A power grid dispatching instruction adjusting method and system

Abstract

This invention discloses a method and system for adjusting power grid dispatching instructions, belonging to the field of power system dispatching security technology. The method comprises: obtaining the original dispatching instructions, the initial target feasible region version, and the initial target feasible region constraints based on the power grid to be dispatched; correcting the original dispatching instructions based on the above data to obtain corrected dispatching instructions; performing a binary search process based on the above data to obtain a first boundary instruction, a second boundary instruction, and a binary search index; if the index does not meet the preset conditions, the binary search process is re-executed based on the above data until the preset conditions are met, and the second boundary instruction is determined as the optimized dispatching instruction; and correcting the optimized dispatching instruction based on the above data to obtain the corrected dispatching instruction. The power grid dispatching instruction adjustment method provided by this invention can complete security verification and instruction correction before the dispatching instructions are issued, ensuring that the dispatching instructions are securely executable and that the dispatching process is continuous and uninterrupted.

Description

Technical Field

[0001] This invention relates to the field of power system dispatch safety technology, and in particular to a method and system for adjusting power grid dispatch instructions. Background Technology

[0002] With the continuous growth of installed capacity of new energy sources such as wind power and photovoltaics, the power grid operation is more frequently affected by various factors such as fluctuations in new energy output, load changes, and adjustments to the grid structure. In daily operation, the dispatch center needs to issue dispatch instructions to various resources such as generating units, energy storage, and flexible loads. This process usually needs to meet various operational constraints such as power flow solvability, section and voltage limits, and resource capacity boundaries to ensure the safe and stable operation of the power grid.

[0003] Under the current technological background, the safety verification and adjustment of dispatching instructions typically rely on power flow calculations, constraint checks, and manual experience corrections. However, when changes occur in the grid structure, cross-sectional limits are adjusted, key equipment is put into operation or decommissioned, or measurement and forecast information fluctuates, the feasibility and safety margin of issued or planned instructions may need to be reassessed and corrected. Because dispatching instructions involve multiple constraint dimensions and strong coupling relationships between them, and because grid operation places high demands on the timeliness of instruction verification and adjustment, how to achieve rapid feasibility recovery and safety correction before instruction issuance, and how to update constraint requirements and safety margins in a timely manner when operating conditions change, while simultaneously generating correction instructions that are easy to review and trace, are critical issues that urgently need to be addressed in the field of renewable energy dispatching. Summary of the Invention

[0004] The present invention aims to provide a method and system for adjusting power grid dispatching instructions to solve the above-mentioned technical problems, avoid frequent correction and readjustment of dispatching instructions, and complete security verification and instruction correction before dispatching instructions are issued, so as to ensure that dispatching instructions are safe and executable and that the dispatching process is continuous and uninterrupted.

[0005] To address the aforementioned technical problems, this invention provides a method for adjusting power grid dispatching commands, comprising: The original dispatching instructions and the initial target feasible region version are obtained based on the power grid to be dispatched, and the initial target feasible region constraints are constructed based on the initial target feasible region version. The original scheduling instructions are corrected based on the initial target feasible region constraints and the preset scheduling instruction correction model to obtain the corrected scheduling instructions; A binary search process is executed based on the original scheduling instruction and the corrected scheduling instruction to obtain a first boundary instruction, a second boundary instruction, and a binary search index. If the binary search index does not meet the preset binary search termination condition, a first update instruction and a second update instruction are obtained based on the first boundary instruction and the second boundary instruction. The first update instruction and the second update instruction are used as the first boundary instruction and the second boundary instruction, respectively, and the binary search process is re-executed until the binary search index meets the preset binary search termination condition. The second boundary instruction is then determined as the optimized scheduling instruction. The system acquires real-time power grid operation status data, and corrects the optimized scheduling instructions based on the real-time power grid operation status data, corrected scheduling instructions, initial target feasible region constraints, and preset scheduling instruction correction model, thereby obtaining corrected scheduling instructions.

[0006] In the above scheme, by constructing initial target feasible region constraints, a safety verification benchmark can be established for dispatch instructions, avoiding unfounded corrections and readjustments. Next, the original dispatch instructions are corrected using the initial target feasible region constraints and a preset correction model to obtain corrected dispatch instructions. This allows for safety verification and correction before instruction issuance, preventing readjustments after issuance. Then, by outputting optimized dispatch instructions when the binary search index meets the preset binary search termination condition, the number of optimizations and optimization time for the corrected dispatch instructions can be controlled, preventing dispatch interruptions due to repeated recalculations, while ensuring the output optimized dispatch instructions are safe and executable. Finally, the optimized dispatch instructions can be further corrected using real-time grid operation status data, corrected dispatch instructions, initial target feasible region constraints, and the preset dispatch instruction correction model, ensuring that the obtained corrected dispatch instructions continuously meet the constraints and guaranteeing uninterrupted dispatching.

[0007] Furthermore, the step of obtaining the original dispatching instruction and the initial target feasible region version based on the power grid to be dispatched, and constructing the initial target feasible region constraints based on the initial target feasible region version, includes: Based on the grid to be dispatched, the original dispatch instructions, the original feasible domain version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point ratio coefficients, and grid topology events are obtained. The initial target feasible region version is determined based on the original feasible region version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point proportion coefficients, and grid topology events. Initial target feasible region constraints are then constructed based on the initial target feasible region version.

[0008] In the above scheme, by obtaining the original dispatch instructions, the original feasible region version, and various data of the grid to be dispatched, a complete basis for judgment can be provided for subsequently determining the initial target feasible region version. Then, the initial target feasible region version that best suits the grid's conditions is automatically determined using the various data of the grid to be dispatched. This satisfies adaptive constraint control while balancing safety and economy. Finally, by constructing initial target feasible region constraints based on the initial target feasible region version, a constraint basis is provided for subsequent dispatch instruction correction, binary search, and dispatch instruction modification, ensuring that the entire adjustment process is carried out under the correct constraint criteria.

[0009] Furthermore, the step of determining the initial target feasible region version based on the original feasible region version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly proportion coefficients, and grid topology events, and constructing initial target feasible region constraints based on the initial target feasible region version, includes: Operational risk indicators are obtained based on several key cross-sectional margins, several key voltage margins, several key resource margins, and preset operational risk selection conditions. The risk triggering feasible domain version is determined based on the aforementioned operational risk indicators, the preset first operational risk threshold, and the preset second operational risk threshold. Information quality level indicators are constructed based on power grid state estimation residuals, key measurement missing rates, and power grid anomaly proportion coefficients. The quality triggering feasible domain version is determined based on the information quality level index, the preset first information quality level threshold, and the preset second information quality level threshold. The topology triggering feasible domain version is determined based on power grid topology events and a preset power grid topology event table; The initial target feasible domain version is determined based on the original feasible domain version, the risk-triggered feasible domain version, the quality-triggered feasible domain version, and the topology-triggered feasible domain version; Construct initial target feasible region constraints based on the initial target feasible region version.

