Layered optimization method, system and equipment for power distribution network and medium

Abstract

The invention relates to a hierarchical optimization method, system and device for a power distribution network and a medium. The method comprises the following steps: adjusting operation parameters of power equipment based on a local unit operation boundary instruction to obtain a local control parameter set; based on the local control parameter set and the local unit operation boundary instruction, calculating the adjustable capability of the power equipment to obtain a local unit adjustable capability report; based on a region control target instruction and the local unit adjustable capability report, calculating control set values of all units in a control region to obtain a local unit operation boundary instruction; based on the local unit operation boundary instruction, calculating the total adjustable capability of the control region in the next time period, and obtaining a region aggregation adjustable capability report; and performing global centralized optimization based on the regional aggregation adjustable capability reports of all the control regions to obtain the regional control target instruction. By adopting the method, the power distribution network can be optimized in a layered manner, and the safety is improved.

Description

Technical Field

[0001] This invention belongs to the field of intelligent optimization decision-making, and in particular relates to a hierarchical optimization method, system, equipment and medium for power distribution networks. Background Technology

[0002] In traditional power distribution network optimization techniques, optimization instructions strictly follow a serial processing flow: first, the dispatch center completes global optimization based on ultra-short-term forecasts and issues power targets to regional controllers; the regional controllers then decompose the instructions, allocating regional targets to each local unit; finally, the local units execute the set values ​​and report the status. This "top-down planning, bottom-up execution" approach relies on the authority of the central computation and the time-sequential connection of tasks at each level. However, the current serial processing method suffers from significant real-time bottlenecks and coordination failures. Because optimization calculations at each level are strictly performed sequentially, computational delays at any level accumulate and amplify in the serial link, leading to a delayed response across the entire network. In complex scenarios such as high-proportion renewable energy integration and rapid fluctuations in electric vehicle loads, the speed of grid status changes far exceeds the computation cycle of serial optimization, and risks such as voltage exceeding limits and line overloads cannot be suppressed in a timely manner. More seriously, traditional methods are unable to handle the tight coupling between levels. High-level optimization lacks accurate perception of the real-time status of the lower levels, while the autonomy of the lower levels is limited by the rigid constraints of high-level instructions. This can easily lead to a chain of contradictions, such as over-adjustment of the high-voltage level, difficulty in balancing the medium-voltage level, and control mismatch of the low-voltage level, ultimately threatening the safety and power quality of the distribution network. Summary of the Invention

[0003] Therefore, it is necessary to provide a method, system, equipment, and medium for hierarchical optimization of distribution networks to avoid various risks in distribution networks through parallel and rapid optimization, in order to address the above-mentioned technical problems.

[0004] Firstly, this application provides a hierarchical optimization method for a distribution network, applied to a distribution network system including first-level nodes, second-level nodes, and third-level nodes. The first-level nodes and second-level nodes are connected; the second-level nodes and third-level nodes are connected, including:

[0005] The first-level node receives the local unit operation boundary command; and based on the local unit operation boundary command, adjusts the operating parameters of the power equipment to obtain the local control parameter set;

[0006] The first-level node calculates the adjustable capability of the power equipment within the operating boundary based on the local control parameter set and the local unit operating boundary command, and obtains the local unit adjustable capability report.

[0007] The second-level node calculates the control setpoints of all units within the control area based on the regional control target command and the local unit adjustable capability report, and obtains the local unit operating boundary command.

[0008] The second-layer nodes calculate the total adjustable capacity of the control area in the next time period based on the local unit operation boundary instructions, and obtain the regional aggregate adjustable capacity report.

[0009] The third-layer nodes perform global centralized optimization based on the regional aggregation adjustable capability reports of all control areas to obtain regional control target instructions; the regional control target instructions are then sent to each second-layer node.

[0010] Furthermore, based on the regional aggregate adjustable capability reports of all control areas, global centralized optimization is performed to obtain regional control target instructions, including:

[0011] Based on the regional aggregated adjustable capacity report, the adjustable capacity of each control area is mapped to the corresponding node of the preset power grid model to generate a distribution map of adjustable resources across the entire network.

[0012] Starting from the current operating state of the power grid model, we infer the circuit problems that occur in the baseline scenario and obtain the baseline scenario analysis report; the baseline scenario represents the state of the control area that is not optimized.

[0013] Based on the distribution map of adjustable resources across the entire network, a security-constrained economic scheduling problem is constructed, leading to an optimization problem.

[0014] Based on the baseline scenario analysis report, the mathematical programming solver is invoked to solve the optimization problem and obtain the original optimization solution;

[0015] The original optimized solution is converted into control commands that can be issued, thus obtaining the area control target commands.

[0016] Furthermore, based on the distribution map of adjustable resources across the entire network, a security-constrained economic scheduling problem is constructed, resulting in optimization problems, including:

[0017] Based on the distribution map of adjustable resources across the entire network, the variables that the optimization algorithm needs to solve are identified, and the set of decision variable definitions is obtained;

[0018] Based on the defined set of decision variables and market information, the mathematical expression of the objective function is constructed using the following formula:

[0019]

[0020] in, The objective function is expressed mathematically. Let be the number of distributed power sources, and i be the index of the distributed power source. For the power generation cost function, To contribute one's efforts For equivalent electricity price, Active power This is the network loss cost coefficient. For total active network loss, The number of load aggregators participating in demand response, where j is the index of the demand response resource. Compensation price per unit, To reduce active power load;

[0021] Based on the set of decision variable definitions and the predicted load data, constraints are constructed to obtain a set of constraints.

[0022] The objective function mathematical expression and the set of constraints are combined into an optimization problem.

[0023] Furthermore, taking the current operating state as the starting point, the circuit problems that occur in the benchmark scenario are inferred, resulting in a benchmark scenario analysis report, including:

[0024] Use the current running state as the initial point; and freeze the initial point to obtain the initial conditions of the baseline scene.

[0025] The load data is mapped to specific grid nodes in the grid model to obtain the power data table;

[0026] Based on the initial conditions of the baseline scenario and the power data table, the power flow of the power grid over a future period is simulated through time-series power flow calculation, and the time-series power flow results are obtained.

[0027] By comparing the time-series power flow results with the equipment operating limitations, a list of potential security issues is obtained;

[0028] Based on the operating points of the time-series power flow results, sensitivity parameters and linearization parameters are calculated to obtain the power grid linearization parameter set; and the power grid linearization parameter set and the list of potential safety issues are integrated to obtain a baseline scenario analysis report.

[0029] Furthermore, the area control target command and local unit adjustable capability report calculate the control setpoints of all units within the control area to obtain the local unit operating boundary commands, including:

[0030] Boundary constraints are defined based on regional control target instructions; and the adjustment range is defined based on local unit adjustability reports.

[0031] Based on boundary constraints and adjustment range, an optimization proposition is constructed and solved to obtain an initial optimization solution.

