Demand side power load division method and system based on adaptive interval

Abstract

The invention discloses a demand side power load division method and system based on a self-adaptive interval, relates to the technical field of power system dispatching control, and solves the problems that interval rollback and repeated recalculation are easy to occur during landing execution of a traditional method of first division and later review, and the execution-oriented one-time issuing requirement is difficult to meet. Feasible region constraints of voltage margin and current margin are embedded in the interval generation process, a graph self-encoder is trained through physical quantities such as voltage influence intensity and current influence intensity constructed by sensitivity and node margin to obtain low-dimensional features, and interval division can keep sensitive to out-of-limit risks during formation; constructing a load conflict graph based on the low-dimensional features and a sensitivity threshold value, and coloring the graph to obtain a peak shifting period, so that nodes which still possibly generate a superimposed effect in the same interval are decoupled on a time axis; and finally outputting a structured interval scheduling parameter set by combining the sensitivity continuous correction interval and the time sequence which are identified and updated on line, so as to ensure the implementability and the stability.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power system dispatching control, and more particularly to a demand side power load division method and system based on adaptive intervals. BACKGROUND

[0002] In the demand side management of distribution networks, common load clustering often relies on power consumption curves or business portraits for similarity clustering, and then safety checking and dispatching are performed in the execution stage through optimal power flow or rule base. This path does not explicitly consider the voltage and current constraints of the distribution network when the interval is formed, nor does it map the coupling relationship of topology, line impedance and distributed power injection to the feature space that can be used for optimization, resulting in concurrent peak shifting under the same feeder or the same bus, which easily induces power flow migration and adjacent line over-limit. At the same time, line parameter aging, switch state change and scene load fluctuation will cause the linearization model to deviate, and the traditional method of first division and then review is prone to interval rollback and repeated calculation when landing execution, which is difficult to meet the one-time issuance requirement for execution.

[0003] In view of the above problems, the present application provides a solution. SUMMARY

[0004] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a demand side power load division method and system based on adaptive intervals to solve the problems raised in the background art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical solutions:

[0006] The demand side power load division method based on adaptive intervals comprises the following steps:

[0007] Obtain the topology, node voltage, current and load power time series data of the distribution network, construct a power flow linearization equation based on LinDistFlow approximation, calculate the sensitivity matrix of voltage to active and reactive power and the sensitivity matrix of current to active and reactive power, and use recursive least squares to identify the sensitivity matrix online;

[0008] Extract the voltage influence intensity, current influence intensity, voltage margin and current margin characteristics of the node, input the characteristics into the trained graph autoencoder for dimension reduction, and generate a low-dimensional feature vector that retains the topology and physical coupling information of the distribution network;

[0009] Construct a load conflict graph based on the low-dimensional feature vector, the edges of the conflict graph represent the risk of over-limit when two loads are adjusted concurrently, and use a graph coloring algorithm to assign off-peak periods to the nodes of the load conflict graph according to a predetermined rule to form an off-peak scheme;

[0010] In combination with the staggered peak scheme and the voltage margin and current margin constraints, a differentiable clustering algorithm is used to divide the load intervals, and a load interval division result is output.

[0011] In a preferred embodiment, the online identification includes: presetting a sensitivity matrix element drift threshold value, comparing a current value with a historical reference value in real time, controlling the distributed power supply to inject a small perturbation within a preset amplitude interval when the difference exceeds the threshold value, collecting node voltage and current data after the perturbation, and updating the sensitivity matrix by recursive least squares.

[0012] In a preferred embodiment, the training of the graph autoencoder includes: using node features and topological adjacency of the power distribution network for a preset historical time length as a training set, corresponding to the number of node features, setting a preset number of hidden layers and using linear rectification activation, and simultaneously reconstructing the features and adjacency matrix at the output end, and optimizing the mean square error loss to a preset convergence threshold.

[0013] In a preferred embodiment, the construction of the load conflict graph includes: estimating the influence of concurrent adjustment of any two loads on branch current and node voltage according to the sensitivity matrix, establishing an edge between the corresponding node pairs when any branch is occupied or any node margin is lower than a preset safety threshold, and obtaining an edge set of the load conflict graph.

[0014] In a preferred embodiment, the graph coloring algorithm uses a greedy strategy and combines capacity constraints: staggered peak periods are assigned in descending order of node degree of the load conflict graph, nodes with edges are prohibited from falling into the same period, the number of staggered peak periods is within a preset interval, and the length of the period is determined according to a preset rule of capacity limit and scheduling period.