[0010] In the above scheme, operational risk indicators are calculated using critical section margins, critical voltage margins, critical resource margins, and preset risk conditions. This directly reflects the current security tension of the grid to be dispatched, allowing for the determination of whether stronger constraints are needed. Next, by comparing the operational risk indicators with preset first and second operational risk thresholds, the degree of grid security risk can be determined, and a risk triggering feasible domain version is automatically matched according to the degree of grid security risk. Then, an information quality level indicator is constructed using state estimation residuals, critical measurement missing rates, and grid anomaly point proportion coefficients. This indicator accurately determines whether the grid is reliable and whether there are significant errors. The information quality level indicator is then compared with preset first and second information quality level thresholds to determine the quality triggering feasible domain version. Finally, the topology triggering feasible domain version is determined using grid topology events and a preset grid topology event table, enabling the identification of the required constraint version when the grid structure changes or equipment operates. Finally, by integrating the original feasible region version, the risk-triggered feasible region version, the quality-triggered feasible region version, and the topology-triggered feasible region version, the final initial target feasible region version is determined. This allows for the selection of the safest and most suitable constraint criteria for the current power grid conditions, preventing operational risks to the power grid. Then, based on the initial target feasible region version, initial target feasible region constraints are constructed, providing a reliable constraint basis for the correction of subsequent scheduling instructions.

[0011] Furthermore, in the process of correcting the original scheduling instructions based on the initial target feasible region constraints and the preset scheduling instruction correction model to obtain the corrected scheduling instructions, the construction process of the preset scheduling instruction correction model includes: Obtain several historical initial scheduling instructions, several historical actual scheduling instructions corresponding to several historical initial scheduling instructions, and several historical actual feasible domain versions corresponding to several historical initial scheduling instructions. Construct several historical feasible domain constraints based on several historical feasible domain versions; Based on several historical initial scheduling instructions, several historical actual feasible domain constraints and a preset initial scheduling instruction correction model, several first scheduling instructions corresponding to several historical initial scheduling instructions are obtained. A loss function is constructed based on the historical initial scheduling instruction and the first scheduling instruction corresponding to the historical initial scheduling instruction. The preset initial scheduling instruction correction model is trained by minimizing the loss function until the preset stopping condition is met, and the preset scheduling instruction correction model is obtained.

[0012] In the above scheme, by acquiring several historical initial scheduling instructions, several historical actual scheduling instructions corresponding to these historical initial scheduling instructions, and several historical actual feasible region versions corresponding to these historical initial scheduling instructions, a data foundation can be provided for training the preset initial scheduling instruction correction model. Next, historical actual feasible region constraints are constructed using the historical actual feasible region versions, providing accurate constraint basis for model training according to historical real constraints. Then, several first scheduling instructions corresponding to the historical initial scheduling instructions are obtained using the historical initial scheduling instructions, historical actual feasible region constraints, and the preset initial scheduling instruction correction model, enabling preliminary correction of the historical initial scheduling instructions. Subsequently, a loss function is constructed using the historical initial scheduling instructions and their corresponding first scheduling instructions, allowing the calculation of the deviation between the model output and the actual output, providing direction for model optimization. Finally, the preset initial scheduling instruction correction model is trained with the goal of minimizing the loss function, without optimizing the model until a preset stopping condition is met, ultimately obtaining an accurate preset scheduling instruction correction model.

[0013] Further, the step of performing a binary search process based on the original scheduling instruction and the corrected scheduling instruction to obtain the first boundary instruction, the second boundary instruction, and the binary search index includes: Midpoint instructions are generated based on the original scheduling instructions and the corrected scheduling instructions. Based on preset power grid constraints, the feasibility of the midpoint command is determined, and the feasibility identifier of the midpoint command is obtained. The first boundary instruction, the second boundary instruction, and the binary search index are obtained based on the midpoint instruction, the midpoint instruction feasibility identifier, and the preset instruction feasibility conditions.

[0014] In the above scheme, a midpoint instruction is generated by combining the original scheduling instruction and the corrected scheduling instruction. This yields an intermediate instruction between the two instructions, providing a basis for subsequent feasibility assessment. Next, the feasibility of the midpoint instruction is determined by preset grid constraints, obtaining a feasibility identifier for the midpoint instruction, which provides a basis for adjusting boundary instructions. Then, the first boundary instruction, the second boundary instruction, and the binary search index are obtained using the midpoint instruction, the midpoint instruction feasibility identifier, and the preset instruction feasibility conditions. This determines the boundary range and index of the binary search, providing a foundation for subsequent iterative searches.

[0015] Further, the binary search metric includes boundary spacing, number of instruction optimizations, and total iteration time; if the binary search metric does not meet the preset binary search termination condition, then a first update instruction and a second update instruction are obtained based on the first boundary instruction and the second boundary instruction, so that the first update instruction and the second update instruction are respectively used as the first boundary instruction and the second boundary instruction, and the binary search process is re-executed until the binary search metric meets the preset binary search termination condition, and the second boundary instruction is determined as the optimization scheduling instruction, including: If the boundary spacing is greater than a preset spacing threshold, the number of instruction optimizations is less than a preset optimization number threshold, and the total iteration time is less than a preset total iteration time threshold, then the first update instruction and the second update instruction are obtained based on the first boundary instruction and the second boundary instruction. The first update instruction and the second update instruction are respectively used as the first boundary instruction and the second boundary instruction, and the binary search process is re-executed until the boundary distance is less than or equal to the preset distance threshold, the number of instruction optimizations is greater than or equal to the preset number of optimizations threshold, or the total iteration time is greater than or equal to the preset total iteration time threshold. The second boundary instruction is then determined as the optimization scheduling instruction.

[0016] In the above scheme, by using boundary spacing, instruction optimization count, and total iteration time as binary search metrics, the scheme can determine whether the binary search has reached the termination condition from three aspects: search range, number of iterations, and computation time. Next, if the boundary spacing is greater than a preset spacing threshold, the number of instruction optimizations is less than a preset optimization count threshold, and the total iteration time is less than a preset total iteration time threshold, the binary search metrics do not meet the preset binary search termination condition. At this point, the scheme obtains a first update instruction and a second update instruction through the first boundary instruction and the second boundary instruction, and uses these as the first boundary instruction and the second boundary instruction to update the boundary of the binary search, gradually narrowing the search range. Then, the binary search process is re-executed to gradually find scheduling instructions that meet the safety requirements. When the boundary spacing is less than or equal to the preset spacing threshold, the number of instruction optimizations is greater than or equal to the preset optimization count threshold, or the total iteration time is greater than or equal to the preset total iteration time threshold, the search stops, and the second boundary instruction is determined as the optimized scheduling instruction, resulting in an optimized scheduling instruction that meets the safety requirements.

[0017] Furthermore, the step of acquiring real-time power grid operating status data, and correcting the optimized scheduling instructions based on the real-time power grid operating status data, the corrected scheduling instructions, the initial target feasible region constraints, and the preset scheduling instruction correction model, to obtain the corrected scheduling instructions, includes: Acquire real-time power grid operation status data, and obtain an optimized network security approximate domain based on the real-time power grid operation status data, initial target feasible domain constraints, optimized scheduling instructions, and corrected scheduling instructions; If the initial target feasible region version is a preset first version, then the optimized network security approximation region is locally tightened based on the preset tightening margin parameter to obtain the tightened network security approximation region; A consistency check is performed on the tightened network security approximation domain. If the check is successful, the initial target feasible domain constraints are updated based on the tightened network security approximation domain to obtain the current target feasible domain constraints. Based on the current feasible domain constraints and the preset scheduling instruction correction model, the optimized scheduling instructions are corrected to obtain the corrected scheduling instructions.