[0032] The feasibility of the initial optimized solution is verified, and the local cell operation boundary instructions are obtained.

[0033] Furthermore, by solving the optimization problem, we obtain the initial optimal solution, including:

[0034] The optimization proposition is transformed into a standard quadratic programming form, resulting in a quadratic programming problem.

[0035] The quadratic programming problem is iteratively calculated using the interior point method to obtain iterative process data.

[0036] When the iteration process data meets the iteration stopping condition, the iteration calculation stops; and the last iteration process data is determined as the initial optimization solution; the iteration stopping condition includes the original feasibility requirement, the dual feasibility requirement, and the dual gap requirement.

[0037] Furthermore, based on the local control parameter set and local unit operating boundary commands, the adjustability of the power equipment within the operating boundary is calculated, resulting in a local unit adjustability report, including:

[0038] Generate the expected operating state diagram based on the local control parameter set;

[0039] Based on the expected operating state diagram and the local unit operating boundary instructions, the adjustable power of each controllable unit is calculated to obtain a detailed table of adjustable capabilities.

[0040] Based on the adjustable capacity details table, the active power and reactive power are aggregated to obtain the basic adjustable capacity;

[0041] Based on internal power grid constraints, the basic adjustable capability is verified for risks, resulting in a local unit adjustable capability report.

[0042] Secondly, this application also provides a hierarchical optimization system for a distribution network, comprising:

[0043] The local execution module is used to receive local unit operation boundary instructions and adjust the operating parameters of the power equipment based on the local unit operation boundary instructions to obtain a local control parameter set.

[0044] The local adjustable module is used to calculate the adjustable capability of power equipment within the operating boundary based on the local control parameter set and the local unit operating boundary command, and to obtain a local unit adjustable capability report.

[0045] The area optimization module is used to calculate the control setpoints of all units within the control area based on the area control target command and the local unit adjustable capability report, and to obtain the local unit operating boundary command.

[0046] The regional adjustable module is used by the second-layer nodes to run boundary instructions based on local units, calculate the total adjustable capacity of the control area in the next time period, and obtain a regional aggregate adjustable capacity report.

[0047] The central optimization module is used by the third-layer nodes to perform global centralized optimization based on the regional aggregate adjustable capability reports of all control areas, and obtain regional control target instructions; the regional control target instructions are used to send them to each second-layer node.

[0048] Thirdly, this application also provides a computer device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement any step of the method provided in the first aspect of this application.

[0049] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any step of the method provided in the first aspect of this application.

[0050] The aforementioned hierarchical optimization method, system, equipment, and medium for a power distribution network comprises the following steps: First-layer nodes receive local unit operating boundary instructions and adjust the operating parameters of power equipment based on these instructions to obtain a local control parameter set. Based on the local control parameter set and the local unit operating boundary instructions, the first-layer nodes calculate the adjustable capacity of the power equipment within the operating boundary, obtaining a local unit adjustable capacity report. Second-layer nodes, based on regional control target instructions and the local unit adjustable capacity reports, calculate the control setpoints of all units within the control area, obtaining local unit operating boundary instructions. Based on these instructions, the second-layer nodes calculate the total adjustable capacity of the control area in the next time period, obtaining a regional aggregate adjustable capacity report. Third-layer nodes perform global centralized optimization based on the regional aggregate adjustable capacity reports of all control areas, obtaining regional control target instructions. These regional control target instructions are then distributed to each second-layer node. By predicting parallel planning, serial waiting is transformed into parallel preparation, which greatly reduces response time. By using boundary management to achieve hierarchical decoupling, even if the central layer has not completed its calculation in the current cycle, the regional layer and the local layer can still perform autonomous operation based on the boundary instructions of the previous cycle, thereby improving reliability. The control objectives of each layer are very clear, avoiding conflicts between upper and lower level controls. Attached Figure Description

[0051] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0052] Figure 1 This is a schematic diagram of a distribution network hierarchical optimization method provided in an embodiment of the present invention;

[0053] Figure 2 This is a schematic diagram of the structure of a distribution network hierarchical optimization system provided in an embodiment of the present invention. Detailed Implementation

[0054] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0055] In one embodiment, such as Figure 1 As shown, a hierarchical optimization method for a distribution network is provided. This embodiment illustrates the application of this method to a distribution network system, which includes first-layer nodes, second-layer nodes, and third-layer nodes. The first-layer nodes can also be understood as the local layer, the second-layer nodes as the regional layer, and the third-layer nodes as the centralized layer. The first-layer nodes and second-layer nodes are connected; the second-layer nodes and third-layer nodes are also connected. It is understood that this method can also be applied to systems including electronic device terminals and a central server, and is implemented through the interaction between the electronic device terminals and the central server. In this embodiment, the method includes the following steps:

[0056] Step 101: The first-level node receives the local unit operation boundary command; and based on the local unit operation boundary command, adjusts the operating parameters of the power equipment to obtain the local control parameter set.

[0057] In this system, the first-level node refers to the lowest-level control unit in the distribution network, directly connected to and controlling specific power equipment, including distributed photovoltaic inverters, energy storage converters, and controllable loads, and is the final executor of commands. The local unit operating boundary command is a control command issued from the second-level node to the first-level node, specifying the upper and lower limits of the permitted operation of the power equipment within the local unit in the next time period, ensuring its operation within a safe range that conforms to the global optimization objective. Power equipment refers to specific physical devices, including generators and energy storage batteries. Operating parameters refer to the specific control setpoints of the power equipment. The local control parameter set is a specific, executable set of setpoints obtained by the first-level node after preliminary adjustments to the operating parameters of the power equipment it controls based on the received local unit operating boundary command, ensuring that the equipment is currently operating within the range allowed by the boundary command. The first-level node obtains the local unit's operating boundary instructions from the upper layer. The internal controller parses the instructions and converts them into direct control commands for one or more power devices. For example, the boundary instruction requires the active power output to be between 10kW and 50kW, while the current output of the device is 5kW. The controller will then adjust the operating parameters to a value not lower than 10kW, which is a specific, real-time device-level control.

[0058] Step 102: The first-level node calculates the adjustable capability of the power equipment within the operating boundary based on the local control parameter set and the local unit operating boundary command, and obtains the local unit adjustable capability report.

[0059] Specifically, the local control parameter set serves as the baseline for the current operating parameters of the equipment. Local unit operating boundary instructions define the upper and lower limits of operation. Adjustability refers to the ability of a power device to increase or decrease power relative to its current operating point within the range specified by the local unit operating boundary instructions, including both active and reactive power adjustable capabilities. The local unit adjustable capability report is a data report generated by the first-level node and prepared for submission to the second-level node. It details the potential for upward and downward adjustment of the unit in the next time period, while adhering to the given operating boundaries, serving as the foundational information for upper-level optimization decisions. The first-level node uses the current local control parameter set as the baseline and the local unit operating boundary instructions as the constraint range to calculate the adjustment potential of each controllable device. For example, if an energy storage device is currently charging at 20kW (kilowatt), and the boundary instructions allow a charging power range of 0kW to 30kW, then the upward adjustment capability is 20kW, and the downward adjustment capability is 10kW. The calculated adjustment capabilities are summarized, formatted, and used to generate the local unit adjustable capability report.

