A multi-element load linkage regulation method, system, device and medium

Abstract

The application discloses a kind of multi-element load linkage regulation method, system, equipment and medium, comprising: constructing the two-stage operation transaction model of multi-element load aggregation main body user, forms lower user response model;Establish two-stage robust regulation model dominated by distribution network operator;Introduce power grid topology adjustable constraint;Coupling user response model and enhanced operator regulation model, build double-layer interactive optimization framework, form two-stage robust double-layer optimization problem containing integer variable;Optimization problem is solved using nested column constraint generation algorithm, output optimal electricity price guide strategy, direct load control amount and topology adjustment scheme, as the total scheme of multi-element load linkage regulation for weak point uncertainty.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a method, system, equipment and medium for multi-load linkage control. Background Technology

[0002] With the continuous improvement and development of the new power system based on new energy sources, load regulation in the market environment increasingly needs to consider the various actual needs of users in multi-load aggregation entities. This requires a comprehensive analysis of many factors, such as the operating characteristics of the power grid itself, the potential for flexible load dispatch, and the linkage and interaction with the electricity market. Currently, with the emergence of distributed energy and local energy sharing mechanisms, multi-load aggregation entities are generally built on the distribution network side. As integrators of demand-side flexible resources, they act as agents for the electricity consumption behavior of small and medium-sized users in the region, helping them participate in the electricity market and make decisions on purchasing and selling electricity. In effect, they serve as a bridge between small and medium-sized users and distribution network operation and dispatch.

[0003] At the distribution network level, the reliability of power equipment is particularly critical to the stable operation of the entire system. Equipment such as transformers, transmission lines, and generators may fail due to inherent problems—such as aging or manufacturing defects—or external environmental factors, such as severe weather and climate disasters like thunderstorms, typhoons, and snowstorms. Once equipment fails, it not only causes power outages but also alters the system topology, disrupts power flow, and breaks the original power balance, leading to safety risks and economic losses. Moreover, due to the limited number of fault samples and long statistical periods, the estimated probability of failure often deviates significantly from the actual situation. Therefore, the weak points in the power grid caused by equipment failure become a significant uncertainty factor in dispatching and operation. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a method, system, device, and medium for multi-load linkage regulation to address the problem that existing technologies, under the background of new power systems, lack collaborative modeling of the participation of multi-load aggregation entities in the market and the uncertainty of distribution network equipment failures. This makes it difficult to accurately characterize the operational weaknesses caused by equipment failures and fails to effectively integrate user-side flexibility resources with grid reliability requirements, resulting in significant deficiencies in dispatch decisions when dealing with security risks under high-proportion renewable energy access.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for multi-load linkage control, comprising: Obtain basic operational data and power distribution network system parameters; Based on the acquired data, a two-stage operational transaction model for multiple load aggregation entities is constructed to form a lower-level user response model. Based on the user response model, a two-stage robust control model led by the distribution network operator is established with the goal of minimizing the system operation cost and load control cost before and after the occurrence of the vulnerability. The direct load control amount is used as the decision variable in the stage after the occurrence of the vulnerability, and the identification mechanism of the most severe vulnerability uncertainty scenario is embedded. In the aforementioned distribution network operator control model, adjustable grid topology constraints are introduced to obtain an enhanced operator control model. Couple the user response model with the enhanced operator control model to construct a two-layer interactive optimization framework, forming a two-stage robust two-layer optimization problem with integer variables. The optimization problem is solved by a nested column constraint generation algorithm, which outputs the optimal electricity price guidance strategy, direct load control amount and topology adjustment scheme as a general scheme for multi-level load linkage control facing the uncertainty of weak points.

[0007] As a preferred embodiment of the multi-load linkage control method described in this invention, the construction of a two-stage operation transaction model for multi-load aggregation subject users includes: Based on the uncertainty characteristics of system operation weaknesses, a two-stage operation model framework for a single multi-dimensional load aggregation subject user is obtained by constructing a user optimization architecture that includes two stages: before and after the occurrence of the weakness. Based on the modeling requirements of the pre-vulnerability stage in the two-stage operation model framework, a user cost model for the pre-vulnerability stage is obtained by establishing a cost structure that includes the power generation cost of small gas turbines, the point-to-point energy transaction cost with other multi-load aggregation main users, and the electricity purchase and sale settlement cost with the distribution network operator. Based on the point-to-point energy transaction items included in the user cost model, by establishing the energy conservation relationship of the transaction volume and the total transaction volume aggregation constraint, and using corresponding dual variables to represent the transaction price, a set of consistency constraints for energy transactions among users of multiple load aggregation entities is obtained. Based on the normal operating state defined by the user cost model and the consistency constraint set, by establishing an active and reactive power balance equation that includes distributed power output, energy storage charging and discharging, main grid power purchase and sale and point-to-point transactions, and setting the gas turbine output limit, photovoltaic power generation upper limit, energy storage charging and discharging power limit, state of charge dynamic range, power purchase and sale mutual exclusion constraint and the binary logic variables corresponding to the energy storage charging and discharging mutual exclusion constraint, the user operation constraint set before the occurrence of the weakness point is obtained. Based on the modeling requirements of the post-vulnerability stage in the two-stage operation model framework, by establishing expressions for power generation cost, point-to-point transaction cost, and main grid power purchase and sale cost with the same structure as the previous stage but updated variables, a user cost model for the post-vulnerability stage is obtained. Based on the direct load regulation introduced in the user cost model after the occurrence of the vulnerability, by establishing an active and reactive power balance equation that includes distributed resource output, transaction behavior and load reduction, and applying similar operating limits to gas turbines, photovoltaics and energy storage as in the previous stage, the user emergency operation constraint set for the stage after the occurrence of the vulnerability is obtained. Based on the aforementioned two-stage operation model framework, user cost model before and after the occurrence of weak points, energy transaction consistency constraint set, user operation constraint set, and user emergency operation constraint set, a two-stage operation transaction model for multi-load aggregation subject users is constructed by integrating all cost items and constraints.

[0008] The beneficial effects of this preferred technical solution are that by constructing a complete user-side model that considers the two stages before and after the weak points, integrates the operating characteristics of distributed resources, the point-to-point transaction mechanism and the direct load control response, it realizes the coordinated optimization of the economics and physical feasibility of multiple load aggregation entities under system uncertainty, and provides an accurate and interactive lower-level response basis for upper-level operator control.