[0018] In the above scheme, by acquiring real-time power grid operation status data and updating the network security approximation domain based on this data, initial target feasible region constraints, optimized dispatch instructions, and corrected dispatch instructions, an optimized network security approximation domain can be obtained, making the constraints more consistent with the actual situation of the power grid. Next, when the initial target feasible region version is a preset first version, the optimized network security approximation domain is locally tightened based on preset tightening margin parameters to obtain a tightened network security approximation domain. This allows for appropriate tightening of constraints, improving the safety of power grid operation. Then, by performing a consistency check on the tightened network security approximation domain, and after passing the check, updating the initial target feasible region constraints based on the tightened network security approximation domain, the current target feasible region constraints are obtained. This ensures that the updated constraints are reasonable and effective, forming the latest constraints applicable to the current power grid. Finally, by processing the optimized dispatch instructions using the current target feasible region constraints and a preset dispatch instruction correction model, the dispatch instructions can be readjusted using the latest constraints, making the final corrected dispatch instructions safer and more reliable.

[0019] Furthermore, the step of acquiring real-time power grid operating status data and, based on the real-time power grid operating status data, initial target feasible region constraints, optimized scheduling instructions, and corrected scheduling instructions, acquiring an optimized network security approximation domain includes: Acquire real-time power grid operating status data, and determine the set of affected constraints based on the real-time power grid operating status data and the initial target feasible region constraints; Determine the subset of network security related constraints based on a pre-defined network security approximation domain and the set of affected constraints; The subset of network security related constraints is updated based on real-time power grid operation status data, optimized scheduling instructions, and corrected scheduling instructions to obtain an approximate domain for optimized network security.

[0020] In the above scheme, by acquiring real-time power grid operation status data and determining the set of affected constraints based on this data and the initial target feasible region constraints, subsequent processing can be performed only on this set of affected constraints. Next, by pre-setting a network security approximation domain and the set of affected constraints, a subset of network security-related constraints is determined. This allows for the selection of network security-related constraints from the affected constraints, narrowing down the range of constraints that need updating. Then, by updating the subset of network security-related constraints using real-time power grid operation status data, optimized scheduling instructions, and corrective scheduling instructions, the relevant constraints are refreshed according to the latest power grid conditions, optimized scheduling instructions, and corrective scheduling instructions, resulting in a more accurate optimized network security approximation domain.

[0021] This invention provides a power grid dispatching command adjustment system, comprising a feasible region construction module, a preliminary correction module, a binary search iteration module, and a command correction module, specifically: The feasible region construction module is used to obtain the original scheduling instructions and the initial target feasible region version based on the power grid to be scheduled, and to construct the initial target feasible region constraints based on the initial target feasible region version. The preliminary correction module is used to correct the original scheduling instructions based on the initial target feasible region constraints and the preset scheduling instruction correction model, and to obtain the corrected scheduling instructions; The binary search iteration module is used to execute a binary search process based on the original scheduling instruction and the correction scheduling instruction to obtain a first boundary instruction, a second boundary instruction, and a binary search index. If the binary search index does not meet the preset binary search termination condition, a first update instruction and a second update instruction are obtained based on the first boundary instruction and the second boundary instruction. The first update instruction and the second update instruction are used as the first boundary instruction and the second boundary instruction, respectively, and the binary search process is re-executed until the binary search index meets the preset binary search termination condition. The second boundary instruction is then determined as the optimized scheduling instruction. The instruction correction module is used to acquire real-time power grid operation status data, and correct the optimized scheduling instructions based on the real-time power grid operation status data, the corrected scheduling instructions, the initial target feasible region constraints, and the preset scheduling instruction correction model, thereby acquiring the corrected scheduling instructions.

[0022] This invention provides a power grid dispatching command adjustment system. In practical applications, it only requires a feasible region construction module. By constructing initial target feasible region constraints, a safety verification benchmark can be established for dispatching commands, avoiding unfounded corrections and readjustments. Next, a preliminary correction module corrects the original dispatching command using the initial target feasible region constraints and a preset correction model, obtaining a corrected dispatching command. This completes safety verification and correction before command issuance, preventing readjustments after issuance. Then, a binary search iteration module outputs an optimized dispatching command when the binary search index meets a preset binary search termination condition. This controls the number of optimizations and optimization time for the corrected dispatching command, preventing dispatching interruptions due to repeated recalculations, while ensuring the output optimized dispatching command is safe and executable. Finally, a command correction module further corrects the optimized dispatching command using real-time power grid operating status data, ensuring the corrected dispatching command continuously meets constraints and the dispatching process remains continuous and uninterrupted.

[0023] Furthermore, the feasible region construction module is used to obtain the original scheduling instructions and the initial target feasible region version based on the power grid to be scheduled, and to construct the initial target feasible region constraints based on the initial target feasible region version, including: Based on the grid to be dispatched, the original dispatch instructions, the original feasible domain version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point ratio coefficients, and grid topology events are obtained. The initial target feasible region version is determined based on the original feasible region version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point proportion coefficients, and grid topology events. Initial target feasible region constraints are then constructed based on the initial target feasible region version.

[0024] In the above scheme, by obtaining the original dispatch instructions, the original feasible region version, and various data of the grid to be dispatched, a complete basis for judgment can be provided for subsequently determining the initial target feasible region version. Then, the initial target feasible region version that best suits the grid's conditions is automatically determined using the various data of the grid to be dispatched. This satisfies adaptive constraint control while balancing safety and economy. Finally, by constructing initial target feasible region constraints based on the initial target feasible region version, a constraint basis is provided for subsequent dispatch instruction correction, binary search, and dispatch instruction modification, ensuring that the entire adjustment process is carried out under the correct constraint criteria. Attached Figure Description

[0025] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0026] Figure 1 A flowchart illustrating a method for adjusting power grid dispatching instructions according to an embodiment of the present invention; Figure 2 This is an architecture diagram of a power grid dispatch command adjustment system provided in an embodiment of the present invention. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0029] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0030] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0031] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0032] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0033] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0034] See Figure 1 To avoid frequent corrections and readjustments of dispatching instructions, and to ensure that security verification and instruction correction are completed before the dispatching instructions are issued, thereby guaranteeing the safe execution of dispatching instructions and the continuous and uninterrupted dispatching process, a method for adjusting power grid dispatching instructions is provided. The flowchart of this method can be found in [link to flowchart]. Figure 1 ,include: Step S1: Obtain the original dispatching instructions and the initial target feasible region version based on the power grid to be dispatched, and construct the initial target feasible region constraints based on the initial target feasible region version; Step S2: Correct the original scheduling instructions based on the initial target feasible region constraints and the preset scheduling instruction correction model to obtain the corrected scheduling instructions; Step S3: Execute a binary search process based on the original scheduling instruction and the corrected scheduling instruction to obtain the first boundary instruction, the second boundary instruction, and the binary search index; if the binary search index does not meet the preset binary search termination condition, obtain the first update instruction and the second update instruction based on the first boundary instruction and the second boundary instruction, and use the first update instruction and the second update instruction as the first boundary instruction and the second boundary instruction respectively to re-execute the binary search process until the binary search index meets the preset binary search termination condition, and determine the second boundary instruction as the optimized scheduling instruction; Step S4: Obtain real-time power grid operation status data, and based on the real-time power grid operation status data, corrected dispatch instructions, initial target feasible region constraints, and preset dispatch instruction correction model, corrected dispatch instructions are modified to obtain corrected dispatch instructions.