[0015] In a preferred embodiment, the initialization of the differentiable clustering includes: setting interval prototype vectors according to the number of staggered peak periods, initializing the prototypes by calculating the mean of the low-dimensional feature vectors of the loads in the corresponding periods, calculating the belonging probability of each load to each interval using a Softmax function, and constraining the sum of the belonging probabilities to be equal to one.

[0016] In a preferred embodiment, the objective function of the differentiable clustering includes: a distance term of the low-dimensional feature vector of the load and the prototype vector of the interval to which it belongs, and a logarithmic barrier penalty term constructed by the voltage margin and the current margin, and an adaptive matrix estimation type optimizer is used to iterate within a preset learning rate interval until a preset convergence criterion is reached.

[0017] In a preferred embodiment, the method further includes: generating a set of interval scheduling parameters based on the load interval division result, the set of parameters including interval identifier, node set, planned adjustment amount and safety threshold of each node, and outputting the set of parameters to the control unit through a communication interface for subsequent execution and docking.

[0018] In a preferred embodiment, the output is verified by the communication authentication module, which includes key verification and identity matching; after authorization, the instruction transmission module issues the interval scheduling parameter set; during execution, the data monitoring module collects node voltage and current data in real time; when the safety threshold is exceeded, the data recording module stores the event information and returns an exception code.

[0019] Demand-side power load allocation system based on adaptive intervals includes:

[0020] The receiving module is configured to acquire distribution network operation data and model parameters. The operation data includes the voltage and current of each node and the power time series of each demand-side load. The model parameters include the distribution network topology and line parameters. If some parameters are unknown, they are recorded as parameters to be identified.

[0021] The processing module is configured to perform physical modeling processing on the data acquired by the receiving module. Based on the LinDistFlow approximate linearized power flow equation, it calculates the sensitivity matrices of voltage to active power and reactive power, and the sensitivity matrices of current to active power and reactive power. It uses recursive least squares to identify the sensitivity matrices online, and then combines the influence intensity of node voltage, the influence intensity of current, and voltage margin and current margin to construct physical feature vectors. It uses a graph autoencoder to reduce the dimensionality and generate low-dimensional feature vectors that retain topological and physical coupling information.

[0022] The determination module is configured to set physical constraint rules based on the feasible domain limits of the distribution network voltage and current. Taking the low-dimensional feature vector generated by the processing module as input, it uses a differentiable clustering algorithm to divide the load interval. The clustering objective function simultaneously minimizes the feature divergence and over-limit risk within the interval and outputs the load interval division result.

[0023] The sending module is configured to generate scheduling instructions based on the load interval division results output by the determining module, determine the execution order of instructions according to a preset priority strategy, send scheduling instruction links to the corresponding control execution units, monitor node voltage and branch current in real time, and synchronously record operation logs containing control execution unit identification information, operation content, and time information.

[0024] The technical effects and advantages of this invention, based on an adaptive interval-based demand-side power load allocation method and system, are as follows:

[0025] This invention embeds feasible domain constraints of voltage and current margins during interval generation and trains a graph autoencoder with physical quantities such as voltage and current influence intensity and node margins constructed using sensitivity to obtain low-dimensional features. The interval division can remain sensitive to the risk of exceeding limits during its formation. Based on the low-dimensional features and sensitivity thresholds, a load conflict graph is constructed and colored to obtain staggered peak periods, decoupling nodes that may still have superimposed effects within the same interval on the time axis. Combined with online identification and updating of sensitivity to continuously correct intervals and time series, a structured interval scheduling parameter set is finally output, ensuring feasibility and stability under different operating conditions and topology switching conditions. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the demand-side power load allocation method based on adaptive intervals according to the present invention.

[0027] Figure 2 This is a schematic diagram of the matching control unit and scheduling process of the present invention;

[0028] Figure 3 This is a structural diagram of the demand-side power load partitioning system based on adaptive intervals according to the present invention. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments.

[0030] This application provides a demand-side power load allocation method based on adaptive intervals, referring to... Figure 1 , Figure 1 This is a flowchart of the intervalization method provided in the embodiments of this application, the method including:

[0031] Step S101: Obtain distribution network operation data and model parameters. The operation data includes the voltage and current of each node and the power time series of each demand-side load. The model parameters include the distribution network topology and line parameters, such as impedance. If some parameters are unknown, they are recorded as parameters to be identified. The operation data is obtained through a data acquisition terminal, SCADA system, or smart meter, and an undirected topology graph model of the distribution network is established.