[0009] As a preferred embodiment of the multi-load linkage control method described in this invention, the establishment of a two-stage robust control model led by the distribution network operator includes: Based on the uncertainty characteristics of system operation weaknesses, an optimization architecture that includes two stages, before and after the occurrence of weaknesses, is constructed, and the identification of the most severe weakness scenario is embedded in the optimization process, resulting in a two-stage robust control model framework for distribution network operators. Based on the modeling requirements of the pre-weakness stage in the two-stage robust control model framework, an operating cost model for the pre-weakness stage is obtained by establishing a cost structure that includes transaction costs with the main grid, transaction settlement items within the user's multi-load aggregation entity, and the power generation costs of the self-owned gas turbine. Based on the normal operating state corresponding to the operating cost model before the occurrence of the weak point, the set of physical operation constraints of the distribution network before the occurrence of the weak point is obtained by establishing node active power balance constraints, reactive power balance constraints, line power flow limit constraints and node voltage amplitude constraints. Based on the modeling requirements of the post-weak point stage in the two-stage robust control model framework, a comprehensive operating cost model for the post-weak point stage is obtained by establishing a comprehensive cost structure that includes main grid transaction costs, internal transaction settlement items, gas turbine power generation costs, and direct load control costs. Based on the direct load regulation introduced in the comprehensive operating cost model after the occurrence of the vulnerability, a refined set of operating constraints is obtained by establishing the active and reactive power balance relationship of user nodes, including distributed power output, electricity purchase and sale behavior and load reduction, and setting the upper and lower limits of gas turbine output. The operating cost model before the occurrence of a weak point, the physical operation constraint set of the distribution network, the comprehensive operating cost model after the occurrence of a weak point, and the refined operation constraint set are integrated into the two-stage robust control model framework to establish a two-stage robust control model led by the distribution network operator.

[0010] As a preferred embodiment of the multi-load linkage control method described in this invention, the enhanced operator control model includes: Based on the network reconfiguration requirements of the distribution network after the occurrence of weak points in system operation, by introducing normal operation state variables and active disconnection state variables of the lines, and establishing upper limit constraints on the number of line faults and logical correlation constraints between the two, a set of constraints for adjusting the power grid topology state is obtained. Based on the adjustable network structure defined by the power grid topology state adjustment constraint set, by constructing node active power balance equations, reactive power balance equations, node voltage limit constraints, branch power flow limit constraints, and coupling constraints between branch power flow and topology state, the distribution network physical operation constraint set for the stage after the occurrence of weak points is obtained. Based on the power balance requirements of the user access nodes of the multi-load aggregation main body in the physical operation constraint set of the distribution network, the refined power balance constraints of the user nodes are obtained by establishing the active and reactive power balance relationship that includes the output of distributed power sources, the main grid's power purchase and sale behavior and the direct load regulation. Based on the direct load regulation introduced in the refined power balance constraints of the user nodes, the feasible domain constraint of the direct load regulation is obtained by setting the lower limit of the regulation value to zero and the upper limit to the original load level of the corresponding user. The set of power grid topology state adjustment constraints, the set of distribution network physical operation constraints, the refined power balance constraints of user nodes and the feasible domain constraint of direct load regulation are then jointly embedded into the distribution network operator control model to obtain the enhanced operator control model.

[0011] As a preferred embodiment of the multi-load linkage control method described in this invention, the construction of a two-layer interactive optimization framework to form a two-stage robust two-layer optimization problem with integer variables includes: Based on the distributed resource allocation and transaction needs of multiple load aggregation entities in the distribution network, a user-side optimization model is obtained by constructing a single user's operation and transaction model. Based on the lower-level response characteristics represented by the user-side optimization model, an operator-side control model is obtained by constructing a multi-level load linkage control model led by the distribution network operator and embedding a two-stage robust optimization structure. Based on the decision variables and physical relationships of the user-side optimization model and the operator-side control model, a two-layer interactive optimization framework is constructed by using the power flow changes caused by the weak points as the coupling link, forming a two-stage robust two-layer optimization problem with integer variables.

[0012] As a preferred embodiment of the multi-load linkage control method of the present invention, wherein: the step of solving the optimization problem using a nested column constraint generation algorithm includes: Based on the initial two-level iterative counter, the initial optimal electricity purchase and sale decision variables on the user side are obtained by solving the multi-variable load aggregation subject user operation transaction model without penalty terms. Based on the initial optimal power purchase and sale decision variables on the user side, the operator's control decisions, including nodal electricity prices, direct load control instructions, and network topology adjustment strategies, are obtained by solving the robust control model dominated by the distribution network operator. Based on the market and control signals formed by the operator's regulation and decision-making, the updated user electricity consumption and transaction behavior scheme is obtained by solving the multi-load aggregation subject user operation transaction model with the introduction of a transaction consistency coordination mechanism. Based on the transaction volume deviation between the updated user behavior scheme and the operator's control decision, a convergence criterion for measuring the degree of inconsistency between upper and lower level decisions is obtained by calculating the transaction consistency penalty term value. Based on the comparison between the penalty term value and the preset convergence threshold, if the penalty term value is less than or equal to the preset convergence threshold, the algorithm is determined to have converged, and the final multi-load linkage control scheme is obtained. If the penalty term value is greater than the preset convergence threshold, a new iteration round is obtained by incrementing the two-layer iteration counter, and the process returns to the initialization step to continue execution, thus forming a closed-loop iteration process.

[0013] The beneficial effects of this preferred technical solution are that, through a closed-loop iterative solution process driven by nested column constraint generation and transaction consistency penalty mechanism, it effectively coordinates the decision-making interaction between multiple load aggregation entities and distribution network operators, and efficiently converges to a robust linkage control scheme for operational weaknesses and uncertainties while ensuring the feasibility of physical and market constraints.

[0014] As a preferred embodiment of the multi-load linkage control method described in this invention, the basic operating data and distribution network system parameters include: the power generation cost and capacity parameters of the small gas turbines configured by the multi-load aggregation main users; the installed capacity of distributed photovoltaics and the photovoltaic power generation resource coefficient; the energy storage charging and discharging efficiency, maximum charging and discharging power and upper and lower limits of state of charge; typical active and reactive load curves; distribution network line impedance parameters and node voltage limits.

[0015] Secondly, the present invention provides a multi-load linkage control system, comprising: The data acquisition module is used to acquire basic operating data and power distribution network system parameters; The operator robust control modeling module is used to construct a two-stage operation transaction model for multiple load aggregation subjects based on the acquired data, forming a lower-level user response model; The topology adjustability enhancement module is used to establish a two-stage robust control model led by the distribution network operator based on the user response model. The goal is to minimize the system operating costs and load control costs before and after the occurrence of the vulnerability. The direct load control amount is used as the decision variable in the stage after the occurrence of the vulnerability. The module also includes an embedded identification mechanism for the most severe vulnerability uncertainty scenario. A two-layer interactive coupling modeling module is used to introduce adjustable grid topology constraints into the distribution network operator control model to obtain an enhanced operator control model. Couple the user response model with the enhanced operator control model to construct a two-layer interactive optimization framework, forming a two-stage robust two-layer optimization problem with integer variables. The nested solution and scheme output module is used to solve the optimization problem using a nested column constraint generation algorithm, and output the optimal electricity price guidance strategy, direct load control amount and topology adjustment scheme as a general scheme for multi-element load linkage control facing the uncertainty of weak points.

