Power supply collaborative control method and system based on sensor edge computing

Abstract

The present invention discloses a power supply collaborative control method and system based on sensor edge computing, which relates to the field of power supply control technology. The method includes: collecting power grid data sets, establishing an initial tree structure diagram and a topology collaborative table; establishing a real-time power supply data set; calculating a local collaborative capability value and updating it to the initial tree structure diagram; performing edge fusion analysis based on the local collaborative capability value; performing regional load analysis on a first collaborative control scheme; using load anomaly identification and a topology collaborative table to perform structural optimization of the initial tree structure diagram, establish a second collaborative control scheme, distinguish between the first and second collaborative control schemes, and output a final collaborative control scheme. The method solves the technical problems existing in the prior art, such as high data transmission delay, inflexible centralized control response, high risk of regional load imbalance, and unreasonable power grid resource allocation, and achieves the technical effects of reducing data transmission delay, improving the real-time and flexibility of power supply control, and realizing accurate and efficient configuration of power grid resources.

Description

Technical Field

[0001] The present application relates to the technical field related to power supply control, and specifically to a power supply collaborative control method and system based on sensor edge computing. Background Art

[0002] With the massive influx of distributed power sources, energy storage devices, and smart power terminals, the grid structure is becoming increasingly complex, and the physical topology of the grid has evolved into a highly interconnected and intertwined complex network. The power consumption characteristics and power generation capacity of different regions vary significantly, and the rationality of regional division directly affects the overall operational efficiency of the grid. The diversity of node capability data also places higher demands on power supply control. Existing centralized power supply control requires that grid data be transmitted to a central control node for processing and decision-making. This not only results in significant data transmission delays, but also makes it difficult to meet power supply scenarios with high real-time requirements, such as sudden peak power consumption and intermittent power generation from renewable energy sources. Furthermore, the centralized transmission of large amounts of data increases network bandwidth pressure and the risk of data loss. Furthermore, centralized control lacks the ability to flexibly respond to local grid characteristics, making it impossible to fully utilize the computing and control potential of edge devices, making it difficult to achieve efficient and coordinated utilization of grid resources, thus affecting grid operational efficiency and quality.

[0003] Therefore, at the current stage, relevant technologies have technical problems such as high data transmission delay, inflexible centralized control response, high risk of regional load imbalance, and unreasonable grid resource allocation. Summary of the Invention

[0004] This application solves the technical problems of high data transmission delay, inflexible centralized control response, high risk of regional load imbalance, and unreasonable grid resource allocation in the existing technology by providing a power supply collaborative control method and system based on sensor edge computing. It achieves the technical effects of reducing data transmission delay, improving the real-time and flexibility of power supply control, and realizing accurate and efficient allocation of grid resources.

[0005] The present application provides a power supply collaborative control method based on sensor edge computing, which includes: collecting power grid data sets, establishing an initial tree structure diagram and a topology collaborative table, and the power grid data set includes power grid physical topology, regional division and node capability data; edge nodes perform real-time collection of power supply data to establish a real-time power supply data set, and the real-time power supply data set includes demand data, load data and energy storage data; activate the computing layer of the edge node, calculate the local collaborative capability value according to the real-time power supply data set, and update the local collaborative capability value to the initial tree structure diagram; use the regional division in the initial tree structure diagram to perform edge fusion analysis based on the local collaborative capability value to establish a first collaborative control scheme; perform regional load analysis on the first collaborative control scheme to generate a load anomaly identifier; use the load anomaly identifier and the topology collaborative table to perform structural update optimization of the initial tree structure diagram to establish a second collaborative control scheme, and after distinguishing the first collaborative control scheme and the second collaborative control scheme, output the final collaborative control scheme.

[0006] In a possible implementation, the power supply collaborative control method based on sensor edge computing also performs the following processing: using the first collaborative control scheme to perform control fitting of the power supply network, and using the first control fitting result to establish a first control score; performing control fitting of the second collaborative control scheme, and using the second control fitting result to establish a second control score; obtaining an updated tree structure diagram mapped by the second collaborative control scheme, and establishing a structural change cost based on the updated tree structure diagram and the initial tree structure diagram; using the first control score, the second control score, and the structural change cost to select and judge the first collaborative control scheme and the second collaborative control scheme.

[0007] In a possible implementation, the power supply collaborative control method based on sensor edge computing also performs the following processing: sending the first control fitting result to an evaluation channel connected to the initial tree structure diagram; calling the dynamic prediction sub-channel in the evaluation channel, and using the dynamic prediction sub-channel to predict the performance of the trend change control results of energy storage and load, and generating a dynamic prediction score; calling the local collaborative sub-channel in the evaluation channel, and using the local collaborative sub-channel to evaluate the power collaborative coordination under the first control fitting result, and establish a collaborative coordination score; calling the control evaluation sub-channel in the evaluation channel, and using the control evaluation sub-channel to evaluate the delay, energy consumption, reliability, and load matching under the first control fitting result, and establish a control evaluation score; outputting the first control score according to the dynamic prediction score, collaborative coordination score, and control evaluation score.