[0032] Step S102 involves performing physical modeling on the acquired data to obtain processed data characterizing the feasible region of the distribution network. This processed data includes the electrical sensitivity matrix obtained based on the LinDistFlow approximation and the physical characteristic representations of each node. Specifically, using the radial topology and line impedance parameters of the distribution network, the power flow equation is linearized to establish an approximate linear relationship between node-injected power and voltage and current changes. , .in , This is a sensitivity matrix for voltage and current to changes in active and reactive power. When some line parameters are missing or drift, the elements of the sensitivity matrix can be identified online through small disturbance tests. For example, if a small power step is introduced by a distributed power source or load device during certain periods, the sensitivity matrix estimate is continuously corrected using a recursive least squares algorithm until it converges to the true value. Then, based on the sensitivity matrix, the physical influence characteristics of each node are calculated, including voltage influence strength and current influence strength. Voltage influence strength refers to the degree of influence of the node's injected power change on the voltage of each neighboring node, and current influence strength refers to the degree of influence of the node's injected power change on the current of each related branch. The above influence characteristics, along with the node's local voltage margin and branch current margin, are combined to form the physical feature vector of each node. The physical feature vector of each node is dimensionality-reduced using a graph embedding algorithm to obtain the low-dimensional representation vector h of the node. During this dimensionality reduction process, a graph autoencoder is trained to reconstruct the original features and adjacency relationships, ensuring that the node representation vector h retains the grid topology and physical coupling information. The processed data obtained in this step includes physical semantic information such as sensitivity matrix and node representation, ensuring that subsequent steps can fully perceive the electrical feasible domain of the distribution network.

[0033] Step S103: Based on preset physical constraint rules, determine the load interval division scheme corresponding to the processed data. The physical constraint rules are based on the feasible region constraints of the distribution network voltage and current, and impose constraints on the load adjustment actions executed simultaneously. On this basis, differentiable clustering or optimization algorithms are used to divide the demand-side loads into intervals, and loads with similar electrical impact characteristics and which will not exceed limits when executed concurrently are grouped into the same interval. Specifically, the representation vector h of the node to which each load belongs, obtained in step S102, is used as the clustering input to initialize several interval prototype vectors. Through iterative optimization, each load is softly assigned to a certain interval with a certain probability. During the optimization process, an objective function is set to simultaneously minimize the feature divergence and limit-exceeding risk within the interval: on the one hand, the representation vectors of loads within the same interval are constrained to be close to each other while the distance between different intervals is large; on the other hand, the physical feasible region constraints are incorporated into the clustering objective to penalize interval combinations that may cause voltage or current to exceed the safety margin. Specifically, a certain interval is estimated through the aforementioned sensitivity matrix. The concurrent adjustment actions specifically include changes in the active and reactive power of each node. , This has an impact on the overall network voltage and branch current. , Calculate the voltage of each node relative to the upper limit under the action of this interval. remaining margin And the current in each branch relative to the thermal stability limit. remaining margin The minimum voltage margin and minimum current margin for the entire range are introduced into the objective function. A logarithmic barrier is used to penalize situations where the margin of any node or branch approaches zero, thereby guiding the clustering results to avoid combinations with high risk of exceeding limits. By solving this objective function, the range label for each load is determined. This enables automatic interval partitioning based on physical feasible domain awareness.

[0034] Step S104: A scheduling instruction link is sent to the corresponding control execution unit to prompt the execution unit to perform concurrent scheduling control on the processed data. Specifically, based on the interval division result of step S103, a set of scheduling task instructions corresponding to each interval is generated, with one instruction for each interval. Then, the execution order of tasks in each interval is determined according to a preset priority strategy, and the instructions are sent to the corresponding control equipment or scheduler for load adjustment operations in this order. The scheduling instruction is used to prompt the control execution unit to perform the following operations to complete the load adjustment within the interval: click the instruction link and verify identity or permissions; if the verification is successful, enter the execution interface and adjust the power of the load within the interval according to the predetermined plan. By directly pushing the scheduling instruction to the relevant control personnel or automatic control device, the rapid issuance and execution of tasks are achieved. During the instruction execution process, the system monitors the voltage of key nodes and the branch current in real time to ensure that the adjustment operation remains within a safe threshold. If an abnormal trend is detected, the system can promptly send a warning or automatically stop the relevant operation.