[0035] This embodiment focuses on a coastal power grid where renewable energy accounts for approximately 60% of the installed capacity. The area includes centralized wind farms, photovoltaic power stations, and electrochemical energy storage power stations, and contains several critical cross-sections and voltage-weak nodes. In this embodiment, by constructing initial target feasible region constraints, a safety verification benchmark can be established for dispatch instructions, avoiding unfounded corrections and readjustments. Then, the original dispatch instructions are corrected using the initial target feasible region constraints and a preset correction model to obtain corrected dispatch instructions. This allows for safety verification and correction before the instructions are issued, preventing readjustments after issuance. In this embodiment, after correcting the original scheduling command, a corrected initial scheduling command is obtained. This initial command is then subjected to power flow residual checks, safety boundary checks, and margin checks. If all three checks pass, the corrected scheduling command is directly obtained. If any check fails, a more conservative version is switched to, for example, from a normal version to a conservative version, or from a conservative version to an emergency version. The initial target feasible region constraints are then reconstructed based on the switched version, and the three checks are re-executed until all three checks pass, at which point the corrected scheduling command is obtained. Then, by outputting an optimized scheduling command when the binary search index meets the preset binary search termination condition, the number of optimizations and optimization time for the corrected scheduling command can be controlled, preventing scheduling interruptions due to repeated recalculations, while ensuring the output optimized scheduling command is safe and executable. Finally, the optimized scheduling command can be further corrected using real-time power grid operating status data, the corrected scheduling command, the initial target feasible region constraints, and the preset scheduling command correction model, ensuring that the obtained corrected scheduling command continuously meets the constraints and that the scheduling process is continuous and uninterrupted.

[0036] This embodiment enables rapid feasibility recovery and safety correction of original dispatch instructions in a rolling dispatch scenario with a high proportion of renewable energy power grids. Through the switching of multiple versions of feasible domains and the incremental update mechanism triggered by events, it ensures that the feasible domain constraint boundaries corresponding to the feasible domains can adapt to changes in risk level, information quality, and grid structure. Finally, the corrected dispatch instructions are obtained as the basis for adjusting the original dispatch instructions, providing direct support for online decision-making and handling by operators.

[0037] Furthermore, the step of obtaining the original dispatching instruction and the initial target feasible region version based on the power grid to be dispatched, and constructing the initial target feasible region constraints based on the initial target feasible region version, includes: Based on the grid to be dispatched, the original dispatch instructions, the original feasible domain version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point ratio coefficients, and grid topology events are obtained. The initial target feasible region version is determined based on the original feasible region version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point proportion coefficients, and grid topology events. Initial target feasible region constraints are then constructed based on the initial target feasible region version.

[0038] In this embodiment, by obtaining the original dispatching instruction, the original feasible region version, and various data of the grid to be dispatched from the grid to be dispatched, a complete basis for judgment can be provided for subsequently determining the initial target feasible region version. Specifically, the original dispatching instruction is... ,in This indicates the original conventional unit dispatching instructions; This represents the original energy storage dispatch command; This represents the original flexible load dispatching command. Next, the initial target feasible region version that best suits the conditions of the grid to be dispatched is automatically determined using various data from the grid to be dispatched. This version satisfies adaptive constraint control while balancing safety and economy. Then, by constructing initial target feasible region constraints based on the initial target feasible region version, a constraint basis is provided for subsequent dispatching command correction, binary search, and dispatching command modification, ensuring that the entire adjustment process is carried out under the correct constraint criteria.

[0039] Furthermore, the step of determining the initial target feasible region version based on the original feasible region version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly proportion coefficients, and grid topology events, and constructing initial target feasible region constraints based on the initial target feasible region version, includes: Operational risk indicators are obtained based on several key cross-sectional margins, several key voltage margins, several key resource margins, and preset operational risk selection conditions. The risk triggering feasible domain version is determined based on the aforementioned operational risk indicators, the preset first operational risk threshold, and the preset second operational risk threshold. Information quality level indicators are constructed based on power grid state estimation residuals, key measurement missing rates, and power grid anomaly proportion coefficients. The quality triggering feasible domain version is determined based on the information quality level index, the preset first information quality level threshold, and the preset second information quality level threshold. The topology triggering feasible domain version is determined based on power grid topology events and a preset power grid topology event table; The initial target feasible domain version is determined based on the original feasible domain version, the risk-triggered feasible domain version, the quality-triggered feasible domain version, and the topology-triggered feasible domain version; Construct initial target feasible region constraints based on the initial target feasible region version.

[0040] In this embodiment, the operational risk index is calculated using key section margin, key voltage margin, key resource margin, and preset risk conditions. Specifically, the minimum margins corresponding to the key section margin, key voltage margin, and key resource margin are calculated and normalized. The minimum value among the normalized minimum margins of the three is taken as the operational risk index, which can intuitively reflect the current security tension of the grid to be dispatched, so as to determine whether constraints need to be strengthened. Next, by comparing the operational risk index with preset first operational risk thresholds and preset second operational risk thresholds, the security risk level of the grid can be determined, and the feasible risk triggering domain version is automatically matched according to the security risk level of the grid. Specifically: It was determined to be low risk, among which, To implement risk indicators, To preset the first operational risk threshold, It was determined to be of medium risk, among which, To preset a second operational risk threshold, The risk level is determined to be high. At low risk, the risk-triggered feasible domain version is matched with the preset normal version feasible domain; at medium risk, the risk-triggered feasible domain version is matched with the preset conservative version feasible domain; and at high risk, the risk-triggered feasible domain version is matched with the preset emergency version feasible domain. The preset first operational risk threshold and preset second operational risk threshold can be adjusted based on historical power grid operational statistics.

[0041] The preset normal version feasible domain, preset conservative version feasible domain, and preset emergency version feasible domain can be composed of a preset static hard constraint domain, a preset network security approximation domain, and a preset emergency safety verification domain. The preset static hard constraint domain is used to limit the output and regulation range of units, energy storage devices, and flexible loads in the power grid, ensuring that dispatching instructions do not exceed the equipment's own capacity and the planned allowable boundaries. It is the basic constraint that all versions of feasible domains must meet. The preset network security approximation domain can quickly calculate the impact of key section power flow and key node voltage on changes in dispatching instructions based on current power grid operation data, and is used to quickly determine whether dispatching instructions will cause over-limits. The preset emergency safety verification domain is used to directly verify key equipment, key sections, and voltage constraints through fast power flow calculations, ensuring that dispatching instructions meet safety requirements in high-risk, topology-changing, or poor data quality situations.