[0060] Step 103: The second-layer node calculates the control setpoints of all units within the control area based on the regional control target command and the local unit adjustable capability report, and obtains the local unit operating boundary command.

[0061] Specifically, the second-level node is the area controller in the distribution network, managing a control area containing multiple first-level nodes. The area control target instruction, issued by the third-level node to the second-level node, specifies the total task that the control area needs to complete in the next time period; it is a relatively macro-level target for the entire area. The local unit adjustability report comes from all first-level nodes under the jurisdiction of the second-level node. The local unit operating boundary instruction is a personalized operating boundary instruction calculated by the second-level node for each first-level node within its jurisdiction; it is the direct source of the instructions received by the first-level nodes. The second-level node receives the area control target instruction from the third level and collects the local unit adjustability reports from all first-level nodes, solving an optimization problem: how to fairly, economically, and safely decompose and distribute the total area task assigned by the superior to each subordinate first-level node. During the allocation, the capability range of each node must be fully considered, generating a unique and feasible local unit operating boundary instruction for each first-level node.

[0062] Step 104: The second-layer node calculates the total adjustable capacity of the control area in the next time period based on the local unit running boundary instructions, and obtains the regional aggregate adjustable capacity report.

[0063] Specifically, the local unit operating boundary command is the command just assigned to each first-level node. The control area is the power grid region comprised of all first-level nodes managed by the same second-level node. Total adjustable capacity is the collective regulating capacity of the entire control area under a given allocation scheme, considering inter-unit coordination and internal area constraints. The regional aggregate adjustable capacity report is a data report generated by the second-level node and prepared for submission to the third-level node. It describes how much power the entire area, as a whole, can adjust upwards or downwards in the next time period after the node has assigned tasks to its subordinate units; it is a crucial input for global optimization. Based on the recently issued local unit operating boundary command, the second-level node calculates the aggregate adjustable capacity of the entire area. At the new operating point, it calculates the net regulating capacity that the entire area can provide after satisfying internal safety constraints, and summarizes the calculation results into the regional aggregate adjustable capacity report.

[0064] Step 105: The third-layer node performs global centralized optimization based on the regional aggregation adjustable capability reports of all control areas to obtain the regional control target instructions; the regional control target instructions are used to send them to each second-layer node.

[0065] The third-level node is the highest-level control center of the distribution network, responsible for the overall optimization of the entire power grid. The regional aggregated adjustable capacity reports for all control areas originate from reports from all second-level nodes under the jurisdiction of the third-level node. Global centralized optimization is the core operation performed by the third-level node. Starting from the safety and economy of the entire power grid, it establishes a complex mathematical model to solve for the optimal operating point of the entire network. Regional control target instructions are the control targets set for the next time period for each control area after the third-level node performs global optimization calculations; they are the direct source of instructions received by the second-level nodes. The third-level node collects regional aggregated adjustable capacity reports from all second-level nodes. These reports collectively depict the distribution map of adjustable resources across the entire network. Based on the network model, predicted load, and electricity price information, it solves a large-scale optimization problem with the goal of minimizing the total operating cost and network losses, while ensuring that all lines are not overloaded and voltage does not exceed limits. It determines the optimal operating state that each area should achieve, thereby generating regional control target instructions sent to each second-level node.

[0066] This embodiment provides a hierarchical optimization method for a distribution network. First-layer nodes receive local unit operating boundary instructions and adjust the operating parameters of power equipment based on these instructions to obtain a local control parameter set. Based on the local control parameter set and the local unit operating boundary instructions, the first-layer nodes calculate the adjustable capacity of the power equipment within the operating boundary, obtaining a local unit adjustable capacity report. Second-layer nodes calculate the control setpoints of all units within the control area based on the regional control target instructions and the local unit adjustable capacity reports, obtaining local unit operating boundary instructions. Based on the local unit operating boundary instructions, the second-layer nodes calculate the total adjustable capacity of the control area in the next time period, obtaining a regional aggregated adjustable capacity report. Third-layer nodes perform global centralized optimization based on the regional aggregated adjustable capacity reports of all control areas, obtaining a regional control target instruction. The regional control target instruction is then distributed to each second-layer node. Through the above methods, we can predict parallel planning, transform serial waiting into parallel preparation, greatly compressing response time. By using boundary management, we can achieve hierarchical decoupling. Even if the central layer has not completed its calculation in the current cycle, the regional layer and the local layer can still perform autonomous operation based on the boundary instructions of the previous cycle, improving reliability. The control objectives of each layer are very clear, avoiding conflicts between upper and lower level controls.

[0067] In one embodiment, based on the regional aggregate adjustable capability reports of all control regions, global centralized optimization is performed to obtain regional control target instructions, including:

[0068] Step 201: Based on the regional aggregated adjustable capacity report, map the adjustable capacity of each control area to the corresponding node of the preset power grid model to generate a distribution map of adjustable resources across the entire network.

[0069] The regional aggregated adjustable capacity report is a data report submitted by each second-level node, describing the range of power generation or consumption capacity that the entire jurisdiction can increase or decrease in the next time period. The power grid model is a mathematical model used for computer simulation, representing the physical characteristics of the actual distribution network, such as topology, line parameters, and transformer parameters. The nationwide adjustable resource distribution map is a data set that maps abstract adjustable capacity values ​​to specific physical nodes in the power grid model, clearly indicating where, how much, and what type of adjustable capacity is available across the entire network. Third-level nodes read all regional aggregated adjustable capacity reports and, according to preset mapping relationships, allocate this capacity data to the corresponding nodes in the power grid model under their jurisdiction. For example, the 50kW increase in adjustable capacity from the regional A report is assigned to the node in the power grid model representing the photovoltaic power station cluster in that region.

[0070] Step 202: Taking the current operating state of the power grid model as the starting point, reason out the circuit problems that occur in the baseline scenario and obtain the baseline scenario analysis report; the baseline scenario represents the state of the control area that is not optimized.