[0035] The execution entity for the above steps can be the backend process of a distribution network dispatch control system or energy management platform. By comprehensively processing distribution network data through these steps, intelligent automation of load optimization dispatch is achieved while ensuring that voltage, current, and other data meet safety constraints. First, raw data of distribution network operation is collected, and the sensitivity parameters and representation vectors of each node's impact on voltage and current are obtained through physical model analysis. This standardized processing not only solves the problem of difficulty in quantifying the physical coupling relationship between different loads but also allows subsequent interval division and dispatch optimization to be established on a unified physical semantic space, significantly reducing errors that may arise from traditional experience-based division. Based on preset physical constraint rules, the system automatically matches each load adjustment task to the most suitable interval, significantly improving the accuracy and efficiency of decision-making compared to manual grouping based on experience. For example, load adjustment tasks involving heavy line loads are automatically assigned to intervals that reduce the power flow of that line, and reactive power compensation tasks involving voltage anomalies are assigned to intervals that support voltage. After receiving the dispatch instruction link, the control execution unit only needs to click to verify to jump to the execution interface and implement load adjustments. This seamless experience reduces the average scheduling response time, while the system records complete operation logs in real time for subsequent auditing and analysis.

[0036] The effectiveness of this technical solution is mainly reflected in three aspects: First, in terms of security, by integrating electrical feasible domain constraints into interval partitioning, coupled with pre-execution identity and permission verification and full-process log recording, the safe and controllable load concurrent adjustment process is ensured, avoiding risks such as voltage exceeding limits or line overload. Second, in terms of efficiency, the automatic completion of load clustering and dispatch instruction generation, and direct transmission to the execution unit, significantly saves the time of manual analysis and issuing instructions one by one, achieving rapid response. Finally, in terms of application scenarios, this solution can be widely applied to peak shaving and valley filling control, distributed power source absorption optimization, grid demand response, and emergency fault control in distribution networks, which is of great significance for improving the level of intelligent grid dispatching.

[0037] In an optional embodiment, the step of performing physical modeling processing on the acquired data to obtain processed data characterizing the feasible region of the distribution network specifically includes: calculating the Jacobian matrix of the LinDistFlow equation based on the distribution network topology and line parameters to obtain the voltage sensitivity matrix. and current sensitivity matrix When line parameters are incomplete or changing, the sensitivity matrix is ​​identified online using a recursive least squares method, taking advantage of distributed generation output fluctuations or load step disturbances. This yields updated sensitivity matrix estimates and their covariance. For example, during distribution network operation, when a slight change in the current of a branch is detected, it can be inferred that a disturbance has occurred in the active or reactive power output of nearby nodes. The data from this disturbance is used to correct the corresponding elements of the sensitivity matrix, and after multiple iterations and convergence, more accurate sensitivity parameters are obtained. The processed data obtained in this way can dynamically reflect the physical characteristics of the distribution network, providing an accurate basis for subsequent interval division.

[0038] In the above embodiments, the system constructs the physical representation vector of nodes through electrical influence characteristics and uses a graph autoencoder to maintain topology-related adjacency patterns. This means that when partitioning intervals, the system can consider the factor of "which load adjustments simultaneously will cause a significant increase in current on which line," thereby avoiding potential problems caused by grouping solely based on load curve similarity. Specifically, the voltage influence strength of node i on neighboring node j is defined. And the strength of the influence of node i on the current of branch e. Using the intensity of these influences within a defined radius as the main feature, and combining information such as the node's own voltage and current margin indices and distributed power output, a physical feature vector of the node is constructed. Then train the autoencoder network. Make low-dimensional vectors Able to reconstruct the original features And its adjacency relationships, thus implicitly including physical factors such as topology and impedance. In this way, the node representation vectors obtained are used for interval partitioning. They inherently possess physical semantics, allowing potentially highly interdependent nodes to be partitioned further apart, thus reducing concurrency risks.

[0039] In an optional embodiment, a load interval division scheme is determined according to preset physical constraint rules, including: determining a first control unit and a first scheduling process corresponding to the target load when the target load data does not contain a preset marker; and determining a second control unit and a second scheduling process corresponding to the target load when the target load data contains a preset marker. The target load data is any data item in the processed data, such as a load adjustment task. The first control unit is the first execution node of the first scheduling process, which includes a second control unit, and the second scheduling process also includes a second control unit. The control execution unit includes both the first and second control units. The preset marker can be used to identify load adjustment tasks with special urgency or importance levels, thereby enabling cascading processing. After obtaining the processed data, the system first determines whether each data item, such as a load adjustment task, has a special marker to decide which scheduling process to use. If it does not contain a preset marker, the first scheduling process is followed, assigning the task to the first control unit for initial adjustment; if it contains a preset marker, the second scheduling process is followed, directly escalating the task to a higher-level second control unit for processing, simplifying the process. The second scheduling process is equivalent to a simplification or cascading approach to the first scheduling process. For example, if a load adjustment task involves eliminating significant safety risks and requires immediate attention, the system will assign a pre-defined emergency flag. This will cause the task to bypass the conventional first control unit, such as the regional control station, and be directly sent to the distribution dispatch center as the second control unit for processing. The process will end after the task is completed. This flag-based mechanism enables cascading dispatch for special requests, improving response speed and targeting. Figure 2 This is a schematic diagram of a matching control unit and scheduling process in an embodiment of this application. The following is in conjunction with... Figure 2 The two processes described above will be further explained. Figure 2 The diagram shows two scheduling paths for a target task in the interval division results: the left path is the standard process when there is no marker, where the regional control station, as the first control unit, approves the task and then sends it to the distribution dispatch center, the second control unit, for review and completion; the right path is the bypass process when there is a marker, where the distribution dispatch center directly processes the task, and then the process ends. By determining whether the task data contains a preset marker, the system can flexibly select different processes, ensuring both hierarchical approval for general tasks to prevent misoperation and rapid processing for urgent tasks.