[0042] Based on the three-layer feasible domain, construct the preset normal version feasible domain, the preset conservative version feasible domain, and the preset emergency version feasible domain: Preset normal version feasible domain ,in, To pre-define the static hard constraint domain, To define a predefined network security approximation domain, based on the predefined normal version feasible domain, the optimized network security approximation domain is locally tightened using predefined tightening margin parameters to obtain the tightened network security approximation domain. Then, the predefined conservative version feasible domain is obtained by combining the predefined static hard constraint domain and the tightened network security approximation domain. ,in, To define the feasible domain for the conservative version, To tighten the network security approximation domain, a pre-defined emergency security verification domain is introduced for strong constraints, under the premise of a pre-defined static hard constraint domain. The approximation domain can also be retained as a pre-defined emergency version feasible domain. ,in, To pre-define the feasible domains for emergency versions, This is a preset emergency safety verification domain. From the above, the inclusion relationship of the three feasible domains can be obtained: .

[0043] Then, an information quality level index is constructed using state estimation residuals, key measurement missing rates, and the proportion coefficient of power grid anomalies. This index can accurately determine the reliability of the power grid and whether there are significant errors. The quality level index is then compared with preset first and second information quality level thresholds to determine the feasible version of the quality triggering domain. Specifically: When determining the quality trigger feasible domain version as the preset normal version, among which, As an indicator of information quality level, To preset the first information quality level threshold, When determining the quality triggering feasible domain version, it is set to the preset conservative version, where, To preset a second information quality level threshold, The feasible version for quality triggering is determined to be the preset emergency version. The preset first information quality level threshold and the preset second information quality level threshold can be adjusted based on historical power grid operation statistics.

[0044] Subsequently, the feasible version of the topology trigger is determined through grid topology events and a preset grid topology event table. This allows for the determination of the required constraint version when the grid structure changes or equipment operates. Specifically: if a grid topology event occurs within a critical section, critical voltage point associated channel, or critical equipment list listed in the preset grid topology event table, causing the sensitivity change of critical equipment to exceed a preset sensitivity threshold, then the feasible version of the topology trigger is determined to be the preset conservative version; if the grid topology event belongs to a critical section margin rapidly tightening event, a critical voltage out-of-bounds risk increase event, or a critical equipment failure to complete critical sensitivity update and consistency verification event within a preset time period, then the feasible version of the topology trigger is determined to be the preset emergency version; if it is not determined to be the preset conservative version or the preset emergency version, then the feasible version of the topology trigger is determined to be the preset normal version.

[0045] Finally, by comprehensively considering the original feasible region version, the risk-triggered feasible region version, the quality-triggered feasible region version, and the topology-triggered feasible region version, the final initial target feasible region version is determined. This allows for the selection of the safest and most suitable constraint caliber for the current power grid conditions, preventing operational risks to the power grid. Then, based on the initial target feasible region version, initial target feasible region constraints are constructed to provide a reliable constraint basis for the correction of subsequent dispatch instructions. Specifically, the most conservative feasible region version among the risk-triggered, quality-triggered, and topology-triggered feasible region versions is selected as the initial feasible region version. Among these, the normal version is more conservative than the conservative version, and the conservative version is more conservative than the emergency version. After obtaining the initial feasible region version, it is compared with the original feasible region version. If they are different, the initial feasible region version is used as the initial target feasible region version, and initial target feasible region constraints are constructed based on the initial target feasible region version. At the same time, the original feasible region constraints corresponding to the original feasible region version in the power grid are unloaded, and the initial target feasible region constraints are loaded to provide a reliable constraint basis for the correction of subsequent scheduling instructions. If they are the same, the original feasible region version is directly used as the initial target feasible region version, and initial target feasible region constraints are constructed based on the initial target feasible region version to provide a reliable constraint basis for the correction of subsequent scheduling instructions.

[0046] During the process of switching feasible domain versions, if the feasible domain version is switched from the normal version to the conservative version, or from the conservative version to the emergency version, it will take effect immediately. If the feasible domain version is switched from the emergency version to the conservative version, or from the conservative version to the normal version, the following conditions must be met: the stability duration corresponding to the operational risk indicator and the stability duration corresponding to the information quality level indicator must both be no less than the preset version retention time to avoid frequent switching. The preset version retention time can be adjusted based on the historical operational statistics of the power grid.

[0047] Furthermore, in the process of correcting the original scheduling instructions based on the initial target feasible region constraints and the preset scheduling instruction correction model to obtain the corrected scheduling instructions, the construction process of the preset scheduling instruction correction model includes: Obtain several historical initial scheduling instructions, several historical actual scheduling instructions corresponding to several historical initial scheduling instructions, and several historical actual feasible domain versions corresponding to several historical initial scheduling instructions. Construct several historical feasible domain constraints based on several historical feasible domain versions; Based on several historical initial scheduling instructions, several historical actual feasible domain constraints and a preset initial scheduling instruction correction model, several first scheduling instructions corresponding to several historical initial scheduling instructions are obtained. A loss function is constructed based on the historical initial scheduling instruction and the first scheduling instruction corresponding to the historical initial scheduling instruction. The preset initial scheduling instruction correction model is trained by minimizing the loss function until the preset stopping condition is met, and the preset scheduling instruction correction model is obtained.

[0048] In this embodiment, by acquiring several historical initial scheduling instructions, several historical actual scheduling instructions corresponding to these historical initial scheduling instructions, and several historical actual feasible domain versions corresponding to these historical initial scheduling instructions, a data foundation can be provided for training a preset initial scheduling instruction correction model. Next, historical actual feasible domain constraints are constructed using the historical actual feasible domain versions, providing accurate constraint basis for model training according to historical real constraints. Then, several first scheduling instructions corresponding to the historical initial scheduling instructions are obtained using the historical initial scheduling instructions, historical actual feasible domain constraints, and the preset initial scheduling instruction correction model, enabling preliminary correction of the historical initial scheduling instructions. Subsequently, a loss function is constructed using the historical initial scheduling instructions and the corresponding first scheduling instructions, allowing the calculation of the deviation between the model output and the actual output, providing direction for model optimization, specifically: ,in, This refers to historical power grid operating status data corresponding to historical initial dispatch instructions. This is the first scheduling instruction. These are historical actual dispatch instructions. Let i be the preset slack variables for the i safety constraints in the historical feasible region constraints. , As a preset margin threshold, . This indicates that there are m safety constraints in the historical feasible region constraints, and W is a preset device weight matrix. ,in Let j be the weight of the j-th device in the power grid. The preset interior point barrier coefficient, , used to control the strength of the preset slack variables away from the constraint boundary, is . Finally, the preset initial scheduling instruction correction model is trained with the goal of minimizing the loss function. This allows the model to be trained without optimization until the preset stopping condition is met, ultimately yielding an accurate preset scheduling instruction correction model.

[0049] Further, the step of performing a binary search process based on the original scheduling instruction and the corrected scheduling instruction to obtain the first boundary instruction, the second boundary instruction, and the binary search index includes: Midpoint instructions are generated based on the original scheduling instructions and the corrected scheduling instructions. Based on preset power grid constraints, the feasibility of the midpoint command is determined, and the feasibility identifier of the midpoint command is obtained. The first boundary instruction, the second boundary instruction, and the binary search index are obtained based on the midpoint instruction, the midpoint instruction feasibility identifier, and the preset instruction feasibility conditions.

