence according to the first planning template and the second logical constraint.

[0069] Furthermore, the microgrid capacity configuration optimization system under abnormal operating conditions is also used to: deploy a first initialization threshold according to the operating condition system, and deploy a second initialization threshold according to the action sequence; cascade the first initialization threshold and the second initialization threshold, and perform logic training until convergence to form the initialization port and embed it in the power grid central control.

[0070] Furthermore, the microgrid capacity configuration optimization system under abnormal operating conditions is also used to: perform modal matching and task parameter writing at the first operating condition layer for the abnormal operating condition configuration task according to the first initialization threshold, perform configuration matching and configuration guidance writing at the second operating condition layer, and determine the first task system; import the first task system into the second initialization threshold, perform time axis discretization processing and timing logic constraints on the first task system, and generate the task configuration requirements.

[0071] Furthermore, the microgrid capacity configuration optimization system under abnormal operating conditions is also used to: divide the target microgrid into fast response devices and adaptive response devices; use the fast response devices and adaptive response devices as the configuration environment, and construct a capacity configurator by forming a loop with a main problem module, a sub-problem module, and a post-processing module.

[0072] Furthermore, the microgrid capacity configuration optimization system under abnormal operating conditions is also used for: generating a master problem based on the task configuration requirements, wherein the master problem initially only includes cost constraints and has an upper bound; generating a first candidate capacity configuration scheme and a first optimistic cost by solving the master problem; generating a sub-problem, substituting the first candidate capacity configuration scheme into the sub-problem to determine the target value, wherein the sub-problem is an operational feasibility check problem, and the sub-problem aims to minimize the violation of operational constraints; and generating the task configuration scheme by analyzing the target value and performing iterative solving.

[0073] Furthermore, the microgrid capacity configuration optimization system under abnormal operating conditions is also used for: if the target value is zero, marking the first candidate capacity configuration scheme as feasible and updating the upper bound; if the target value is non-zero, generating a first linear constraint and adding the first linear constraint to the main problem.

[0074] Furthermore, the microgrid capacity configuration optimization system under abnormal operating conditions is also used to: decompose the task configuration scheme into multiple sub-schemes based on the execution of response devices, and determine multiple sub-schemes; constrain the execution timestamps of the multiple sub-schemes, and perform multi-threaded targeted distribution and capacity configuration management.

[0075] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The microgrid capacity configuration optimization method and specific example under abnormal operating conditions in the aforementioned Embodiment 1 are also applicable to the microgrid capacity configuration optimization system under abnormal operating conditions in this embodiment. Through the foregoing detailed description of the microgrid capacity configuration optimization method under abnormal operating conditions, those skilled in the art can clearly understand the microgrid capacity configuration optimization system under abnormal operating conditions in this embodiment. Therefore, for the sake of brevity, it will not be described in detail here.

[0076] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0077] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for optimizing microgrid capacity configuration under abnormal operating conditions, characterized in that, The method includes: For the target microgrid, the operating condition system is determined by decomposing abnormal operating conditions, and the action sequence is determined by explicitly modeling the action timing under abnormal operating conditions. The decomposition criteria include at least the operating condition mode and configuration dimension. Based on the aforementioned operating condition system and action sequence, an initialization port is deployed in the power grid central control unit to convert the uploaded abnormal operating condition configuration tasks and determine the task configuration requirements; Input the task configuration requirements into the capacity configurator, perform column constraint optimization to generate and determine the task configuration scheme; According to the task configuration scheme, capacity configuration and management are performed on the target microgrid.

2. The microgrid capacity configuration optimization method under abnormal operating conditions as described in claim 1, characterized in that, Based on the operating mode, a first operating mode layer is determined, wherein the first operating mode layer includes at least an emergency resilience mode, a degraded operation mode, and a recovery and reconfiguration mode; Based on the configuration dimension, a second operating condition layer is determined, wherein the second operating condition layer includes at least short-term configuration and continuous configuration, and the continuous configuration includes survival configuration under continuous energy deficit and dynamic balancing configuration under continuous incomplete information state; The first operating condition layer and the second operating condition layer are cascaded to determine the operating condition system.