[0040] Optionally, when the target load data does not contain a preset marker, the first control unit corresponding to the target load is determined. This specifically includes: identifying keywords or features in the target data to obtain identification results; and determining the first control unit corresponding to the identification results according to a preset correspondence table, wherein the preset correspondence table records the correspondence between different keyword or feature patterns and control units. In the above embodiment, automatically matching a suitable control execution unit according to preset identification rules can improve the efficiency and accuracy of task allocation. The correspondence table is preset by distribution network operation experience and rules. For example, if the data content of a load adjustment task contains keywords related to "voltage anomaly," the correspondence table indicates that this task is assigned to the voltage control team as the first control unit; or, for example, if the content contains characteristics of large load transfer, the task is assigned to the load management personnel. By matching control units with keywords or patterns, intelligent classification and allocation of load adjustment tasks are achieved, avoiding subjective errors that may be caused by manual judgment, and further improving overall scheduling efficiency.

[0041] In an optional embodiment, sending scheduling instruction links to the control execution unit specifically includes: generating a set of scheduling instruction links corresponding to each processed data item, wherein each data item in the processed data corresponds to one instruction link; determining the priority order of each processed data item according to a preset priority strategy; and sequentially sending the corresponding instruction links to each unit in the set of control execution units according to the priority order, wherein each processed data item corresponds to one control execution unit, and the control execution unit includes multiple units. In other words, the system first generates a corresponding instruction link for each load adjustment task in the processed data and stores these links in a queue to be sent; then, it sorts the tasks in the queue according to the priority strategy to determine the processing order of the tasks; finally, it sends each instruction sequentially to the execution unit responsible for the task through a message or communication module according to the sorting order. For example, if there are N load adjustment tasks to be executed at a certain time, the system generates N instruction links corresponding to these N tasks respectively; if priority strategy A is used to sort these N tasks, the system will send these N instructions one by one to the responsible control personnel or devices according to the order obtained by strategy A, ensuring that the tasks are executed in the predetermined order.

[0042] In an optional embodiment, the priority strategy includes at least one of the following: a first-to-first-execute strategy, which sorts tasks in the processed data according to their creation time, with earlier submissions having higher priority; a short-duration-first strategy, which sorts tasks in the processed data according to their estimated execution time, with shorter estimated execution times having higher priority; and a balanced sorting strategy, which calculates a balance ratio for each task in the processed data as follows: (estimated task execution time + task waiting time) / estimated task execution time, with a higher balance ratio having higher priority, where the task waiting time is the difference between the current time and the task's submission time. The above embodiments provide three different sorting methods for selection. The first-to-first-execute strategy processes tasks entirely according to their entry time into the system, ensuring fairness in scheduling and adhering to the first-come, first-served principle. Its characteristic is simplicity and directness, but it may lack flexibility in certain emergency situations. The short-duration-first strategy sorts tasks based on their required execution time, prioritizing tasks that are easy to complete and have short execution times. This allows for the completion of more tasks in a short time, making it suitable for scenarios with a large backlog of tasks. However, its disadvantage is that important, time-consuming tasks may be repeatedly postponed. The balanced sorting strategy comprehensively considers the waiting time and execution time of each task through the formula above. When a task has a long waiting time, its balanced ratio increases, thus increasing its priority; while tasks with short execution times will also have larger ratios due to their smaller denominators. Therefore, this strategy balances fairness and efficiency. For example, at a certain moment, assuming the current time is 10:00, there are three tasks a, b, and c to be executed. Task a was submitted at 8:00 with a waiting time of 120 minutes, task b was submitted at 9:30 with a waiting time of 30 minutes, and task c was submitted at 9:00 with a waiting time of 60 minutes. The estimated execution times are a = 10 minutes, b = 30 minutes, and c = 20 minutes, respectively. Then, the balanced ratio of a is (10 + 120) / 10 = 13, the balanced ratio of b is (30 + 30) / 30 = 2, and the balanced ratio of c is (20 + 60) / 20 = 4. Therefore, the balanced ratios are a > c > b, and the corresponding task execution order is a, c, b. The advantage of a balanced sorting strategy is that it considers both task waiting time and execution speed, preventing any task from waiting too long while ensuring the system prioritizes easier tasks, thus combining the advantages of the previous two strategies. Among the three sorting methods, the scheduler can flexibly choose according to different business needs: when the number of tasks to be executed is large and needs to be processed quickly, the shortest execution time priority strategy can be selected; when there are few tasks and strict adherence to the first-come, first-served principle is required, the first-submitted, first-executed strategy should be chosen; when a balance between fairness and efficiency is desired, the balanced sorting strategy should be adopted. Providing multiple priority sorting methods increases the flexibility of task scheduling and further enhances the user experience.