[0016] Thirdly, the present invention provides an electronic device, comprising: Memory, used to store programs; A processor is configured to execute the computer-executable instructions, which, when executed by the processor, implement the steps of the multi-load linkage control method.

[0017] Fourthly, the present invention provides a computer-readable storage medium, comprising: when the program is executed by a processor, the step of implementing the multi-load linkage control method.

[0018] The beneficial effects of this invention are as follows: By constructing a two-stage operation and transaction model for multiple load aggregation main users, this invention achieves refined modeling of the entire process of user-side distributed resources, peer-to-peer transactions, and main grid power purchase and sales behavior before and after the occurrence of system weaknesses, thereby improving the economy and physical feasibility of the lower-level response model. By establishing a two-stage robust control model with topology adjustability led by distribution network operators and embedding a mechanism for identifying the most severe vulnerability uncertainty scenario, the system achieves coordinated optimization of network reconfiguration, direct load control, and electricity price guidance, thereby enhancing its proactive control capability in response to operational uncertainties. By coupling the user response model with the enhanced operator model, a two-stage robust bi-level optimization problem with integer variables is constructed. A nested column constraint generation algorithm combined with a transaction consistency penalty mechanism is used for closed-loop iterative solution, achieving efficient coordination and convergence of upper and lower level decisions. Finally, a comprehensive multi-level load linkage control scheme that balances security, economy, and feasibility is output. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the basic process of a multi-load linkage control method provided in one embodiment of the present invention.

[0020] Figure 2 This is an overall framework diagram of a multi-load linkage control method provided in one embodiment of the present invention. Detailed Implementation

[0021] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0022] Example 1, referring to Figure 1 As an embodiment of the present invention, a multi-load linkage control method is provided, comprising: S100: Acquire basic operating data and power distribution system parameters; S200: Based on the acquired data, construct a two-stage operation transaction model for multiple load aggregation entities and form a lower-level user response model; S300: Based on the user response model, a two-stage robust control model led by the distribution network operator is established with the goal of minimizing the system operation cost and load control cost before and after the occurrence of the vulnerability. The direct load control amount is used as the decision variable in the stage after the occurrence of the vulnerability, and the identification mechanism of the most severe vulnerability uncertainty scenario is embedded. S400: In the distribution network operator control model, an adjustable grid topology constraint is introduced to obtain an enhanced operator control model; S500: Couple the user response model with the enhanced operator control model to construct a two-layer interactive optimization framework, forming a two-stage robust two-layer optimization problem with integer variables; S600: It uses a nested column constraint generation algorithm to solve the optimization problem and outputs the optimal electricity price guidance strategy, direct load control amount and topology adjustment scheme as a general scheme for multi-level load linkage control facing the uncertainty of weak points.

[0023] It should be noted that existing technologies face a series of challenges during distribution network operation, including a lack of effective modeling mechanisms for uncertainties in system operation, traditional dispatching methods being mostly based on deterministic scenarios or simple stochastic models, making it difficult to accurately depict the worst-case operating state after a fault, resulting in insufficient robustness of control strategies; secondly, diverse distributed resources on the user side (such as gas turbines, photovoltaics, and energy storage) and flexible trading behaviors (such as point-to-point energy sharing and grid power purchase and sale) are not fully incorporated into a unified optimization framework, making it difficult for operators to accurately utilize load response capabilities; distribution network operators have limited control methods, usually relying only on electricity price signals or fixed network topology, lacking the ability to coordinate direct load control, dynamic topology adjustment, and market mechanisms; and the decision-making of upper and lower level entities (users and operators) is disconnected, lacking an effective two-level interaction and consistency coordination mechanism, resulting in problems such as large trading deviations, convergence difficulties, and physical infeasibility in the actual implementation of the formulated schemes, making it difficult to achieve efficient, coordinated, and feasible load control in the face of uncertainties in weak points.

[0024] Therefore, existing technologies, under the background of new power systems, lack collaborative modeling of the uncertainties between the participation of multiple load aggregation entities in the market and distribution network equipment failures. This makes it difficult to accurately characterize operational weaknesses caused by equipment failures and fails to effectively integrate user-side flexibility resources with grid reliability requirements. Consequently, dispatch decisions are significantly inadequate in addressing security risks under high-proportion renewable energy access. Through steps S100-S600, a two-stage user-side operation and trading model, an enhanced operator robust control model, and a two-layer interactive optimization framework are constructed. A nested column constraint generation algorithm is used for collaborative solving. This achieves efficient linkage between electricity price guidance, direct load control, and network topology adjustment under uncertainties in system operation weaknesses, significantly improving the security, economy, and collaborative control capabilities of the distribution network.

[0025] Example 2, refer to Figure 2 As one embodiment of the present invention, based on the previous embodiment, a multi-load linkage control method is provided, including: In the embodiments of this application, such as Figure 2 As shown, this is the overall framework of the invention, divided into two stages: "before system uncertainty and fault" and "after fault," reflecting the collaborative response mechanism between the Distribution Network Operator (DSO) and the users of the multi-load aggregation entity. Before the fault occurs, the DSO guides users to conduct electricity trading through the Price Management Platform (DLMP), enabling users to engage in peer-to-peer (P2P) energy sharing and interact with the main grid to achieve economically optimized operation under normal conditions. When a fault event (such as line tripping or equipment failure) causes a weakness in system operation, the DSO initiates an emergency control strategy. On the one hand, it continues to issue price signals through the DLMP to incentivize user response; on the other hand, it introduces Direct Load Control (DLC) instructions to proactively reduce the load of some users to maintain system safety and stability. At the same time, the user side retains P2P trading capabilities, forming a dual-track linkage mechanism of "market guidance + direct control," achieving efficient, flexible, and collaborative control of distributed resources under uncertain disturbances.

[0026] In this embodiment of the application, the basic operating data and distribution network system parameters in step S100 include: the power generation cost and capacity parameters of small gas turbines configured by the main users of the multi-load aggregation; the installed capacity of distributed photovoltaics and the photovoltaic power generation resource coefficient; the energy storage charging and discharging efficiency, maximum charging and discharging power and upper and lower limits of state of charge; typical active and reactive load curves; distribution network line impedance parameters and node voltage limits.