[0071] Specifically, the current operating state refers to the real-time operating parameters of the power grid at this very moment, including the voltage of each node and the power flow of each branch. The initial point is the starting state set during computer simulation; in this embodiment, it refers to the current operating state of the power grid. The baseline scenario is a hypothetical scenario that simulates the future development of the power grid over a period of time without issuing optimization instructions to each region, i.e., keeping the current control strategy unchanged. Circuit problems refer to potential safety hazards that may occur during power grid operation, exemplified by line power exceeding thermal stability limits and node voltage exceeding safe ranges. The baseline scenario analysis report contains two core components: a list of predicted potential safety problems and linearized parameters such as power grid sensitivity calculated based on the current operating point, used to accelerate subsequent optimization solutions. The third-layer nodes perform power flow calculation simulation, using the current operating state as the simulation starting point and freezing optimization control. Predicted future load data is input into the power grid model, and through time-series power flow calculation, the evolution of the power grid over a period of time is simulated. The simulation results are compared with the safe operating limits of the equipment to identify all possible limit violations and record the time and location of occurrence.

[0072] Step 203: Based on the distribution map of adjustable resources across the entire network, construct a security-constrained economic scheduling problem to obtain an optimization problem.

[0073] Specifically, the safety-constrained economic dispatch problem is a specific mathematical optimization problem. Its goal is to find the most economical dispatch scheme while satisfying all power grid safety operation constraints. The optimization problem consists of an objective function and a set of constraints, forming a standardized mathematical model that can be recognized and processed by a mathematical solver. The third-level nodes, based on the distribution map of adjustable resources across the entire network, determine the variables to be optimized. The objective function is constructed with the goal of minimizing total operating cost. The physical laws of the power grid, equipment safety limits, and risk points identified in the baseline scenario are collectively used to construct the constraint set. This transforms the complex power grid dispatch decision problem into a clearly structured and computable mathematical optimization problem.

[0074] Step 204: Based on the baseline scenario analysis report, call the mathematical programming solver to solve the optimization problem and obtain the original optimization solution.

[0075] Among them, the mathematical programming solver is a specialized software tool used to solve various mathematical optimization problems. The original optimization solution refers to the theoretical optimal solution directly calculated by the mathematical solver without engineering processing; it is a set of continuous values, which can be the power setpoint of each control device accurate to the decimal point. The third-level node calls the mathematical programming solver, inputting the optimization problem model constructed in the previous step into the solver. The solver uses the linearization parameters provided in the baseline scenario analysis report as initial guidance and performs iterative calculations using algorithms such as the interior-point method to ultimately find a solution that mathematically satisfies all constraints and optimizes the objective function.

[0076] Step 205: Convert the original optimized solution into control commands that can be issued to obtain the area control target commands.

[0077] Among them, the regional control target instruction is a control instruction formed by transforming and formatting the original optimization solution, which can be issued to the second-level nodes. It is a macro-level target at the regional level. The third-level nodes aggregate the optimization results of all devices belonging to the same region in the original optimization solution to obtain the total power target of the region. They perform standardization checks, rounding, or smoothing on the instructions to ensure that they comply with the requirements of communication protocols and engineering applications, and to avoid instructions that are too frequent or too detailed to be executed.

[0078] This embodiment constructs and solves a global optimization problem through a centralized layer, achieving macro-level control. It only guides the operation of equipment in each managed area in a general direction, improving the accuracy and flexibility of hierarchical optimization.

[0079] In one embodiment, based on the distribution map of adjustable resources across the entire network, a security-constrained economic scheduling problem is constructed, resulting in an optimization problem, including:

[0080] Step 301: Based on the distribution map of adjustable resources across the entire network, identify the variables that the optimization algorithm needs to solve and obtain the decision variable definition set.

[0081] In this context, the decision variable definition set is the set of all unknowns that need to be automatically solved by the algorithm in a mathematical optimization problem, representing the controllable elements. Based on the distribution map of adjustable resources across the entire network, the third-layer nodes identify all adjustable control units and define the controllable parameters of these units as decision variables. For example, typical decision variables include: the active power output of each distributed power source, the active power exchanged between the power grid and the upper-level network, and the amount of load to be reduced by each load aggregator participating in demand response. The decision variable definition set clarifies the mathematical symbol, physical meaning, and value range of each variable, transforming control methods in the physical world into unknowns in a mathematical problem, laying the foundation for establishing the objective function and constraints.

[0082] Step 302: Based on the defined set of decision variables and market information, construct the mathematical expression of the objective function using the following formula:

[0083]

[0084] in, The objective function is expressed mathematically. Let be the number of distributed power sources, and i be the index of the distributed power source. For the power generation cost function, To contribute one's efforts For equivalent electricity price, Active power This is the network loss cost coefficient. For total active network loss, The number of load aggregators participating in demand response, where j is the index of the demand response resource. Compensation price per unit, To reduce active power load.

[0085] Specifically, the objective function is a mathematical formula used to quantitatively evaluate the merits of any set of decision variable values. The goal of the optimization algorithm is to find the combination of variables that minimizes this function value; in this embodiment, it is a minimization problem. The third-layer nodes, based on the decision variables and market information, construct a total cost expression. This formula includes four parts: distributed generation cost (the total cost of electricity production from all distributed power sources); electricity purchase cost (the cost of purchasing electricity from the main grid); grid loss cost (the equivalent cost of energy loss within the grid); and demand response cost (the compensation required to incentivize users to reduce load). This precisely transforms the economic objective of grid dispatch into a calculable mathematical objective.

[0086] Step 303: Based on the decision variable definition set and the predicted load data, construct the constraint conditions to obtain the constraint condition set.

[0087] Specifically, the constraint set is a collection of mathematical relationships and inequalities that all decision variables must adhere to, ensuring the technical and physical feasibility of the optimal solution. The third-layer nodes construct various types of constraints based on the decision variables and predicted load data. For example, these may include power balance constraints: all power injected into the grid must equal all power consumed plus grid losses, a fundamental physical law; equipment operation constraints: the value of each decision variable cannot exceed its physical limit; safe operation constraints: based on baseline scenario analysis, ensuring that the optimized system state will not experience voltage exceedances, line overloads, or other problems; and operational constraints: including generator ramp rate limits, ensuring that the optimized results are not only mathematically optimal but also physically achievable, and guaranteeing the safe and stable operation of the power grid.

[0088] Step 304: Combine the mathematical expression of the objective function and the set of constraints into an optimization problem.

[0089] The optimization problem is a complete and standardized mathematical problem statement, providing all the necessary information for the standard solver. The third-level nodes combine the mathematical expression of the objective function obtained from the first two steps with the set of constraints to form a complete mathematical proposition with clear requirements, which is: Under the premise of satisfying the set of constraints, find a set of values ​​for decision variables that minimize the objective function value.

[0090] This embodiment transforms a complex engineering problem into a well-structured and clearly defined standardized mathematical problem, preparing it for automatic solution by calling readily available mathematical programming solvers. This enables the optimization problem to be solved automatically, improving the efficiency and reliability of hierarchical optimization.

[0091] In one embodiment, starting from the current operating state, circuit problems occurring in a baseline scenario are inferred, resulting in a baseline scenario analysis report, including:

[0092] Step 401: Take the current running state as the initial point; and freeze the initial point to obtain the initial conditions of the baseline scene.