[0043] In an optional embodiment, the feasible domain-aware interval division method further includes: automatically recording and storing operation logs; wherein, the operation logs are generated during the process of each control execution unit entering the execution interface to perform load adjustment operations, and the operation logs include the identification information of the control execution unit, specific operation content, and time information, etc. By automatically recording the execution process of each load adjustment task, it can be ensured that every step of the scheduling process is traceable. When the adjustment results cause disputes or require tracing responsibility, the log records can provide a detailed chain of evidence. In addition, the collected log data can also be used for subsequent analysis and model training to continuously optimize interval division and scheduling strategies, and achieve continuous improvement of the system. In practical applications, the log recording module can asynchronously save the operation data to a secure storage medium to ensure that it does not interfere with real-time control.

[0044] It should be noted that the method steps provided in the above embodiments are divided into logical sequences. In practical applications, some steps can be executed in different orders by software or hardware as needed, or several steps can be combined. For example, steps S103 and S104 can be executed in parallel under certain circumstances, that is, corresponding instructions can be issued step by step while dividing the interval to shorten the overall response time. Those skilled in the art should understand that the description of the order of method steps in this application does not constitute a limitation.

[0045] This application also provides a demand-side power load allocation system based on adaptive intervals, such as Figure 3 As shown, Figure 3 This is a structural block diagram of the system provided in the embodiments of this application. The system includes:

[0046] The receiving module 301 is used to acquire power distribution network operation data and model parameters, which corresponds to step S101 of the above method.

[0047] Processing module 302 is used to perform physical modeling processing on the acquired data to obtain processed data that characterizes the feasible domain of the distribution network, corresponding to step S102 of the above method.

[0048] The determination module 303 is used to determine the load interval division scheme that matches the processed data according to the preset physical constraint rules, which corresponds to step S103 of the above method.

[0049] The sending module 304 is used to send a scheduling instruction link to the control execution unit, prompting the execution unit to perform concurrent scheduling control on the processed data, corresponding to step S104 of the above method.

[0050] In an optional embodiment, the processing module 302 includes: a modeling unit, used to calculate an electrical sensitivity matrix based on the distribution network topology and monitoring data, perform physical modeling on the operating data, and obtain processed data; the modeling unit may also employ online identification technology, and when the model parameters are incomplete, iteratively correct the sensitivity matrix through small disturbance data to ensure that the processed data is accurate and effective.

[0051] In an optional embodiment, the determining module 303 includes: a first determining unit, configured to determine a first controlling unit and a first scheduling process corresponding to the target load when the target load data does not contain a preset marker, wherein the target load data is any data item in the processed data; and a second determining unit, configured to determine a second controlling unit and a second scheduling process corresponding to the target load when the target load data contains a preset marker. The first controlling unit is the first execution node in the first scheduling process, the first scheduling process includes a second controlling unit, the second scheduling process includes a second controlling unit, and the control execution unit includes both the first and second controlling units. This module implements the function of selecting different scheduling processes based on whether a special marker exists in the task data, ensuring the expedited processing of urgent tasks.

[0052] In an optional embodiment, the first determining unit includes: an identification subunit, used to identify keywords or features in the target load data when the target load data does not contain a preset tag, and obtain an identification result; and a mapping subunit, used to determine the first control unit corresponding to the identification result according to a preset correspondence table, wherein the preset correspondence table records the correspondence between different keyword or feature patterns and control units. Through the cooperation of the identification subunit and the mapping subunit, ordinary load adjustment tasks can be automatically assigned to appropriate area control personnel or devices, thereby improving scheduling efficiency.