[0027] In this embodiment of the application, step S200, which involves constructing a two-stage operational transaction model for multiple load aggregation entities, includes: Based on the uncertainty characteristics of system operation weaknesses, a two-stage operation model framework for a single multi-dimensional load aggregation subject user is obtained by constructing a user optimization architecture that includes two stages: before and after the occurrence of the weakness. Based on the modeling requirements of the pre-vulnerability stage in the two-stage operation model framework, a user cost model for the pre-vulnerability stage is obtained by establishing a cost composition that includes the power generation cost of small gas turbines, the point-to-point energy transaction cost with other multi-load aggregate main users, and the electricity purchase and sale settlement cost with the distribution network operator. Based on the point-to-point energy transaction items included in the user cost model, by establishing the energy conservation relationship of the transaction volume and the total transaction volume aggregation constraint, and using corresponding dual variables to represent the transaction price, a set of consistency constraints for energy transactions among users of multiple load aggregation entities is obtained. Based on the normal operating state defined by the user cost model and the consistency constraint set, the active and reactive power balance equations that include distributed power output, energy storage charging and discharging, main grid power purchase and sale and point-to-point transactions are established. The corresponding binary logic variables are set for gas turbine output limit, photovoltaic power generation upper limit, energy storage charging and discharging power limit, state of charge dynamic range, power purchase and sale mutual exclusion constraint and energy storage charging and discharging mutual exclusion constraint. The user operation constraint set before the occurrence of the vulnerability is obtained. Based on the modeling requirements of the post-vulnerability stage in the two-stage operation model framework, a user cost model for the post-vulnerability stage is obtained by establishing expressions for generation cost, point-to-point transaction cost, and main grid power purchase and sale cost with the same structure as the previous stage but updated variables. Based on the direct load regulation introduced in the user cost model after the occurrence of the vulnerability, by establishing an active and reactive power balance equation that includes distributed resource output, transaction behavior and load reduction, and applying similar operating limits to gas turbines, photovoltaics and energy storage as in the previous stage, the user emergency operation constraint set for the stage after the occurrence of the vulnerability is obtained. Based on a two-stage operation model framework, user cost models before and after the occurrence of weak points, energy trading consistency constraint sets, user operation constraint sets, and user emergency operation constraint sets, a two-stage operation trading model for multi-load aggregation entities is constructed by integrating all cost items and constraints. The point-to-point energy trading price is endogenously represented by the dual variable of the energy conservation constraint, achieving a unified modeling of market clearing price and physical feasibility.

[0028] In this embodiment of the application, a complete two-stage operation and transaction model on the user side is formed through the above modeling steps. The model aims to minimize the total cost before and after the occurrence of the vulnerability. The decision variables cover distributed generation, peer-to-peer transactions, grid power purchase and sales, and energy storage scheduling. It also embeds physical operation constraints and market transaction consistency conditions to form the lower-level response model in the two-layer optimization framework.

[0029] In this embodiment of the application, the vulnerability uncertainty modeling method in step S200 automatically identifies the vulnerability scenario that has the most serious impact on the system operating cost among all possible line fault scenarios by embedding an inner search mechanism in the optimization model, and feeds the worst scenario back to the upper-level control decision, thereby achieving robust modeling and response to operational vulnerability uncertainty.

[0030] In an optional implementation, the weak point uncertainty modeling method in step S200 can also generate several typical line fault scenarios and their probabilities in advance based on historical data or probability models. In the two-stage scheduling framework, with the goal of minimizing the expected total cost, the original inner worst-case scenario search mechanism is replaced by solving a stochastic optimization problem that includes the weighted average of all scenarios.

[0031] In an alternative implementation, the weak point uncertainty modeling method in step S200 can also introduce a fuzzy set (such as the mean or support set constraint based on historical data) about the probability distribution of line faults into the two-stage framework, and change the original inner worst-case scenario search to maximizing the expected system cost within this fuzzy set, thereby solving the control strategy that is still robust to the most unfavorable situation among all possible distributions.

[0032] In this embodiment of the application, the modeling method for network topology adjustment in step S200 introduces binary variables to represent the switching status of lines, and combines the Big-M method to dynamically couple network topology changes with physical constraints such as power flow and voltage, thereby directly embedding the ability to actively reconstruct the distribution network topology after a fault into the optimization model.

[0033] In an optional implementation, the modeling method for network topology adjustment in step S200 can also pre-enumerate several feasible topologies that satisfy safety constraints such as radial operation, introduce discrete selection variables to select the optimal topology in the optimization, and embed its corresponding network parameters (such as admittance matrix and connectivity relationship) into power flow and operation constraints, thereby avoiding online dynamic modeling of line status.

[0034] In an optional implementation, the modeling method for network topology adjustment in step S200 can also be to train a GNN model outside the optimization process, take the current operating state and weak point information as input, output a set of high-potential feasible topology candidates, and embed these candidate topologies as constraints or variables into the original optimization problem to achieve intelligent reconstruction that integrates data-driven and physical model.

[0035] In this embodiment, the objective function of the two-stage transaction model on the user side is expressed as: in, The cost for a single, multi-load aggregation entity user before operational vulnerabilities arise. Costs incurred after a vulnerability arises in the operation of a single multi-load aggregator. Costs incurred by a single multi-load aggregator before the vulnerability arise include the small gas turbine power generation costs of the multi-load aggregator, transaction costs between the multi-load aggregator users, and transaction costs between the multi-load aggregator user and the distribution network operator. in, The unit power generation cost of a small gas turbine generator. For power generation, , These are the transaction prices between users in a multi-load aggregation entity, and the node marginal prices between users in a multi-load aggregation entity and distribution network operators. This refers to the transaction volume between multiple user entities that aggregate their workloads. , It aggregates the electricity purchased and sold by main users and distribution network operators for diverse loads.

[0036] In this embodiment of the application, based on the point-to-point energy transaction items included in the user cost model, a set of consistency constraints for energy transactions among users of multiple load aggregation entities is obtained by establishing a conservation relationship for the transaction electricity volume and a total transaction volume aggregation constraint, and by using corresponding dual variables to represent the transaction price.

[0037] Specifically, any two adjacent users and The peer-to-peer transactions between them satisfy the energy conservation constraint: in, Indicates user To users During the period The transaction volume (positive value for sold) is represented by the equation, which ensures that bidirectional transactions are opposites; its corresponding Lagrange multiplier... This refers to the endogenously generated peer-to-peer transaction price.

[0038] Furthermore, users During the period The total net power of peer-to-peer transactions is obtained by summing the transaction volumes of the peer-to-peer user with all its neighboring users: in, Indicates to users There exists a set of neighboring users connected through peer-to-peer transactions. These constraints collectively constitute the consistency constraint set for energy transactions between users.

[0039] In this embodiment, the user's constraint set includes active and reactive power balance equations to ensure energy conservation in each time period. Specifically, the active power balance constraint for user i in time period t is: Accordingly, the reactive power balance constraint only considers the reactive power support of the gas turbine and the load demand, and is expressed as: in, This aggregates the photovoltaic power generation of multiple load entities. , It aggregates the energy storage charging and discharging of multiple load main users. , Aggregating the active and reactive loads of multiple load groups for the main users, , It is a binary variable, representing the power purchase and sale status of multiple load aggregation entities, namely users and distribution network operators.

[0040] In this embodiment of the application, the following power generation operation constraints exist for distributed small gas turbines and distributed photovoltaics owned by multiple load aggregation main users: in, , This represents the maximum power generation of a small gas turbine. For photovoltaic installed capacity, The photovoltaic power generation resource coefficient is used to characterize the uncertainty of photovoltaic resources.