[0093] The current operating state is a snapshot of all key operating parameters of the power grid at a certain moment, which may include the voltage of each node, the power of each branch, and the output of generators. The initial point is the starting state used for simulation calculations; in this embodiment, it represents the current operating state. Freezing the initial point is a model setting operation that keeps the output or setpoint of all controllable devices fixed in subsequent simulations. The baseline scenario initial condition is a simulation starting point where a control strategy is fixed. The third-layer nodes collect real-time operating data of the power grid and set it as the initial state of the simulation. In the simulation model, the control mode of all controllable units is set to constant output or constant setpoint, and their operating points are locked at the values ​​of the initial state, creating a reference scenario without optimization intervention. This provides a benchmark for accurately assessing the future evolution trend and potential risks of the power grid under natural conditions.

[0094] Step 402: Map the load data to specific grid nodes in the grid model to obtain the power data table.

[0095] Specifically, load data is the predicted curve of the total load of the power grid over a future period. The power data table is a structured data set that distributes the total load forecast data to specific nodes of the power grid according to time and spatial dimensions. The third-level nodes read the load forecast data and, based on historical statistical patterns or preset allocation coefficients, decompose and distribute the total load value of each time segment to each load node of the power grid model, forming a detailed table containing information on time, node, active power, and reactive power.

[0096] Step 403: Based on the initial conditions of the baseline scenario and the power data table, the power flow of the power grid in the future is simulated through time-series power flow calculation to obtain the time-series power flow results.

[0097] Specifically, time-series power flow calculation is a simulation method that performs power flow calculations sequentially over time to simulate the dynamic processes of a power grid. The time-series power flow result is a complete data set recording the electrical states of all nodes and branches in the entire network at each calculation point within a future period. The third-level nodes begin calculations from initial conditions. For the first simulation period, the system reads the load data of each node at time T1 from the power data table, substitutes it into the power flow equations, and calculates the grid state at time T1. The calculation result at time T1 is used as the initial state at time T2, and the load data at time T2 is substituted to perform the power flow calculation at time T2. This iterative process proceeds sequentially over time until the entire simulation cycle is completed, dynamically and continuously simulating the behavior of the power grid under future load changes, revealing the evolution trajectory of the system state, rather than just an isolated instantaneous state.

[0098] Step 404: Compare the timing flow results with the equipment operation limitations to obtain a list of potential security issues.

[0099] Among them, equipment operating limits are the boundary values ​​for the safe operation of power grid equipment, including upper and lower voltage limits and line thermal stability limit capacity. The potential safety issue list is a detailed report recording all limit-crossing events and their attributes found in the simulation. The third-level node automatically compares each data point in the time-series power flow results with the preset operating limits of the corresponding equipment. Once an electrical quantity is found to exceed its limit, it is recorded as a potential safety issue, including: the identified problematic equipment, the type of issue, the time of occurrence, and the severity of the limit exceedance. After traversing all equipment and all time points, all issues are recorded and summarized into a list.

[0100] Step 405: Based on the operating point of the time-series power flow results, calculate the sensitivity parameters and linearization parameters to obtain the power grid linearization parameter set; and integrate the power grid linearization parameter set and the list of potential security issues to obtain the baseline scenario analysis report.

[0101] Specifically, sensitivity parameters and linearization parameters are mathematical parameters reflecting the physical characteristics of the power grid, calculated based on the current operating point. These can be elements of the Jacobian matrix, power transmission distribution factors, etc. The power grid linearization parameter set is a collection of these parameters. The baseline scenario analysis report is a comprehensive analysis report integrating a list of potential safety issues and the power grid linearization parameter set. The third-layer nodes use the operating state at a key time point in the time-series power flow results as a baseline. By solving the partial derivatives of the power flow equations, the sensitivity relationship between changes in node injected power and the impact on branch power and node voltage at that operating point is calculated, thus obtaining the linearization parameter set. The generated list of potential safety issues is then integrated with this power grid linearization parameter set into a unified report.

[0102] This embodiment identifies the objects to be optimized through a list of safety issues, providing crucial initial gradient information for the subsequent mathematical programming solver. This greatly improves the solution efficiency. By utilizing these linearization parameters, the solver can more accurately predict the search direction, reduce the number of iterations, find the optimal solution faster, and thus improve the processing speed of hierarchical optimization.

[0103] In one embodiment, the area control target command and local unit adjustable capability report calculate the control setpoints of all units within the control area to obtain the local unit operating boundary command, including:

[0104] Step 501: Define boundary constraints based on the regional control target instructions; and define the adjustment range based on the local unit adjustability report.

[0105] Boundary constraints are the mathematical representation of the overall regional task issued from the upper level. Originating from regional control target instructions, they define the overall operational objectives that the entire control area must meet in the next time period, typically expressed as an equality or inequality constraint. Adjustment range is the mathematical quantification of the adjustable capabilities of each lower-level unit. Derived from the adjustable capability report of each local unit, it precisely describes the upper and lower limits of active and reactive power that each first-level node can adjust upwards and downwards in the next time period, defining the operational space of each unit. Second-level nodes parse the regional control target instructions, transforming the described macroscopic objectives into one or more strict mathematical constraints. They also parse the adjustable capability reports of all subordinate units, extracting the upper and lower limits of the adjustable active / reactive power for each unit, forming a clear set of value ranges for each decision variable.

[0106] Step 502: Based on boundary constraints and adjustment range, construct and solve the optimization proposition to obtain the initial optimization solution.

[0107] Specifically, an optimization proposition is a complete mathematical optimization problem description, including decision variables representing the unknowns to be solved, an objective function representing the optimization objective, and constraints. The initial optimization solution is the theoretical result directly calculated by the mathematical optimization algorithm; it is a set of specific values ​​indicating the theoretically optimal power setting value for each local unit. The second-level nodes, based on defined boundary constraints and adjustment ranges, set specific adjustment command values ​​for each unit and select a suitable objective function. The objective function and all constraints are combined to form a complete optimization proposition. The built-in mathematical programming solver is then used to iteratively calculate this proposition. Through complex mathematical algorithms, the solver automatically searches for the set of decision variable values ​​that optimizes the objective function, while satisfying all constraints.

[0108] Step 503: Verify the feasibility of the initial optimized solution to obtain the local cell running boundary instructions.

[0109] Specifically, feasibility verification is the process of checking the safety and executability of the initial optimized solution, verifying whether the theoretical solution is truly feasible in actual engineering. Local unit operating boundary commands are verified and standardized final control commands prepared for issuance to each first-level node. Second-level nodes perform additional checks on the initial optimized solution to ensure it truly satisfies all constraints, assess whether the commands will cause the unit operating point to be too close to its capability boundary, thus lacking flexibility in responding to emergencies, and appropriately round or smooth continuous values ​​to conform to the communication protocols and control precision of field equipment. If verification passes, the initial optimized solution is converted into a standard command format; if potential problems are found during verification, recalculation or activation of backup safety strategies is triggered.