[0008] In a possible implementation, the power supply collaborative control method based on sensor edge computing also performs the following processing: establishing an evaluation feature set of local collaborative capability, the evaluation feature set including load flexibility features, energy storage support features, and local energy balance features; and using the evaluation feature set to calculate the local collaborative capability value of the real-time power supply data set.

[0009] In a possible implementation, the power supply collaborative control method based on sensor edge computing also performs the following processing: predicting demand stability of the demand data and generating a demand stability influencing factor; using the demand stability influencing factor to identify the risk of local energy balance characteristics in the evaluation feature set, and using the risk identification result to update the local collaborative capability value.

[0010] In a possible implementation, the power supply collaborative control method based on sensor edge computing also performs the following processing: using the load anomaly identifier to perform inverse positioning of the abnormal area; querying the topology collaborative table, and constructing a candidate sub-topology set based on the inverse positioning of the abnormal area and the query results; establishing a candidate tree structure diagram with all candidate sub-topology sets, and performing control fitting simulation on each candidate tree structure diagram; using a multi-index scoring matrix to output the optimal control scheme for each candidate tree structure diagram, and screening all the optimal control schemes, and outputting the screening results as the second collaborative control scheme.

[0011] In a possible implementation, the power supply collaborative control method based on sensor edge computing also performs the following processing: performing computational testing on the computing layer of the edge node to generate a test error; locating the abnormal edge node based on the test error, and matching the neighborhood assistance node according to the positioning result; and using the matched neighborhood assistance node to perform data sharing processing of the abnormal edge node.

[0012] In a possible implementation, the power supply collaborative control method based on sensor edge computing also performs the following processing: using the positioning results to obtain all neighborhood nodes; performing adaptation analysis of all neighborhood nodes by independent calculation to establish a first adaptation analysis result; performing adaptation analysis of all neighborhood node combination calculations to establish a second adaptation analysis result; if the highest adaptation result in the second adaptation analysis result is higher than the highest adaptation result in the first adaptation analysis result, and meets the combination breakthrough threshold, then the neighborhood node combination corresponding to the highest adaptation result in the second adaptation analysis result is output as a neighborhood assistance node.

[0013] In a possible implementation, the power supply collaborative control method based on sensor edge computing also performs the following processing: collaborative scheduling of the power grid is performed based on the final collaborative control scheme, and a feedback verification data set is established; the feedback verification data set is used to evaluate the effect of collaborative scheduling, and collaborative scheduling optimization is performed based on the effect evaluation feedback.

[0014] The present application also provides a power supply collaborative control system based on sensor edge computing, which includes: a power grid data set acquisition module, which is used to acquire power grid data sets and establish an initial tree structure diagram and a topology collaborative table, wherein the power grid data set includes power grid physical topology, regional division and node capability data; a real-time power supply data set establishment module, which is used to perform real-time acquisition of power supply data by edge nodes and establish a real-time power supply data set, wherein the real-time power supply data set includes demand data, load data and energy storage data; a local collaborative capability value calculation module, which is used to activate the computing layer of the edge node, calculate the local collaborative capability value according to the real-time power supply data set, and update the local collaborative capability value to the initial tree structure diagram; an edge fusion analysis module, which is used to perform edge fusion analysis based on the local collaborative capability value using the regional division within the initial tree structure diagram, and establish a first collaborative control scheme; a regional load analysis module, which is used to perform regional load analysis on the first collaborative control scheme and generate a load anomaly identifier; a collaborative control scheme output module, which is used to perform structural optimization of the initial tree structure diagram using the load anomaly identifier and the topology collaborative table, establish a second collaborative control scheme, and output the final collaborative control scheme after distinguishing the first collaborative control scheme and the second collaborative control scheme.

[0015] The power supply collaborative control method and system based on sensor edge computing proposed in this application is intended to collect power grid data sets, establish an initial tree structure diagram and a topology collaborative table; establish a real-time power supply data set; calculate the local collaborative capability value and update it to the initial tree structure diagram; perform edge fusion analysis based on the local collaborative capability value; perform regional load analysis on the first collaborative control scheme; use the load anomaly identifier and the topology collaborative table to optimize the structural update of the initial tree structure diagram, establish a second collaborative control scheme, distinguish the first and second collaborative control schemes, and output the final collaborative control scheme. This solves the technical problems existing in the prior art, such as high data transmission delay, inflexible centralized control response, high risk of regional load imbalance, and unreasonable power grid resource allocation, and achieves the technical effects of reducing data transmission delay, improving the real-time and flexibility of power supply control, and realizing accurate and efficient configuration of power grid resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions of the embodiments of the present disclosure, the accompanying drawings of the embodiments of the present disclosure are briefly introduced below. Flowcharts are used in this application to illustrate the operations performed by the systems according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed in precise order. Instead, various steps may be processed in reverse order or simultaneously as needed. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0017] Figure 1 A flow chart of the power supply collaborative control method based on sensor edge computing provided in an embodiment of the present application.