[0041] In this embodiment of the application, the following charging and discharging operation constraints exist for energy storage of a single multi-load aggregation main user: in, As a binary variable characterizing the charge and discharge state of energy storage, This represents the maximum charge and discharge power of the energy storage. The charge / discharge coefficient of energy storage. , These represent the maximum and minimum states of charge (SOC) of the energy storage. The initial SOC value is given by historical operating data obtained from S100, and the weak point scenario refers to the N-1 line fault set.

[0042] In this embodiment of the application, the total operating and transaction costs of user i, the subject of multi-load aggregation, during the emergency phase (i.e., the "post-event" phase) after a system operational vulnerability occurs are expressed as follows: For the operation of multiple load aggregation entities after the emergence of operational weaknesses, the constraints on active and reactive power also exist as follows: in, , This refers to the direct controllable quantities of active and reactive loads of the main users of multi-load aggregators. For distributed small gas turbines, distributed photovoltaic systems, and energy storage in the stage following the occurrence of operational weaknesses in the main users of multi-load aggregators, the constraints are similar to those in the stage before the occurrence of operational weaknesses, as follows: In this embodiment of the application, step S300, which establishes a two-stage robust control model led by the distribution network operator, includes: Based on the uncertainty characteristics of system operation weaknesses, an optimization architecture that includes two stages, before and after the occurrence of weaknesses, is constructed, and the identification of the most severe weakness scenario is embedded in the optimization process, resulting in a two-stage robust control model framework for distribution network operators. Based on the modeling requirements of the pre-weakness stage in the two-stage robust control model framework, an operating cost model for the pre-weakness stage is obtained by establishing a cost structure that includes transaction costs with the main grid, transaction settlement items within the user's multi-load aggregation entity, and the power generation costs of the self-owned gas turbine. Based on the normal operating state corresponding to the operating cost model before the occurrence of the weak point, the set of physical operation constraints of the distribution network before the occurrence of the weak point is obtained by establishing node active power balance constraints, reactive power balance constraints, line power flow limit constraints and node voltage amplitude constraints. Based on the modeling requirements of the post-weak point stage in the two-stage robust control model framework, a comprehensive operating cost model for the post-weak point stage is obtained by establishing a comprehensive cost structure that includes main grid transaction costs, internal transaction settlement items, gas turbine power generation costs, and direct load control costs. Based on the direct load regulation introduced in the comprehensive operating cost model after the occurrence of the vulnerability, a refined set of operating constraints is obtained by establishing the active and reactive power balance relationship of user nodes, including distributed power output, electricity purchase and sale behavior and load reduction, and setting the upper and lower limits of gas turbine output. The operational cost model before the occurrence of a weak point, the physical operation constraint set of the distribution network, the comprehensive operational cost model after the occurrence of a weak point, and the refined operation constraint set are integrated into the two-stage robust control model framework to establish a two-stage robust control model led by the distribution network operator. The most severe weak point scenario is obtained by solving an inner-layer maximization subproblem, which uses line fault state variables... ( Indicates the line The failure event is used as the decision variable, and the objective is to maximize the total operation and load regulation cost of the distribution network under a given user response strategy, thereby characterizing the worst-case operating state of the system under equipment failure uncertainty. This mechanism ensures that the generated control strategy is robust and feasible for all possible vulnerability scenarios.

[0043] In this embodiment, the objective function of the two-stage robust control model is: in, This refers to the operating costs incurred by distribution network operators before operational vulnerabilities emerge. The operating costs after the emergence of operational vulnerabilities. The cost of load regulation in the later stages is generated due to the weak points in operation.

[0044] For the stage before operational vulnerabilities emerge, the operating costs of distribution network operators include transaction costs between the operator and the main grid on behalf of the main users of the diversified load aggregation within the distribution network, transaction costs of the main users of the diversified load aggregation within the distribution network, and a portion of the gas turbine power generation costs belonging to the operator: in, , This refers to the transaction volume and price between the main users and the main network representing the diverse loads within the distribution network, as well as the transaction prices. , The electricity purchased and sold for transactions between multiple load aggregation entities within the distribution network and other users. For the transaction price, , This refers to the unit cost and power generation of gas turbine power generation. For distribution network operation, there are issues of active and reactive power balance at nodes and voltage constraints. in, , For the active and reactive power flow of the line, , For line resistance and reactance, For node voltage, , These are the upper and lower limits of the node voltage. For active and reactive power flow on a line, the following upper and lower limit constraints exist: in, , This represents the maximum active and reactive power flow of the line. For the normal stage without distributed generation resources, the following active and reactive power balance constraints exist at the nodes: For all distributed gas turbines owned by operators, there are upper and lower limits on power generation: In the later stages after operational vulnerabilities are identified, distribution network operators also consider the transaction costs between the operator representing the main users of the diversified load aggregation within the distribution network and the main grid, as well as the transaction costs of the main users of the diversified load aggregation within the distribution network and a portion of the gas turbine power generation costs belonging to the operator: For the stage following the emergence of operational weaknesses, the load control costs are considered as follows: in, This represents the load value of the node where the main user of the multi-load aggregation entity is located. This is for direct load regulation.

[0045] In this embodiment of the application, step S400, obtaining the enhanced operator control model, includes: Based on the network reconfiguration requirements of the distribution network after the occurrence of weak points in system operation, by introducing normal operation state variables and active disconnection state variables of the lines, and establishing upper limit constraints on the number of line faults and logical correlation constraints between the two, a set of constraints for adjusting the power grid topology state is obtained. Based on the adjustable network structure defined by the power grid topology state adjustment constraint set, by constructing node active power balance equations, reactive power balance equations, node voltage limit constraints, branch power flow limit constraints, and coupling constraints between branch power flow and topology state, the distribution network physical operation constraint set for the stage after the occurrence of weak points is obtained. Based on the power balance requirements of the user access nodes of the multi-load aggregation main body in the physical operation constraint set of the distribution network, the refined power balance constraints of the user nodes are obtained by establishing the active and reactive power balance relationship that includes the output of distributed power sources, the main grid's power purchase and sale behavior and the direct load regulation. Based on the direct load regulation introduced in the refined power balance constraints of the user nodes, the feasible domain constraint of the direct load regulation is obtained by setting the lower limit of the regulation value to zero and the upper limit to the original load level of the corresponding user. The set of power grid topology state adjustment constraints, the set of distribution network physical operation constraints, the refined power balance constraints of user nodes and the feasible domain constraint of direct load regulation are then jointly embedded into the distribution network operator control model to obtain the enhanced operator control model.