[0110] This embodiment receives instructions from higher authorities and capability reports from lower authorities, performs rigorous mathematical optimization and engineering verification, and generates refined control instructions that meet both global objectives and local feasibility. This ensures that the mathematically perfect theoretical solution derived from the optimization model is safe, stable, and operable when converted into actual control instructions. This greatly improves the reliability and engineering practicality of hierarchical optimization and avoids malfunctions or risks to field equipment caused by model idealization or calculation errors.

[0111] In one embodiment, solving the optimization proposition to obtain an initial optimization solution includes:

[0112] Step 601: Transform the optimization proposition into a standard quadratic programming form to obtain the quadratic programming problem.

[0113] The optimization proposition is the original optimization problem, containing a user-defined objective function and constraints. The standard quadratic programming form is a standard and well-studied problem form in mathematical optimization theory, consisting of three parts: a quadratic objective function, a set of linear equality constraints, and a set of linear inequality constraints. It is a standardized format that solvers can efficiently process. The quadratic programming problem is essentially an optimization proposition that has been equivalently transformed into the standard quadratic programming form. The second-level nodes perform mathematically equivalent transformations on the original optimization proposition, expanding and rearranging the objective function algebraically to represent a quadratic form of the decision variables. All constraints, whether linear or nonlinear, are transformed into simple linear equality or inequality expressions by introducing auxiliary variables, ensuring that the mathematical essence of the problem remains unchanged while its form becomes more regular and uniform.

[0114] Step 602: Iteratively calculate the quadratic programming problem based on the interior point method to obtain iterative process data.

[0115] Specifically, the interior-point method is an advanced numerical algorithm for solving constrained optimization problems. Its basic principle is to start from the interior of the feasible region and gradually approach the optimal solution on the boundary along a trajectory called the center path. Iterative computation is a numerical solution method that generates a sequence of approximate solutions by repeatedly executing a series of computational steps, hoping that this sequence converges to the true solution. Iterative process data refers to a set of intermediate data generated in each iteration, including: the decision variable values ​​at the current iteration point, the objective function value, the degree of constraint violation, and the values ​​of other auxiliary variables used to determine the convergence status. The second-level node calls the solver, starting from the set initial point, and initiates the iterative process of the interior-point method. In each iteration, the algorithm calculates the gradient of the objective function and the gradient of the constraints at the current point, solves a system of linear equations that is approximately simplified from the original problem, and thus determines an improved search direction and a suitable step size. Moving along this direction by this step size, a new solution estimate is obtained. This new point is closer to the true optimal solution. The new point and its related computational data are recorded as the process data of one iteration. Through a series of controlled and progressively optimized steps, a complex problem that is difficult to solve directly is transformed into a series of relatively simple subproblems, and the solution estimate gradually and stably approaches the high-precision optimal solution from the initial point.

[0116] Step 603: When the iteration process data meets the iteration stopping condition, stop the iteration calculation; and determine the last iteration process data as the initial optimization solution; the iteration stopping condition includes the original feasibility requirement, the dual feasibility requirement, and the dual gap requirement.

[0117] Specifically, iteration stopping conditions are a set of pre-defined numerical criteria used to determine when iterative computation can terminate, i.e., when the current solution is sufficiently close to the optimal solution. Primitive feasibility requirement is one of the stopping conditions, measuring the degree to which the current iteration point satisfies the constraints of the primal problem, requiring that the violation of the constraints be less than a very small tolerance. Dual feasibility requirement is another stopping condition, measuring the degree to which the current solution satisfies the constraints of the dual problem, involving conditions such as the sign of the Lagrange multipliers. Dual gap requirement is yet another stopping condition, representing the difference between the objective function values ​​of the primal and dual problems, always non-negative. When the gap is close to zero, it indicates that the current solution is very close to the global optimum, and strong duality holds. The initial optimal solution is the value of the decision variables contained in the data of the last iteration when iteration stops; this is determined as the solution to the optimization problem. After each iteration, the second-level node checks the current iteration process data, calculates its original feasibility error, dual feasibility error, and dual gap, and compares these three error values ​​with a preset, extremely small precision tolerance. Only when all errors are less than their corresponding tolerances is the convergence condition considered to be met. Once the condition is met, the iteration loop terminates immediately, and the decision variable values ​​obtained from the last iteration are output, which is the initial optimization solution.

[0118] This embodiment ensures that the final solution is not only feasible but also of high quality through strict, multi-faceted stopping conditions. It is mathematically proven to be sufficiently close to the true optimal solution, thus ensuring the scientific rigor and optimality of the hierarchical optimization.

[0119] In one embodiment, based on the local control parameter set and local unit operating boundary instructions, the adjustability of the power equipment within the operating boundary is calculated to obtain a local unit adjustability report, including:

[0120] Step 701: Generate the expected operating state diagram based on the local control parameter set.

[0121] The expected operating state diagram is a data model describing the detailed electrical state of all power equipment managed by the first-level nodes in the next time period, assuming continued operation according to the current local control parameter set. This includes the expected output power, expected power consumption of each device, and the expected voltage and current of key nodes. The first-level nodes use the current local control parameter set as a baseline input, combining the device's own operating characteristics and the internal power grid topology, to internally predict the electrical state distribution within the entire local unit over a short future period. This transforms the static device parameter set into a dynamic, system-level future operating prediction. This provides a reliable baseline scenario and calculation starting point for accurately assessing the regulation potential of each device.

[0122] Step 702: Based on the expected operating state diagram and the local unit operating boundary instructions, calculate the adjustable power of each controllable unit to obtain a detailed table of adjustable capabilities.

[0123] Specifically, the adjustable capacity detail table is a structured data list that records in detail the specific adjustable capacity of each controllable device within the local unit under given constraints. For each device, it includes the upward / downward adjustment amount of its active power and the upward / downward adjustment amount of its reactive power. The first-level nodes compare the expected operating point of each device in the expected operating state diagram with the upper and lower operating limits specified by the local unit's operating boundary instructions for that device. For each device, the difference between the current expected operating point and the upper operating limit is calculated, which is the downward adjustment capacity of that device; the difference between the current expected operating point and the lower operating limit is calculated, which is the upward adjustment capacity of that device. This process is repeated for all controllable devices, and the results are summarized into a detail table.

[0124] Step 703: Based on the adjustable capacity details table, aggregate active power and reactive power to obtain the basic adjustable capacity.

[0125] Specifically, the basic adjustable capacity is a preliminary estimate of the aggregated capacity of the entire local unit under ideal conditions, ignoring the constraints of the local power grid. This includes the total active power up / down adjustment capacity and the total reactive power up / down adjustment capacity. The first-level nodes summarize the data in the adjustable capacity details table, add up the upward active power adjustment capacity of all devices to obtain the unit-level total active power up adjustment capacity, and add up the downward active power adjustment capacity of all devices to obtain the total active power down adjustment capacity. Using the same method, the reactive power adjustment capacity of all devices is aggregated to obtain the total reactive power up and down adjustment capacity.