3. The microgrid capacity configuration optimization method under abnormal operating conditions as described in claim 1, characterized in that, Action sequences are determined by explicitly modeling the timing of actions under abnormal operating conditions, including: A dynamic programming framework is adopted to discretize the multi-stage time axis for abnormal operating conditions and determine the first planning template. A second logical constraint based on operating condition actions is introduced, wherein the operating condition actions include at least equipment startup time, power ramp-up rate, and load switching sequence; The action sequence is deployed based on the first planning template and the second logical constraint.

4. The microgrid capacity configuration optimization method under abnormal operating conditions as described in claim 1, characterized in that, Deploy initialization ports in the power grid central control system, including: Deploy a first initialization threshold according to the operating condition system, and deploy a second initialization threshold according to the action sequence; The first initialization threshold and the second initialization threshold are cascaded and trained to convergence to form the initialization port, which is then embedded in the power grid central control system.

5. The microgrid capacity configuration optimization method under abnormal operating conditions as described in claim 4, characterized in that, Convert the uploaded abnormal operating condition configuration tasks to determine the task configuration requirements, including: Based on the first initialization threshold, perform modal matching and task parameter writing at the first working condition layer for the abnormal working condition configuration task, perform configuration matching and configuration guidance writing at the second working condition layer, and determine the first task system. The first task system is imported into the second initialization threshold, and the first task system is subjected to time axis discretization and timing logic constraints to generate the task configuration requirements.

6. The microgrid capacity configuration optimization method under abnormal operating conditions as described in claim 1, characterized in that, Before inputting the task configuration requirements into the capacity configurator, the construction of the capacity configurator includes: Divide the target microgrid into fast response devices and adaptive response devices; Using the aforementioned rapid response device and adaptive response device as the configuration environment, a capacity configurator is built by forming a loop with the main problem module, sub-problem module, and post-processing module.

7. The microgrid capacity configuration optimization method under abnormal operating conditions as described in claim 6, characterized in that, The execution column constraint optimization generation determines the task configuration scheme, including: Based on the task configuration requirements, a main problem is generated, wherein the main problem initially contains only cost constraints and an upper bound is set; By solving the main problem, a first candidate capacity configuration scheme and a first optimistic cost are generated. Generate a subproblem, substitute the first candidate capacity configuration scheme into the subproblem, and determine the target value, wherein the subproblem is an operational feasibility check problem, and the subproblem aims to minimize the amount of violation of operational constraints; By analyzing the target value and performing iterative solving, the task configuration scheme is generated.

8. The microgrid capacity configuration optimization method under abnormal operating conditions as described in claim 7, characterized in that, If the target value is zero, mark the first candidate capacity configuration scheme as feasible and update the upper bound; If the target value is non-zero, generate a first linear constraint and add the first linear constraint to the main problem.

9. The microgrid capacity configuration optimization method under abnormal operating conditions as described in claim 1, characterized in that, Capacity configuration management of the target microgrid includes: The task configuration scheme is decomposed into multiple sub-schemes based on the execution of the response device. The execution timestamp constraints are applied to the multiple sub-schemes to perform multi-threaded targeted distribution and capacity configuration management.

10. A microgrid capacity configuration optimization system under abnormal operating conditions, characterized in that, The steps for implementing the microgrid capacity configuration optimization method under abnormal operating conditions as described in any one of claims 1 to 9 include: The action sequence determination module is used to determine the operating condition system for a target microgrid by decomposing abnormal operating conditions and to determine the action sequence by explicitly modeling the action timing under abnormal operating conditions. The decomposition criteria include at least the operating condition mode and configuration dimension. The task configuration requirement determination module is used to determine the task configuration requirements by deploying an initialization port in the power grid central control based on the operating condition system and action sequence, converting the uploaded abnormal operating condition configuration tasks. The task configuration scheme determination module is used to input the task configuration requirements into the capacity configurator, perform column constraint optimization generation, and determine the task configuration scheme. The capacity configuration and control module is used to perform capacity configuration and control on the target microgrid according to the task configuration scheme.

Images