[0046] Due to uncertainties in the operation of the distribution network system, the following constraints exist for adjusting the power grid topology: in, For the status of the line, when When =1, it indicates that the line is in normal condition. This represents the number of line faults. This is the active opening and closing state of the line, when =1 indicates that the line is closed without interruption. Even after the emergence of operational weaknesses, the active and reactive power balance, as well as voltage and power flow constraints, still exist as follows: For the node where the main user of multi-load aggregation is located, the active and reactive power balance constraints after load regulation are as follows: For the load control of multiple load aggregation entities, there are upper and lower limit constraints as follows: In this embodiment of the application, step S500 constructs a two-layer interactive optimization framework to form a two-stage robust two-layer optimization problem with integer variables, including: Based on the distributed resource allocation and transaction needs of multiple load aggregation entities in the distribution network, a user-side optimization model is obtained by constructing a single user's operation and transaction model. Based on the lower-level response characteristics represented by the user-side optimization model, an operator-side control model is obtained by constructing a multi-level load linkage control model led by the distribution network operator and embedding a two-stage robust optimization structure. Based on the decision variables and physical relationships of the user-side optimization model and the operator-side control model, a two-layer interactive optimization framework is constructed using power flow changes caused by weaknesses as the coupling link, forming a two-stage robust two-layer optimization problem with integer variables. To ensure consistency between the upper and lower layers in electricity purchase and sale behavior, a linear penalty term based on transaction volume deviation is introduced, and bidirectional inequality constraints are applied to limit the feasible matching range between user-reported quantities and operator dispatch instructions, thereby gradually eliminating decision conflicts during the iteration process.

[0047] In this embodiment of the application, the coordination mechanism between the user side and the operator side in step S500 is achieved through a two-layer iterative mechanism: the operator issues scheduling instructions at the upper layer, and the user optimizes their own electricity consumption behavior based on the instructions at the lower layer, and quantifies their response deviation through a penalty term; the deviation is fed back to the operator layer to update the next round of instructions, and the process is iterated repeatedly until the decisions of the upper and lower layers tend to be consistent.

[0048] In an optional implementation, the coordination mechanism between the user side and the operator side in step S500 can also transform the original lower-level user optimization problem into a set of equivalent KKT optimality conditions under the premise of satisfying the assumptions of convexity and continuity, and directly embed them as constraints into the upper-level operator optimization model, thereby transforming the two-level problem into a single-level mathematical programming problem that can be solved in one go.

[0049] In an optional implementation, the coordination mechanism between the user side and the operator side in step S500 can also construct a decentralized trading platform, where the distribution network operator releases control requirements and incentive signals, and various multi-load aggregation entities autonomously submit response plans and participate in peer-to-peer energy trading through smart contracts. The system automatically executes settlement and resource scheduling through a consensus mechanism, achieving collaborative control without centralized optimization.

[0050] In this embodiment, to ensure consistency in electricity purchase and sale decisions between upper and lower layers, a linear penalty term based on transaction volume deviation is introduced, and a two-way inequality constraint is applied to limit the feasible matching range between user-reported quantities and operator dispatch instructions. Furthermore, by establishing a physical correspondence between user-side and operator-side variables, the unity of energy flow and settlement flow is clarified. Among them, represents the user connected to the distribution network node . The above equation indicates that: the electricity purchase quantity of the user from the distribution network operator is equal to the electricity sales quantity of the operator to the user, and the electricity sales quantity of the user to the operator is equal to the electricity purchase quantity of the operator from the user, thus ensuring the consistency of power balance and market settlement at the physical level.

[0051] To ensure the consistency of the upper and lower layer decisions in the electricity purchase and sales behavior, a transaction consistency penalty term is introduced in the user-side optimization model. Let the scheduling instruction given by the operator in the th round of iteration be , (that is, the optimal solution of the previous round, denoted as the reference value with "*"), then when the user performs optimization in the th round, its electricity purchase and sales decisions , should be as close as possible to this instruction. For this purpose, a penalty term is defined: Among them, etc. represent the electricity purchase and sales instructions obtained by the operator's optimization in the previous round (i.e., the iteration reference value), and is the penalty weight.

[0052] At the same time, to ensure physical feasibility, the following one-way matching constraint is imposed: The above constraint allows the user's declared quantity to be not less than the operator's expected value (to avoid "committing to power supply but not executing"), while the penalty term suppresses excessive deviation. The two together promote the convergence of the upper and lower layer decisions.

[0053] In the embodiment of this application, in step S600, a nested column constraint generation algorithm is used to solve the optimization problem, including: in view of the fact that the inner-layer worst-case scenario identification sub-problem contains integer variables such as topological states and charge-discharge logics, the strong duality theory fails, and it is impossible to transform the max-min problem into a single-layer model through the classical KKT conditions or dual transformation. Therefore, a nested column constraint generation (Nested CCG) framework needs to be adopted to be compatible with the mixed integer structure and ensure convergence; Based on the initialized double-layer iteration counter, by solving the multi-load aggregation main user operation transaction model without the penalty term, the initial optimal electricity purchase and sales decision variables on the user side are obtained; Based on the initial optimal electricity purchase and sales decision variables on the user side, by solving the robust regulation model dominated by the distribution network operator, the operator's regulation decisions including nodal electricity prices, direct load regulation instructions, and network topology adjustment strategies are obtained; Based on the market and control signals formed by the operator's regulation and decision-making, the updated user electricity consumption and transaction behavior scheme is obtained by solving the multi-load aggregation subject user operation transaction model with the introduction of a transaction consistency coordination mechanism. Based on the transaction volume deviation between the updated user behavior scheme and the operator's control decision, a convergence criterion for measuring the degree of inconsistency between upper and lower level decisions is obtained by calculating the transaction consistency penalty term value. Based on the comparison between the penalty term value and the preset convergence threshold, if the penalty term value is less than or equal to the preset convergence threshold, the algorithm is determined to have converged, and the final multi-load linkage control scheme is obtained. If the penalty term value is greater than the preset convergence threshold, a new iteration round is obtained by incrementing the two-layer iteration counter, and the process returns to the initialization step to continue execution, thus forming a closed-loop iteration process.

[0054] In this embodiment of the application, in step S600, direct load regulation (DLC) is modeled as a continuously adjustable reduction of user active and reactive loads. By introducing regulation variables and applying upper and lower limit constraints, it is incorporated into the two-stage robust optimization model as a flexible resource, and the unit regulation value is included in the total system cost to achieve fine-grained scheduling of load-side flexibility.

[0055] In an optional implementation, the modeling granularity of Direct Load Control (DLC) in step S600 can further divide the user-adjustable load into several preset operating levels (such as off, low, medium, and high levels), introduce integer or binary variables to represent the level selection, and embed the power reduction amount and comfort cost corresponding to each level into the optimization model, thereby characterizing the load response behavior in a discrete manner.

[0056] In an optional implementation, the modeling granularity of direct load regulation (DLC) in step S600 can also be achieved by the distribution network operator issuing price-based incentive signals. Users can construct a utility-maximizing response function based on their own electricity consumption utility and regulation costs, and embed this function as a lower-level behavioral model into a two-layer optimization framework, thereby endogenizing the load reduction amount in a demand-elastic manner.