[0050] In this embodiment, a midpoint instruction is generated by using the original scheduling instruction and the corrected scheduling instruction. This yields an intermediate instruction between the two instructions, providing an object for subsequent feasibility assessment. Specifically, the preset lower bound is... Corresponding to the original scheduling instruction The upper limit is preset to be Corresponding to the correction scheduling instruction ,according to The midpoint of the boundary is obtained. Substitute A midpoint command is generated. Next, the feasibility of the midpoint command is determined using preset power grid constraints, which yields a feasibility indicator for the midpoint command, providing a basis for adjusting boundary commands. Specifically: ; Wherein, if an intermediate instruction satisfies the preset grid constraints, the midpoint instruction is marked as feasible; if an intermediate instruction does not satisfy the preset grid constraints, the midpoint instruction is marked as infeasible. The preset grid constraints are... Where x represents the power grid operating status data, and u represents intermediate commands. This indicates that the power flow equations are satisfied, guaranteeing that the power flow is solvable; This indicates that static safety constraints are met. Then, by using the midpoint instruction, the midpoint instruction feasibility flag, and preset instruction feasibility conditions, the first boundary instruction, the second boundary instruction, and the binary search index are obtained, determining the boundary range and index of the binary search, providing a foundation for subsequent iterative searches. Specifically, the preset instruction feasibility conditions are: if the midpoint instruction feasibility flag is feasible, then the original scheduling instruction is used as the first boundary instruction, and the midpoint instruction is used as the second boundary instruction, making the obtained second boundary instruction closer to the original scheduling instruction; if the midpoint instruction feasibility flag is infeasible, then the midpoint instruction is used as the first boundary instruction, and the corrected scheduling instruction is used as the second boundary instruction, making the obtained first boundary instruction closer to the corrected scheduling instruction.

[0051] Further, the binary search metric includes boundary spacing, number of instruction optimizations, and total iteration time; if the binary search metric does not meet the preset binary search termination condition, then a first update instruction and a second update instruction are obtained based on the first boundary instruction and the second boundary instruction, so that the first update instruction and the second update instruction are respectively used as the first boundary instruction and the second boundary instruction, and the binary search process is re-executed until the binary search metric meets the preset binary search termination condition, and the second boundary instruction is determined as the optimization scheduling instruction, including: If the boundary spacing is greater than a preset spacing threshold, the number of instruction optimizations is less than a preset optimization number threshold, and the total iteration time is less than a preset total iteration time threshold, then the first update instruction and the second update instruction are obtained based on the first boundary instruction and the second boundary instruction. The first update instruction and the second update instruction are respectively used as the first boundary instruction and the second boundary instruction, and the binary search process is re-executed until the boundary distance is less than or equal to the preset distance threshold, the number of instruction optimizations is greater than or equal to the preset number of optimizations threshold, or the total iteration time is greater than or equal to the preset total iteration time threshold. The second boundary instruction is then determined as the optimization scheduling instruction.

[0052] In this embodiment, by using boundary spacing, instruction optimization count, and total iteration time as binary search metrics, the termination condition of the binary search can be determined from three aspects: search range, number of iterations, and computation time. Next, if the boundary spacing is greater than a preset spacing threshold, the number of instruction optimizations is less than a preset optimization count threshold, and the total iteration time is less than a preset total iteration time threshold, the binary search metrics do not meet the preset binary search termination condition. At this point, a first update instruction and a second update instruction are obtained through the first boundary instruction and the second boundary instruction, and these are used as the first boundary instruction and the second boundary instruction to update the boundary of the binary search, gradually narrowing the search range. Then, the binary search process is re-executed to gradually find scheduling instructions that meet the safety requirements. When the boundary spacing is less than or equal to a preset spacing threshold, the number of instruction optimizations is greater than or equal to a preset optimization count threshold, or the total iteration time is greater than or equal to a preset total iteration time threshold, the search stops, and the second boundary instruction is determined as the optimized scheduling instruction, resulting in an optimized scheduling instruction that meets the safety requirements.

[0053] Furthermore, the step of acquiring real-time power grid operating status data, and correcting the optimized scheduling instructions based on the real-time power grid operating status data, the corrected scheduling instructions, the initial target feasible region constraints, and the preset scheduling instruction correction model, to obtain the corrected scheduling instructions, includes: Acquire real-time power grid operation status data, and obtain an optimized network security approximate domain based on the real-time power grid operation status data, initial target feasible domain constraints, optimized scheduling instructions, and corrected scheduling instructions; If the initial target feasible region version is a preset first version, then the optimized network security approximation region is locally tightened based on the preset tightening margin parameter to obtain the tightened network security approximation region; A consistency check is performed on the tightened network security approximation domain. If the check is successful, the initial target feasible domain constraints are updated based on the tightened network security approximation domain to obtain the current target feasible domain constraints. Based on the current feasible domain constraints and the preset scheduling instruction correction model, the optimized scheduling instructions are corrected to obtain the corrected scheduling instructions.

[0054] In this embodiment, by acquiring real-time power grid operating status data and updating the network security approximation domain based on the real-time power grid operating status data, initial target feasible region constraints, optimized scheduling instructions, and corrected scheduling instructions, an optimized network security approximation domain can be obtained, making the constraints more consistent with the actual situation of the power grid. Next, when the initial target feasible region version is a preset first version, the optimized network security approximation domain is locally tightened based on preset tightening margin parameters to obtain a tightened network security approximation domain. This allows for appropriate tightening of constraints, improving the safety of power grid operation. Then, by performing a consistency check on the tightened network security approximation domain, and after passing the check, updating the initial target feasible region constraints based on the tightened network security approximation domain, the current target feasible region constraints are obtained. This ensures that the updated constraints are reasonable and effective, forming the latest constraints applicable to the current power grid. Finally, by processing the optimized scheduling instructions using the current target feasible region constraints and a preset scheduling instruction correction model, the scheduling instructions can be readjusted using the latest constraints, making the final corrected scheduling instructions safer and more reliable. Simultaneously, the obtained corrected scheduling instructions must meet a preset minimum modification correction criterion. ,in, To correct the scheduling instructions, To optimize scheduling instructions, These are scheduling instructions to be adjusted within the feasible domain. W represents the current feasible region constraint for the target, and W is the preset device weight matrix.

[0055] In this embodiment, if a corrected scheduling instruction that satisfies the current target feasible domain constraints cannot be found in the initial target feasible domain version, or if the time taken from obtaining the original scheduling instruction to obtaining the corrected scheduling instruction exceeds a preset adjustment scheduling instruction time threshold, a preset hierarchical rollback mechanism is immediately and automatically triggered, prioritizing the reuse of the legal and valid historical scheduling instructions from the previous moment. If the historical scheduling instruction is not applicable (i.e., the consistency check of the historical scheduling instruction fails), a preset scheduling strategy adapted to the current power grid operating conditions is quickly called from the preset emergency instruction feasible strategy library. If the preset emergency instruction feasible strategy library cannot match the current power grid requirements, the system further switches to the emergency version and uses a preset safety backup template to generate basic scheduling instructions.