[0053] In an optional embodiment, the sending module 304 includes: a generation unit, configured to generate a set of scheduling instruction links corresponding to the processed data, wherein each data item in the processed data corresponds to one instruction link; a third determination unit, configured to determine the priority order of each task in the processed data according to a preset priority strategy; and a sending unit, configured to send the corresponding instruction links sequentially to each unit in a set of control execution units according to the priority order, wherein each data item in the processed data corresponds to one control execution unit, and the control execution units include multiple units. The sending module 304 may further include a priority strategy library, with built-in multiple scheduling order decision strategies for the third determination unit to call, to meet the needs of different scenarios.

[0054] In an optional embodiment, the priority strategy includes at least one of the following: a first-to-submit, first-execute strategy (sorting tasks according to their submission times in the processed data); a shortest execution time priority strategy (sorting tasks based on their estimated execution time, with shorter estimated times having higher priority); and a balanced sorting strategy (calculating the balance ratio of each task using the formula (estimated execution time + waiting time) / estimated execution time, with higher balance ratios having higher priority, where the waiting time is the difference between the current time and the task submission time). In the balanced sorting strategy, the longer the waiting time or the shorter the execution time of each task, the higher its balance ratio, thus occupying a higher position in the sorting. For example, at the current moment, if task a has waited for 120 minutes, task b for 30 minutes, and task c for 60 minutes, and their estimated execution times are 10 minutes, 30 minutes, and 20 minutes respectively, then the calculated balance ratio for a is 13, for b is 2, and for c is 4. Therefore, the execution order will be a, c, b in that order. By flexibly selecting from these three strategies, a balance can be struck between efficiency and fair scheduling based on actual circumstances. Administrators can freely choose appropriate strategies according to the needs of the scenario: prioritizing more tasks when the workload is heavy, ensuring first-come-first-served fairness when the workload is light, and using a balanced strategy when both need to be considered. Diverse priority strategies help improve the adaptability of the scheduling system in different application environments.

[0055] In an optional embodiment, the system further includes a recording module for automatically recording and storing operation logs. These operation logs are generated during the process of the control execution unit entering the execution interface to perform load adjustment tasks. The operation logs include information such as the identification information of the control execution unit, the operation content, and a timestamp. The recording module stores key process information for each task execution, facilitating future auditing and analysis of the scheduling process. In the event of an anomaly or accident, the logs can be used to trace the operational details of each control unit and clarify the attribution of responsibility.

[0056] It should be noted that the system provided in the above embodiments can be deployed in hardware according to the above module division when implementing its functions, or several modules can be integrated into the same processing unit through software logic. Similarly, the module division is only for the purpose of clearly describing this solution, and the specific names and division methods of each functional module do not constitute a limitation on this application. The method embodiments and device embodiments of this application belong to the same concept, and their specific implementation process has been described in detail in the method section, and will not be repeated here.

[0057] This application also provides a computer-readable storage medium storing instructions that, when executed by a processor, cause the processor to perform the steps of the distribution network physical feasible domain sensing intervalization method described in any of the preceding claims. The computer-readable storage medium may include, but is not limited to, media capable of storing program code such as USB flash drives, read-only memory, random access memory, portable hard drives, magnetic disks, and optical discs. These instructions can be pre-programmed into the storage medium and, when installed on a device with processing capabilities, can be invoked and executed to implement the functions of the method described in this application.

[0058] It should be noted that, for the sake of brevity, the foregoing method embodiments are described as a series of actions, but this does not mean that the application limits the order of the steps. Based on the ideas of this application, some steps can be executed in different orders or in parallel without affecting the functional implementation. Secondly, those skilled in the art should also understand that the specific embodiments described in the specification are preferred embodiments of the technical solutions of this application, and not limitations on the scope of protection of this application. All equivalent improvements or substitutions made within the spirit and principles of this application should be covered within the scope of protection of this application.

[0059] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A demand-side power load allocation method based on adaptive intervals, characterized in that, Includes the following steps: The distribution network topology, node voltage, current and load power time series data are obtained. Based on the LinDistFlow approximation, a power flow linearization equation is constructed. The sensitivity matrices of voltage and current to active and reactive power are calculated. The sensitivity matrices are identified online using recursive least squares. Extract the voltage and current influence strength, voltage margin and current margin features of the nodes, and input the features into the trained graph autoencoder for dimensionality reduction to generate a low-dimensional feature vector that retains the topology and physical coupling information of the distribution network. A load conflict graph is constructed based on low-dimensional feature vectors. The edges of the load conflict graph represent the risk of exceeding limits when two loads are adjusted concurrently. A graph coloring algorithm is used to allocate staggered time periods to the nodes of the load conflict graph according to preset rules to form a staggered peak scheme. Combining peak shifting schemes with voltage and current margin constraints, a differentiable clustering algorithm is used to divide the load intervals, and the load interval division results are output.