[0057] In this embodiment, the nested column constraint generation (CCG) multivariate load regulation optimization solution framework starts from initialization and sets the number of iterations. Without introducing penalty terms, solve the distribution network user-level problem to obtain the user-side decision variables for electricity purchase, electricity sale, and reactive power. These results are then passed as input to the distribution network operator layer; within the Nested CCG loop, the operator layer optimization problem is solved, and its scheduling instructions are updated based on user responses (e.g., ...). Solve the user-layer problem again, but this time use the operator's scheduling instructions as a reference value and introduce a penalty term to coordinate the deviation between the upper and lower layers; determine the absolute value of the penalty term. Is it less than the preset convergence threshold? If the condition is met, the iteration terminates; otherwise, let... And then proceed to the next iteration; through this alternating solution and feedback mechanism, dynamic game and collaborative optimization between users and operators are realized, ultimately resulting in a robust and cost-effective control scheme.

[0058] Example 3 is an embodiment of the present invention. This embodiment differs from the first embodiment in that it provides a multi-load linkage control system.

[0059] It should be noted that the technical solution of this multi-load linkage control system and the technical solution of the above-mentioned multi-load linkage control method belong to the same concept. For details not described in detail in the technical solution of the multi-load linkage control system in this embodiment, please refer to the description of the technical solution of the above-mentioned multi-load linkage control method.

[0060] This embodiment provides a multi-load linkage control system, comprising: The data acquisition module is used to acquire basic operating data and power distribution network system parameters; The operator robust control modeling module is used to construct a two-stage operation transaction model for multiple load aggregation subjects based on the acquired data, forming a lower-level user response model; The topology adjustability enhancement module is used to establish a two-stage robust control model led by the distribution network operator based on the user response model. The goal is to minimize the system operating costs and load control costs before and after the occurrence of the vulnerability. The direct load control amount is used as the decision variable in the stage after the occurrence of the vulnerability. The module also includes an embedded identification mechanism for the most severe vulnerability uncertainty scenario. A two-layer interactive coupling modeling module is used to introduce adjustable grid topology constraints into the distribution network operator control model to obtain an enhanced operator control model. Couple the user response model with the enhanced operator control model to construct a two-layer interactive optimization framework, forming a two-stage robust two-layer optimization problem with integer variables. The nested solution and scheme output module is used to solve the optimization problem using a nested column constraint generation algorithm, and output the optimal electricity price guidance strategy, direct load control amount and topology adjustment scheme as a general scheme for multi-element load linkage control facing the uncertainty of weak points.

[0061] This embodiment also provides an electronic device applicable to a multi-load linkage control method, including: The system includes a memory and a processor. The memory stores computer-executable instructions, and the processor executes these instructions to implement a multi-load linkage control method as described in the above embodiments.

[0062] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements a multi-load linkage control method as proposed in the above embodiments.

[0063] The storage medium proposed in this embodiment and the method for implementing multi-load linkage control proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0064] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0065] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for multi-load linkage control, characterized in that, include: Obtain basic operational data and power distribution network system parameters; Based on the acquired data, a two-stage operational transaction model for multiple load aggregation entities is constructed to form a lower-level user response model. Based on the user response model, a two-stage robust control model led by the distribution network operator is established with the goal of minimizing the system operation cost and load control cost before and after the occurrence of the vulnerability. The direct load control amount is used as the decision variable in the stage after the occurrence of the vulnerability, and the identification mechanism of the most severe vulnerability uncertainty scenario is embedded. In the aforementioned distribution network operator control model, adjustable grid topology constraints are introduced to obtain an enhanced operator control model. Couple the user response model with the enhanced operator control model to construct a two-layer interactive optimization framework, forming a two-stage robust two-layer optimization problem with integer variables. The optimization problem is solved by a nested column constraint generation algorithm, which outputs the optimal electricity price guidance strategy, direct load control amount and topology adjustment scheme as a general scheme for multi-level load linkage control facing the uncertainty of weak points.

2. The multi-load linkage control method as described in claim 1, characterized in that: The two-stage operation transaction model for constructing multiple load aggregation entities includes: Based on the uncertainty characteristics of system operation weaknesses, a two-stage operation model framework for a single multi-dimensional load aggregation subject user is obtained by constructing a user optimization architecture that includes two stages: before and after the occurrence of the weakness. Based on the modeling requirements of the pre-vulnerability stage in the two-stage operation model framework, a user cost model for the pre-vulnerability stage is obtained by establishing a cost structure that includes the power generation cost of small gas turbines, the point-to-point energy transaction cost with other multi-load aggregation main users, and the electricity purchase and sale settlement cost with the distribution network operator. Based on the point-to-point energy transaction items included in the user cost model, by establishing the energy conservation relationship of the transaction volume and the total transaction volume aggregation constraint, and using corresponding dual variables to represent the transaction price, a set of consistency constraints for energy transactions among users of multiple load aggregation entities is obtained. Based on the normal operating state defined by the user cost model and the consistency constraint set, by establishing an active and reactive power balance equation that includes distributed power output, energy storage charging and discharging, main grid power purchase and sale and point-to-point transactions, and setting the gas turbine output limit, photovoltaic power generation upper limit, energy storage charging and discharging power limit, state of charge dynamic range, power purchase and sale mutual exclusion constraint and the binary logic variables corresponding to the energy storage charging and discharging mutual exclusion constraint, the user operation constraint set before the occurrence of the weakness point is obtained. Based on the modeling requirements of the post-vulnerability stage in the two-stage operation model framework, by establishing expressions for power generation cost, point-to-point transaction cost, and main grid power purchase and sale cost with the same structure as the previous stage but updated variables, a user cost model for the post-vulnerability stage is obtained. Based on the direct load regulation introduced in the user cost model after the occurrence of the vulnerability, by establishing an active and reactive power balance equation that includes distributed resource output, transaction behavior and load reduction, and applying similar operating limits to gas turbines, photovoltaics and energy storage as in the previous stage, the user emergency operation constraint set for the stage after the occurrence of the vulnerability is obtained. Based on the aforementioned two-stage operation model framework, user cost model before and after the occurrence of weak points, energy transaction consistency constraint set, user operation constraint set, and user emergency operation constraint set, a two-stage operation transaction model for multi-load aggregation subject users is constructed by integrating all cost items and constraints.