[0056] Furthermore, the step of acquiring real-time power grid operating status data and, based on the real-time power grid operating status data, initial target feasible region constraints, optimized scheduling instructions, and corrected scheduling instructions, acquiring an optimized network security approximation domain includes: Acquire real-time power grid operating status data, and determine the set of affected constraints based on the real-time power grid operating status data and the initial target feasible region constraints; Determine the subset of network security related constraints based on a pre-defined network security approximation domain and the set of affected constraints; The subset of network security related constraints is updated based on real-time power grid operation status data, optimized scheduling instructions, and corrected scheduling instructions to obtain an approximate domain for optimized network security.

[0057] In this embodiment, by acquiring real-time power grid operating status data and determining the set of affected constraints based on the real-time power grid operating status data and the initial target feasible region constraints, subsequent processing can be performed only on the set of affected constraints. Specifically: ,in These are topology impact events obtained based on real-time power grid operating status data. To pre-define the full set of key constraint indexes, The set of affected constraints needs to be incrementally updated. This is an impact discrimination function used to determine whether a constraint in a preset full set of key constraint indices belongs to the set of affected constraints based on topological impact events. The topology-affecting event is determined to have a direct impact on constraint j, thus belonging to the affected constraint set. Next, by pre-setting the network security approximation domain and the affected constraint set, a subset of network security-related constraints is determined. This allows for the filtering of network security-related constraints from the affected constraints, narrowing down the range of constraints requiring updates. Then, the subset of network security-related constraints is updated using real-time power grid operating status data, optimized dispatch instructions, and corrective dispatch instructions. This allows for refreshing the relevant constraints according to the latest power grid conditions, optimized dispatch instructions, and corrective dispatch instructions, resulting in a more accurate optimized network security approximation domain.

[0058] This embodiment provides a power grid dispatch command adjustment system, including a feasible region construction module, a preliminary correction module, a binary search iteration module, and a command correction module, specifically: The feasible region construction module is used to obtain the original scheduling instructions and the initial target feasible region version based on the power grid to be scheduled, and to construct the initial target feasible region constraints based on the initial target feasible region version. The preliminary correction module is used to correct the original scheduling instructions based on the initial target feasible region constraints and the preset scheduling instruction correction model, and to obtain the corrected scheduling instructions; The binary search iteration module is used to execute a binary search process based on the original scheduling instruction and the correction scheduling instruction to obtain a first boundary instruction, a second boundary instruction, and a binary search index. If the binary search index does not meet the preset binary search termination condition, a first update instruction and a second update instruction are obtained based on the first boundary instruction and the second boundary instruction. The first update instruction and the second update instruction are used as the first boundary instruction and the second boundary instruction, respectively, and the binary search process is re-executed until the binary search index meets the preset binary search termination condition. The second boundary instruction is then determined as the optimized scheduling instruction. The instruction correction module is used to acquire real-time power grid operation status data, and correct the optimized scheduling instructions based on the real-time power grid operation status data, the corrected scheduling instructions, the initial target feasible region constraints, and the preset scheduling instruction correction model, thereby acquiring the corrected scheduling instructions.

[0059] This embodiment provides a power grid dispatch command adjustment system. In practical applications, it only requires a feasible region construction module. By constructing initial target feasible region constraints, a safety verification benchmark can be established for dispatch commands, avoiding unfounded corrections and readjustments. Next, a preliminary correction module is used to correct the original dispatch command using the initial target feasible region constraints and a preset correction model, obtaining a corrected dispatch command. This allows for safety verification and correction before command issuance, preventing readjustments after issuance. Then, a binary search iteration module is used. By outputting an optimized dispatch command when the binary search index meets a preset binary search termination condition, the number of optimizations and optimization time for the corrected dispatch command can be controlled, preventing dispatch interruptions due to repeated recalculations, while ensuring the output optimized dispatch command is safe and executable. Finally, a command correction module is used. Using real-time power grid operating status data, the corrected dispatch command, the initial target feasible region constraints, and the preset dispatch command correction model, the optimized dispatch command can be further corrected, ensuring that the obtained corrected dispatch command continuously meets the constraints and that the dispatch process is continuous and uninterrupted.

[0060] Furthermore, the feasible region construction module is used to obtain the original scheduling instructions and the initial target feasible region version based on the power grid to be scheduled, and to construct the initial target feasible region constraints based on the initial target feasible region version, including: Based on the grid to be dispatched, the original dispatch instructions, the original feasible domain version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point ratio coefficients, and grid topology events are obtained. The initial target feasible region version is determined based on the original feasible region version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point proportion coefficients, and grid topology events. Initial target feasible region constraints are then constructed based on the initial target feasible region version.

[0061] In this embodiment, by obtaining the original dispatching command, the original feasible region version, and various data of the grid to be dispatched, a complete basis for determining the initial target feasible region version can be provided. Next, the initial target feasible region version that best suits the grid's conditions is automatically determined using the data of the grid to be dispatched. This satisfies adaptive constraint control while balancing safety and economy. Then, by constructing initial target feasible region constraints based on the initial target feasible region version, a constraint basis is provided for subsequent dispatching command correction, binary search, and dispatching command modification, ensuring that the entire adjustment process is carried out under the correct constraint criteria.

[0062] This embodiment also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the functions of the system as described above.

[0063] It is understood that the above system item embodiments correspond to the method item embodiments of the present invention, and can implement the method for adjusting power grid dispatching instructions provided by any of the above method item embodiments of the present invention.

[0064] It should be noted that the system embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0065] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for adjusting power grid dispatching instructions, characterized in that, include: The original dispatching instructions and the initial target feasible region version are obtained based on the power grid to be dispatched, and the initial target feasible region constraints are constructed based on the initial target feasible region version. The original scheduling instructions are corrected based on the initial target feasible region constraints and the preset scheduling instruction correction model to obtain the corrected scheduling instructions; A binary search process is executed based on the original scheduling instruction and the corrected scheduling instruction to obtain a first boundary instruction, a second boundary instruction, and a binary search index. If the binary search index does not meet the preset binary search termination condition, a first update instruction and a second update instruction are obtained based on the first boundary instruction and the second boundary instruction. The first update instruction and the second update instruction are used as the first boundary instruction and the second boundary instruction, respectively, and the binary search process is re-executed until the binary search index meets the preset binary search termination condition. The second boundary instruction is then determined as the optimized scheduling instruction. The system acquires real-time power grid operation status data, and corrects the optimized scheduling instructions based on the real-time power grid operation status data, corrected scheduling instructions, initial target feasible region constraints, and preset scheduling instruction correction model, thereby obtaining corrected scheduling instructions.

2. The method for adjusting power grid dispatching instructions according to claim 1, characterized in that, The step of obtaining the original dispatching instructions and the initial target feasible region version based on the power grid to be dispatched, and constructing the initial target feasible region constraints based on the initial target feasible region version, includes: Based on the grid to be dispatched, the original dispatch instructions, the original feasible domain version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point ratio coefficients, and grid topology events are obtained. The initial target feasible region version is determined based on the original feasible region version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point proportion coefficients, and grid topology events. Initial target feasible region constraints are then constructed based on the initial target feasible region version.