[0126] Step 704: Based on internal power grid constraints, perform a hazard verification of the basic adjustable capability to obtain a local unit adjustable capability report.

[0127] Internal grid constraints refer to the physical safety limitations of the local unit's internal grid, including line and transformer capacity limitations and allowable fluctuations in node voltage. Hazard verification is a simulation process used to assess whether extreme adjustments to the local unit's basic adjustable capacity will cause the internal grid to exceed limits. The local unit adjustable capacity report is a final, confirmed, safe, and practically usable unit-level adjustable capacity data report after internal safety verification. Using the expected operating state diagram as a baseline, the first-level nodes simulate calling the extreme values ​​in the basic adjustable capacity upwards and downwards. Through rapid sensitivity analysis, it checks whether any equipment or node within the local unit would exceed its internal grid constraints under such extreme adjustments. If a risk of exceeding limits is found, the reported aggregated capacity value is reduced accordingly until a safe adjustable capacity range is found that will not cause internal safety problems even with full adjustment.

[0128] This embodiment decomposes the macroscopic operating boundary instructions into specific and quantifiable adjustment ranges for each device, ensuring the safety of the entire system. It ensures that the adjustable capability reported by the first-level node is not the theoretical maximum value on paper, but a practically usable capability that takes into account local safety constraints. This avoids the risk that the instructions issued by the upper-level node may meet the global optimization goal but cause local power grid failures.

[0129] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0130] Based on the same inventive concept, this application also provides a distribution network hierarchical optimization system for implementing the aforementioned distribution network hierarchical optimization method. The solution provided by this system is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more distribution network hierarchical optimization system embodiments provided below can be found in the limitations of a distribution network hierarchical optimization method described above, and will not be repeated here.

[0131] In one exemplary embodiment, such as Figure 2 As shown, a hierarchical optimization system 800 for a power distribution network is provided, comprising:

[0132] The local execution module 801 is used to receive local unit operation boundary instructions and adjust the operating parameters of the power equipment based on the local unit operation boundary instructions to obtain a local control parameter set.

[0133] The local adjustable module 802 is used to calculate the adjustable capability of the power equipment within the operating boundary based on the local control parameter set and the local unit operating boundary command, and to obtain a local unit adjustable capability report.

[0134] The area optimization module 803 is used to calculate the control setpoints of all units within the control area based on the area control target command and the local unit adjustable capability report, and to obtain the local unit operating boundary command.

[0135] The regional adjustable module 804 is used by the second-layer node to run boundary instructions based on the local unit, calculate the total adjustable capacity of the control area in the next time period, and obtain a regional aggregate adjustable capacity report.

[0136] The central optimization module 805 is used by the third-layer nodes to perform global centralized optimization based on the regional aggregation adjustable capability reports of all control areas, and obtain regional control target instructions; the regional control target instructions are used to send them to each second-layer node.

[0137] Furthermore, the central optimization module 805 is also used for:

[0138] Based on the regional aggregated adjustable capacity report, the adjustable capacity of each control area is mapped to the corresponding node of the preset power grid model to generate a distribution map of adjustable resources across the entire network.

[0139] Starting from the current operating state of the power grid model, we infer the circuit problems that occur in the baseline scenario and obtain the baseline scenario analysis report; the baseline scenario represents the state of the control area that is not optimized.

[0140] Based on the distribution map of adjustable resources across the entire network, a security-constrained economic scheduling problem is constructed, leading to an optimization problem.

[0141] Based on the baseline scenario analysis report, the mathematical programming solver is invoked to solve the optimization problem and obtain the original optimization solution;

[0142] The original optimized solution is converted into control commands that can be issued, thus obtaining the area control target commands.

[0143] Furthermore, the central optimization module 805 is also used for:

[0144] Based on the distribution map of adjustable resources across the entire network, the variables that the optimization algorithm needs to solve are identified, and the set of decision variable definitions is obtained;

[0145] Based on the defined set of decision variables and market information, the mathematical expression of the objective function is constructed using the following formula:

[0146]

[0147] in, The objective function is expressed mathematically. Let be the number of distributed power sources, and i be the index of the distributed power source. For the power generation cost function, To contribute one's efforts For equivalent electricity price, Active power This is the network loss cost coefficient. For total active network loss, The number of load aggregators participating in demand response, where j is the index of the demand response resource. Compensation price per unit, To reduce active power load;

[0148] Based on the set of decision variable definitions and the predicted load data, constraints are constructed to obtain a set of constraints.

[0149] The objective function mathematical expression and the set of constraints are combined into an optimization problem.

[0150] Furthermore, the central optimization module 805 is also used for:

[0151] Use the current running state as the initial point; and freeze the initial point to obtain the initial conditions of the baseline scene.

[0152] The load data is mapped to specific grid nodes in the grid model to obtain the power data table;

[0153] Based on the initial conditions of the baseline scenario and the power data table, the power flow of the power grid over a future period is simulated through time-series power flow calculation, and the time-series power flow results are obtained.

[0154] By comparing the time-series power flow results with the equipment operating limitations, a list of potential security issues is obtained;

[0155] Based on the operating points of the time-series power flow results, sensitivity parameters and linearization parameters are calculated to obtain the power grid linearization parameter set; and the power grid linearization parameter set and the list of potential safety issues are integrated to obtain a baseline scenario analysis report.

[0156] Furthermore, the region optimization module 803 is also used for:

[0157] Boundary constraints are defined based on regional control target instructions; and the adjustment range is defined based on local unit adjustability reports.

[0158] Based on boundary constraints and adjustment range, an optimization proposition is constructed and solved to obtain an initial optimization solution.

[0159] The feasibility of the initial optimized solution is verified, and the local cell operation boundary instructions are obtained.

[0160] Furthermore, the region optimization module 803 is also used for:

[0161] The optimization proposition is transformed into a standard quadratic programming form, resulting in a quadratic programming problem.

[0162] The quadratic programming problem is iteratively calculated using the interior point method to obtain iterative process data.

[0163] When the iteration process data meets the iteration stopping condition, the iteration calculation stops; and the last iteration process data is determined as the initial optimization solution; the iteration stopping condition includes the original feasibility requirement, the dual feasibility requirement, and the dual gap requirement.

[0164] Furthermore, the locally adjustable module 802 is also used for:

[0165] Generate the expected operating state diagram based on the local control parameter set;

[0166] Based on the expected operating state diagram and the local unit operating boundary instructions, the adjustable power of each controllable unit is calculated to obtain a detailed table of adjustable capabilities.

[0167] Based on the adjustable capacity details table, the active power and reactive power are aggregated to obtain the basic adjustable capacity;

[0168] Based on internal power grid constraints, the basic adjustable capability is verified for risks, resulting in a local unit adjustable capability report.