3. The multi-load linkage control method as described in claim 1 or 2, characterized in that: The establishment of a two-stage robust control model led by the distribution network operator includes: Based on the uncertainty characteristics of system operation weaknesses, an optimization architecture that includes two stages, before and after the occurrence of weaknesses, is constructed, and the identification of the most severe weakness scenario is embedded in the optimization process, resulting in a two-stage robust control model framework for distribution network operators. Based on the modeling requirements of the pre-weakness stage in the two-stage robust control model framework, an operating cost model for the pre-weakness stage is obtained by establishing a cost structure that includes transaction costs with the main grid, transaction settlement items within the user's multi-load aggregation entity, and the power generation costs of the self-owned gas turbine. Based on the normal operating state corresponding to the operating cost model before the occurrence of the weak point, the set of physical operation constraints of the distribution network before the occurrence of the weak point is obtained by establishing node active power balance constraints, reactive power balance constraints, line power flow limit constraints and node voltage amplitude constraints. Based on the modeling requirements of the post-weak point stage in the two-stage robust control model framework, a comprehensive operating cost model for the post-weak point stage is obtained by establishing a comprehensive cost structure that includes main grid transaction costs, internal transaction settlement items, gas turbine power generation costs, and direct load control costs. Based on the direct load regulation introduced in the comprehensive operating cost model after the occurrence of the vulnerability, a refined set of operating constraints is obtained by establishing the active and reactive power balance relationship of user nodes, including distributed power output, electricity purchase and sale behavior and load reduction, and setting the upper and lower limits of gas turbine output. The operating cost model before the occurrence of a weak point, the physical operation constraint set of the distribution network, the comprehensive operating cost model after the occurrence of a weak point, and the refined operation constraint set are integrated into the two-stage robust control model framework to establish a two-stage robust control model led by the distribution network operator.

4. The multi-load linkage control method as described in claim 3, characterized in that: The enhanced operator control model includes: Based on the network reconfiguration requirements of the distribution network after the occurrence of weak points in system operation, by introducing normal operation state variables and active disconnection state variables of the lines, and establishing upper limit constraints on the number of line faults and logical correlation constraints between the two, a set of constraints for adjusting the power grid topology state is obtained. Based on the adjustable network structure defined by the power grid topology state adjustment constraint set, by constructing node active power balance equations, reactive power balance equations, node voltage limit constraints, branch power flow limit constraints, and coupling constraints between branch power flow and topology state, the distribution network physical operation constraint set for the stage after the occurrence of weak points is obtained. Based on the power balance requirements of the user access nodes of the multi-load aggregation main body in the physical operation constraint set of the distribution network, the refined power balance constraints of the user nodes are obtained by establishing the active and reactive power balance relationship that includes the output of distributed power sources, the main grid's power purchase and sale behavior and the direct load regulation. Based on the direct load regulation introduced in the refined power balance constraints of the user nodes, the feasible domain constraint of the direct load regulation is obtained by setting the lower limit of the regulation value to zero and the upper limit to the original load level of the corresponding user. The set of power grid topology state adjustment constraints, the set of distribution network physical operation constraints, the refined power balance constraints of user nodes and the feasible domain constraint of direct load regulation are then jointly embedded into the distribution network operator control model to obtain the enhanced operator control model.

5. The multi-load linkage control method as described in claim 4, characterized in that: The construction of the two-layer interactive optimization framework forms a two-stage robust two-layer optimization problem with integer variables, including: Based on the distributed resource allocation and transaction needs of multiple load aggregation entities in the distribution network, a user-side optimization model is obtained by constructing a single user's operation and transaction model. Based on the lower-level response characteristics represented by the user-side optimization model, an operator-side control model is obtained by constructing a multi-level load linkage control model led by the distribution network operator and embedding a two-stage robust optimization structure. Based on the decision variables and physical relationships of the user-side optimization model and the operator-side control model, a two-layer interactive optimization framework is constructed by using the power flow changes caused by the weak points as the coupling link, forming a two-stage robust two-layer optimization problem with integer variables.

6. The multi-load linkage control method as described in claim 5, characterized in that: The method of solving the optimization problem using a nested column constraint generation algorithm includes: Based on the initial two-level iterative counter, the initial optimal electricity purchase and sale decision variables on the user side are obtained by solving the multi-variable load aggregation subject user operation transaction model without penalty terms. Based on the initial optimal power purchase and sale decision variables on the user side, the operator's control decisions, including nodal electricity prices, direct load control instructions, and network topology adjustment strategies, are obtained by solving the robust control model dominated by the distribution network operator. Based on the market and control signals formed by the operator's regulation and decision-making, the updated user electricity consumption and transaction behavior scheme is obtained by solving the multi-load aggregation subject user operation transaction model with the introduction of a transaction consistency coordination mechanism. Based on the transaction volume deviation between the updated user behavior scheme and the operator's control decision, a convergence criterion for measuring the degree of inconsistency between upper and lower level decisions is obtained by calculating the transaction consistency penalty term value. Based on the comparison between the penalty term value and the preset convergence threshold, if the penalty term value is less than or equal to the preset convergence threshold, the algorithm is determined to have converged, and the final multi-load linkage control scheme is obtained. If the penalty term value is greater than the preset convergence threshold, a new iteration round is obtained by incrementing the two-layer iteration counter, and the process returns to the initialization step to continue execution, thus forming a closed-loop iteration process.

7. The multi-load linkage control method as described in claim 6, characterized in that: The basic operating data and distribution network system parameters include: the power generation cost and capacity parameters of small gas turbines configured for multi-load aggregation main users; the installed capacity and photovoltaic power generation resource coefficient of distributed photovoltaics; the energy storage charging and discharging efficiency, maximum charging and discharging power and upper and lower limits of state of charge; typical active and reactive load curves; distribution network line impedance parameters and node voltage limits.

8. A multi-load linkage control system, using the method described in any one of claims 1-7, characterized in that, include: The data acquisition module is used to acquire basic operating data and power distribution network system parameters; The operator robust control modeling module is used to construct a two-stage operation transaction model for multiple load aggregation subjects based on the acquired data, forming a lower-level user response model; The topology adjustability enhancement module is used to establish a two-stage robust control model led by the distribution network operator based on the user response model. The goal is to minimize the system operating costs and load control costs before and after the occurrence of the vulnerability. The direct load control amount is used as the decision variable in the stage after the occurrence of the vulnerability. The module also includes an embedded identification mechanism for the most severe vulnerability uncertainty scenario. A two-layer interactive coupling modeling module is used to introduce adjustable grid topology constraints into the distribution network operator control model to obtain an enhanced operator control model. Couple the user response model with the enhanced operator control model to construct a two-layer interactive optimization framework, forming a two-stage robust two-layer optimization problem with integer variables. The nested solution and scheme output module is used to solve the optimization problem using a nested column constraint generation algorithm, and output the optimal electricity price guidance strategy, direct load control amount and topology adjustment scheme as a general scheme for multi-element load linkage control facing the uncertainty of weak points.

9. An electronic device, characterized in that, include: Memory, used to store programs; A processor for loading the program to perform the steps of the method as claimed in any one of claims 1-7.

10. A computer-readable storage medium storing a program, characterized in that, When the program is executed by a processor, it implements the steps of the method as described in any one of claims 1-7.

Images