3. The method for adjusting power grid dispatching instructions according to claim 2, characterized in that, The initial target feasible region version is determined based on the original feasible region version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly proportion coefficients, and grid topology events. Initial target feasible region constraints are then constructed based on these initial target feasible region versions, including: Operational risk indicators are obtained based on several key cross-sectional margins, several key voltage margins, several key resource margins, and preset operational risk selection conditions. The risk triggering feasible domain version is determined based on the aforementioned operational risk indicators, the preset first operational risk threshold, and the preset second operational risk threshold. Information quality level indicators are constructed based on power grid state estimation residuals, key measurement missing rates, and power grid anomaly proportion coefficients. The quality triggering feasible domain version is determined based on the information quality level index, the preset first information quality level threshold, and the preset second information quality level threshold. The topology triggering feasible domain version is determined based on power grid topology events and a preset power grid topology event table; The initial target feasible domain version is determined based on the original feasible domain version, the risk-triggered feasible domain version, the quality-triggered feasible domain version, and the topology-triggered feasible domain version; Construct initial target feasible region constraints based on the initial target feasible region version.

4. The method for adjusting power grid dispatching instructions according to claim 1, characterized in that, In obtaining corrected scheduling instructions by correcting the original scheduling instructions based on the initial target feasible region constraints and the preset scheduling instruction correction model, the construction process of the preset scheduling instruction correction model includes: Obtain several historical initial scheduling instructions, several historical actual scheduling instructions corresponding to several historical initial scheduling instructions, and several historical actual feasible domain versions corresponding to several historical initial scheduling instructions. Construct several historical feasible domain constraints based on several historical feasible domain versions; Based on several historical initial scheduling instructions, several historical actual feasible domain constraints and a preset initial scheduling instruction correction model, several first scheduling instructions corresponding to several historical initial scheduling instructions are obtained. A loss function is constructed based on the historical initial scheduling instruction and the first scheduling instruction corresponding to the historical initial scheduling instruction. The preset initial scheduling instruction correction model is trained by minimizing the loss function until the preset stopping condition is met, and the preset scheduling instruction correction model is obtained.

5. The method for adjusting power grid dispatching instructions according to claim 1, characterized in that, The process of performing a binary search based on the original scheduling instruction and the corrected scheduling instruction to obtain the first boundary instruction, the second boundary instruction, and the binary search index includes: Midpoint instructions are generated based on the original scheduling instructions and the corrected scheduling instructions. Based on preset power grid constraints, the feasibility of the midpoint command is determined, and the feasibility identifier of the midpoint command is obtained. The first boundary instruction, the second boundary instruction, and the binary search index are obtained based on the midpoint instruction, the midpoint instruction feasibility identifier, and the preset instruction feasibility conditions.

6. The method for adjusting power grid dispatching instructions according to claim 1, characterized in that, The binary search metric includes boundary spacing, number of instruction optimizations, and total iteration time. If the binary search metric does not meet the preset binary search termination condition, a first update instruction and a second update instruction are obtained based on the first boundary instruction and the second boundary instruction, respectively, and the first update instruction and the second update instruction are used as the first boundary instruction and the second boundary instruction, and the binary search process is re-executed until the binary search metric meets the preset binary search termination condition. The second boundary instruction is then determined as the optimization scheduling instruction, including: If the boundary spacing is greater than a preset spacing threshold, the number of instruction optimizations is less than a preset optimization number threshold, and the total iteration time is less than a preset total iteration time threshold, then the first update instruction and the second update instruction are obtained based on the first boundary instruction and the second boundary instruction. The first update instruction and the second update instruction are respectively used as the first boundary instruction and the second boundary instruction, and the binary search process is re-executed until the boundary distance is less than or equal to the preset distance threshold, the number of instruction optimizations is greater than or equal to the preset number of optimizations threshold, or the total iteration time is greater than or equal to the preset total iteration time threshold. The second boundary instruction is then determined as the optimization scheduling instruction.

7. The method for adjusting power grid dispatching instructions according to claim 1, characterized in that, The process of acquiring real-time power grid operation status data, and modifying the optimized scheduling instructions based on the real-time power grid operation status data, corrected scheduling instructions, initial target feasible region constraints, and a preset scheduling instruction correction model, to obtain corrected scheduling instructions, includes: Acquire real-time power grid operation status data, and obtain an optimized network security approximate domain based on the real-time power grid operation status data, initial target feasible domain constraints, optimized scheduling instructions, and corrected scheduling instructions; If the initial target feasible region version is a preset first version, then the optimized network security approximation region is locally tightened based on the preset tightening margin parameter to obtain the tightened network security approximation region; A consistency check is performed on the tightened network security approximation domain. If the check is successful, the initial target feasible domain constraints are updated based on the tightened network security approximation domain to obtain the current target feasible domain constraints. Based on the current feasible domain constraints and the preset scheduling instruction correction model, the optimized scheduling instructions are corrected to obtain the corrected scheduling instructions.

8. The method for adjusting power grid dispatching instructions according to claim 7, characterized in that, The process of acquiring real-time power grid operation status data and obtaining an optimized network security approximation domain based on the real-time power grid operation status data, initial target feasible region constraints, optimized scheduling instructions, and corrected scheduling instructions includes: Acquire real-time power grid operating status data, and determine the set of affected constraints based on the real-time power grid operating status data and the initial target feasible region constraints; Determine the subset of network security related constraints based on a pre-defined network security approximation domain and the set of affected constraints; The subset of network security related constraints is updated based on real-time power grid operation status data, optimized scheduling instructions, and corrected scheduling instructions to obtain an approximate domain for optimized network security.

9. A power grid dispatching command adjustment system, characterized in that, It includes a feasible region construction module, a preliminary correction module, a binary search iteration module, and an instruction correction module, specifically: The feasible region construction module is used to obtain the original scheduling instructions and the initial target feasible region version based on the power grid to be scheduled, and to construct the initial target feasible region constraints based on the initial target feasible region version. The preliminary correction module is used to correct the original scheduling instructions based on the initial target feasible region constraints and the preset scheduling instruction correction model, and to obtain the corrected scheduling instructions; The binary search iteration module is used to execute a binary search process based on the original scheduling instruction and the correction scheduling instruction to obtain a first boundary instruction, a second boundary instruction, and a binary search index. If the binary search index does not meet the preset binary search termination condition, a first update instruction and a second update instruction are obtained based on the first boundary instruction and the second boundary instruction. The first update instruction and the second update instruction are used as the first boundary instruction and the second boundary instruction, respectively, and the binary search process is re-executed until the binary search index meets the preset binary search termination condition. The second boundary instruction is then determined as the optimized scheduling instruction. The instruction correction module is used to acquire real-time power grid operation status data, and correct the optimized scheduling instructions based on the real-time power grid operation status data, the corrected scheduling instructions, the initial target feasible domain constraints, and the preset scheduling instruction correction model, so as to obtain the corrected scheduling instructions.

10. The power grid dispatching instruction adjustment system according to claim 9, characterized in that, The feasible region construction module is used to obtain the original dispatching instructions and the initial target feasible region version based on the power grid to be dispatched, and to construct the initial target feasible region constraints based on the initial target feasible region version, including: Based on the grid to be dispatched, the original dispatch instructions, the original feasible domain version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point ratio coefficients, and grid topology events are obtained. The initial target feasible region version is determined based on the original feasible region version, several key section margins, several key voltage margins, several key resource margins, grid state estimation residuals, key measurement missing rates, grid anomaly point proportion coefficients, and grid topology events. Initial target feasible region constraints are then constructed based on the initial target feasible region version.

Images