2. The demand-side power load allocation method based on adaptive intervals according to claim 1, characterized in that: Online identification includes: setting a preset sensitivity matrix element drift threshold, comparing the current value with the historical benchmark value in real time, controlling the distributed power source to inject a small disturbance within a preset amplitude range when the difference exceeds the threshold, collecting node voltage and current data after the disturbance, and updating the sensitivity matrix using recursive least squares.

3. The demand-side power load allocation method based on adaptive intervals according to claim 1, characterized in that: The training of the graph autoencoder includes: using node features and topological adjacency of the distribution network with a preset historical duration as the training set, the input layer dimension corresponding to the number of node features, setting a preset number of hidden layers and using linear rectified activation, and simultaneously reconstructing the features and adjacency matrix at the output end, and optimizing to a preset convergence threshold with mean squared error loss.

4. The demand-side power load allocation method based on adaptive intervals according to claim 1, characterized in that... ； Constructing a load conflict diagram involves estimating the impact of concurrent adjustments of any two loads on branch currents and node voltages based on a sensitivity matrix. When any branch is occupied or any node margin is lower than a preset safety threshold, an edge is established between the corresponding node pairs to obtain the edge set of the load conflict diagram.

5. The demand-side power load allocation method based on adaptive intervals according to claim 4, characterized in that: The graph coloring algorithm employs a greedy strategy combined with capacity constraints: staggered time slots are allocated according to the descending degree of nodes in the load conflict graph, nodes with edges are prohibited from falling into the same time slot, the number of staggered time slots is within a preset range, and the duration of each time slot is determined based on preset rules of capacity limits and scheduling cycles.

6. The demand-side power load allocation method based on adaptive intervals according to claim 1, characterized in that, The initialization of differentiable clustering includes: setting the interval prototype vector according to the number of off-peak periods, initializing the prototype with the statistical mean of the low-dimensional feature vector of the load in the corresponding period, using the Softmax function to calculate the probability of each load belonging to each interval, and constraining the sum of the belonging probabilities to be equal to one.

7. The demand-side power load allocation method based on adaptive intervals according to claim 6, characterized in that, The objective function of differentiable clustering includes: the distance term between the low-dimensional feature vector of the load and the prototype vector of the interval to which it belongs, and the logarithmic barrier penalty term constructed by voltage margin and current margin. An adaptive moment estimation optimizer is used to iterate within a preset learning rate interval until the preset convergence criterion is reached.

8. The demand-side power load allocation method based on adaptive intervals according to claim 1, characterized in that, The method also includes: generating a set of interval scheduling parameters based on the load interval division results. The parameter set includes the interval identifier, node set, planned adjustment amount and safety threshold of each node, and outputting the parameter set to the control unit through the communication interface for subsequent execution and docking.

9. The demand-side power load allocation method based on adaptive intervals according to claim 8, characterized in that, Before output, the communication authentication module performs verification, which includes key verification and identity matching. After authorization, the instruction transmission module sends the interval scheduling parameter set. During execution, the data monitoring module collects node voltage and current data in real time. When the safety threshold is exceeded, the data recording module stores the event information and returns an exception code.

10. A demand-side power load allocation system based on adaptive intervals, used to implement the method of any one of claims 1-9, characterized in that, include: The receiving module is configured to acquire distribution network operation data and model parameters. The operation data includes the voltage and current of each node and the power time series of each demand-side load. The model parameters include the distribution network topology and line parameters. If some parameters are unknown, they are recorded as parameters to be identified. The processing module is configured to perform physical modeling on the data acquired by the receiving module. Based on the LinDistFlow approximate linearized power flow equation, it calculates the sensitivity matrices of voltage to active power and reactive power, and the sensitivity matrices of current to active power and reactive power. It uses recursive least squares to identify the sensitivity matrices online, and then combines the influence intensity of node voltage, the influence intensity of current, and voltage margin and current margin to construct physical feature vectors. It uses a graph autoencoder to reduce the dimensionality and generate low-dimensional feature vectors that retain topological and physical coupling information. The determination module is configured to set physical constraint rules based on the feasible region limits of the distribution network voltage and current. Taking the low-dimensional feature vector generated by the processing module as input, it uses a differentiable clustering algorithm to divide the load interval. The clustering objective function simultaneously minimizes the feature divergence within the interval and the risk of exceeding the limit, and outputs the load interval division result. The sending module is configured to generate scheduling instructions based on the load interval division results output by the determination module, determine the execution order of instructions according to a preset priority strategy, send scheduling instruction links to the corresponding control execution units, monitor node voltage and branch current in real time, and synchronously record operation logs containing control execution unit identification information, operation content, and time information.

Images