[0169] In one embodiment, a computer device is provided, including a memory and a processor, the memory storing a computer program, the processor executing the computer program to implement the steps of the distribution network hierarchical optimization method as described above.

[0170] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0171] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The components described as separate parts may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this disclosure according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0172] The above-described embodiments are merely illustrative of several implementation methods of the embodiments of this application, and their descriptions are relatively specific and detailed. However, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the embodiments of this application, and these modifications and improvements all fall within the protection scope of the embodiments of this application.

Claims

1. A hierarchical optimization method for a power distribution network, characterized in that, Applied to a power distribution network system, the power distribution network system including a first-layer node, a second-layer node, and a third-layer node, wherein the first-layer node and the second-layer node are connected; and the second-layer node and the third-layer node are connected, the method includes: The first-layer node receives the local unit operation boundary instruction; and based on the local unit operation boundary instruction, adjusts the operating parameters of the power equipment to obtain a local control parameter set; The first-layer node calculates the adjustable capability of the power equipment within the operating boundary based on the local control parameter set and the local unit operating boundary command, and obtains a local unit adjustable capability report. The second-layer node calculates the control setpoints of all units within the control area based on the regional control target command and the local unit adjustable capability report, and obtains the local unit operating boundary command. The second-layer node calculates the total adjustable capacity of the control area in the next time period based on the local unit's operating boundary instructions, and obtains a regional aggregate adjustable capacity report. The third-layer node performs global centralized optimization based on the regional aggregate adjustable capability reports of all control regions to obtain the regional control target instruction; the regional control target instruction is used to issue to each second-layer node.

2. The method according to claim 1, characterized in that, The region aggregate adjustable capability report based on all control regions is used for global centralized optimization to obtain the region control target instruction, including: Based on the regional aggregated adjustable capacity report, the adjustable capacity of each control region is mapped to the corresponding node of the preset power grid model to generate a network-wide adjustable resource distribution map; Starting from the current operating state of the power grid model, circuit problems occurring in the baseline scenario are inferred, and a baseline scenario analysis report is obtained; the baseline scenario represents the state of the control region without optimization. Based on the aforementioned network-wide adjustable resource distribution map, a security-constrained economic scheduling problem is constructed, leading to an optimization problem. Based on the baseline scenario analysis report, the mathematical programming solver is invoked to solve the optimization problem and obtain the original optimization solution. The original optimized solution is converted into control commands that can be issued, thus obtaining the area control target command.

3. The method according to claim 2, characterized in that, Based on the network-wide adjustable resource distribution map, a security-constrained economic scheduling problem is constructed, resulting in an optimization problem, including: Based on the aforementioned distribution map of adjustable resources across the entire network, the variables that the optimization algorithm needs to solve are identified, and the set of decision variable definitions is obtained; Based on the defined set of decision variables and market information, the mathematical expression of the objective function is constructed using the following formula: in, The objective function is expressed mathematically. Let be the number of distributed power sources, and i be the index of the distributed power source. For the power generation cost function, To contribute one's efforts For equivalent electricity price, Active power This is the network loss cost coefficient. For total active network loss, The number of load aggregators participating in demand response, where j is the index of the demand response resource. Compensation price per unit, To reduce active power load; Based on the defined set of decision variables and the predicted load data, constraints are constructed to obtain a set of constraints. The objective function mathematical expression and the set of constraints are combined to form the optimization problem.

4. The method according to claim 3, characterized in that, The process begins with the current operating state, then deduces circuit problems occurring in a baseline scenario to obtain a baseline scenario analysis report, including: The current running state is taken as the initial point; and the initial point is frozen to obtain the initial conditions of the baseline scene. The load data is mapped to specific grid nodes in the grid model to obtain a power data table; Based on the initial conditions of the baseline scenario and the power data table, the power grid flow over a future period is simulated through time-series power flow calculation to obtain the time-series power flow results. By comparing the time-series power flow results with the equipment operating limitations, a list of potential security issues is obtained; Based on the operating point of the time-series power flow results, sensitivity parameters and linearization parameters are calculated to obtain the power grid linearization parameter set; and the power grid linearization parameter set and the potential security problem list are integrated to obtain the benchmark scenario analysis report.

5. The method according to claim 1, characterized in that, The region control target command and the local unit adjustable capability report are used to calculate the control setpoints of all units within the control region, and the local unit operating boundary command is obtained, including: Boundary constraints are defined based on the regional control target instructions; and the adjustment range is defined based on the local unit adjustability report. Based on the boundary constraints and the adjustment range, an optimization proposition is constructed and solved to obtain an initial optimization solution. The feasibility of the initial optimized solution is verified to obtain the local unit running boundary instructions.

6. The method according to claim 5, characterized in that, Solving the optimization problem to obtain an initial optimization solution includes: The optimization proposition is transformed into a standard quadratic programming form, resulting in a quadratic programming problem. The quadratic programming problem is iteratively calculated using the interior point method to obtain iterative process data. When the iterative process data meets the iteration stopping condition, the iterative calculation stops; and the last iterative process data is determined as the initial optimization solution; the iteration stopping condition includes the original feasibility requirement, the dual feasibility requirement, and the dual gap requirement.

7. The method according to claim 1, characterized in that, The step of calculating the adjustability of the power equipment within the operating boundary based on the local control parameter set and the local unit operating boundary command, and obtaining a local unit adjustability report, includes: Based on the local control parameter set, generate the expected operating state diagram; Based on the expected operating state diagram and the local unit operating boundary instructions, the power that each controllable unit can adjust is calculated to obtain a detailed table of adjustable capabilities. Based on the aforementioned adjustable capacity details table, active power and reactive power are aggregated to obtain the basic adjustable capacity; Based on internal power grid constraints, the basic adjustable capability is subjected to hazard verification to obtain the local unit adjustable capability report.

8. A hierarchical optimization system for a power distribution network, characterized in that, The system includes: The local execution module is used to receive local unit operation boundary instructions and adjust the operating parameters of the power equipment based on the local unit operation boundary instructions to obtain a local control parameter set. The local adjustable module is used to calculate the adjustable capability of the power equipment within the operating boundary based on the local control parameter set and the local unit operating boundary command, and to obtain a local unit adjustable capability report. The area optimization module is used to calculate the control setpoints of all units within the control area based on the area control target command and the local unit adjustable capability report, and to obtain the local unit operating boundary command. The regional adjustable module is used by the second-layer node to calculate the total adjustable capacity of the control area in the next time period based on the local unit running boundary instructions, and obtain a regional aggregate adjustable capacity report. The central optimization module is used by the third-layer nodes to perform global centralized optimization based on the regional aggregate adjustable capability reports of all control regions to obtain the regional control target instructions; the regional control target instructions are used to send them to each second-layer node.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images