[0018] Figure 2 A schematic diagram of the structure of a power supply collaborative control system based on sensor edge computing provided in an embodiment of the present application.

[0019] Explanation of the reference numerals: power grid data set acquisition module 10 , real-time power supply data set establishment module 20 , local collaborative capability value calculation module 30 , edge fusion analysis module 40 , regional load analysis module 50 , collaborative control scheme output module 60 . DETAILED DESCRIPTION

[0020] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0021] In order to make the purpose, technical solutions and advantages of this application clearer, the application will be further described in detail below with reference to the accompanying drawings. The described embodiments should not be regarded as limiting this application. All other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.

[0022] In the following description, reference is made to “some embodiments”, which describes a subset of all possible embodiments, but it will be understood that “some embodiments” may be the same subset or different subsets of all possible embodiments and may be combined with each other without conflict, and the terms “first\second” involved are merely used to distinguish similar objects and do not represent a specific ordering of the objects. The terms “including” and “having” and any variations are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or modules that are not clearly listed or that are inherent to these processes, methods, products or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are for the purpose of describing the embodiments of this application only.

[0023] The embodiment of the present application provides a power supply collaborative control method based on sensor edge computing, such as Figure 1 As shown, the method includes:

[0024] Step S100 : collecting a power grid data set and establishing an initial tree structure diagram and a topology coordination table. The power grid data set includes power grid physical topology, regional division, and node capability data.

[0025] Preferably, the physical topology of the power grid refers to the actual connection method and layout of various electrical equipment in the power grid (such as generators, transformers, transmission lines, switches, etc.), which is used to describe the hardware architecture of the power grid, including the direction of the lines, the location of the nodes, and the interconnection relationship between the equipment. Specifically, the power grid is surveyed on the spot through satellite images, aerial photography, etc., and the location, model, connection method and other information of the power equipment are recorded. At the same time, the power grid planning and design drawings are consulted to obtain detailed power grid architecture information, including the direction of transmission lines, substation layout, etc.; the intelligent sensors and monitoring devices of the power grid equipment, such as smart meters and phasor measurement units (PMUs), are used to collect the operating status data of the equipment in real time, including electrical parameters such as voltage, current, and power.

[0026] Preferably, the regional division is to divide the entire power grid into different areas based on the geographical distribution, load characteristics, voltage level, etc. of the power grid. Different areas may have different electricity demand, power generation capacity and power grid operation characteristics. For example, the load density in the city center is relatively high, mainly for industrial, commercial and residential electricity consumption; while the load in suburban or rural areas is relatively low, and there may be distributed energy access. Specifically, the power load in different areas of the power grid is monitored and analyzed over a long period of time, and the load curve, peak load, load rate, etc. of each area are statistically analyzed. The power grid is divided into different areas according to the similarities and differences of the load characteristics; the geographical distribution information of the power grid is combined with the analysis of the power grid operation data, such as the transmission power, voltage loss, short-circuit current, etc. of the transmission line, to divide the power grid into regions.

[0027] Preferably, node capability data refers to the capability information of each node in the power grid (such as substations, power plants, important electrical equipment, etc.), including the node's power generation capacity, transmission capacity, transformation capacity, load carrying capacity and energy storage capacity, etc., which is used to reflect the role and performance of the node in the power grid, and helps to evaluate the power supply capacity and reliability of the power grid and make reasonable resource allocation. Specifically, the rated parameters of the equipment are obtained from the technical manuals and product manuals of the power equipment, such as the rated power of the generator, the capacity of the transformer, the current carrying capacity of the transmission line, etc.; the operating data of the equipment, such as the voltage, current, power factor, etc. of the node, are collected in real time through sensors and monitoring devices on the node equipment, and the actual operating capacity and remaining capacity of the node equipment are evaluated by real-time data analysis; the grid node equipment is tested and tested regularly, such as the output test of the generator, the short-circuit test of the transformer, the breaking capacity test of the switchgear, etc., to obtain accurate performance data of the equipment and verify whether the equipment meets the design requirements and operating standards.

[0028] Preferably, certain key nodes in the power grid (such as substations, power supply points, etc.) are used as tree nodes, and transmission lines or other connection relationships are used as tree edges to construct an initial tree structure diagram, which can reflect the power grid structure and hierarchical relationship, that is, a graphical model that uses a tree structure to represent the topological relationship of the power grid. Through the initial tree structure diagram, the connection relationship and hierarchical architecture between the various parts of the power grid can be intuitively displayed to facilitate the analysis, planning and control of the power grid. A topology coordination table is a specialized database that records in detail the actual connections between various devices and nodes in the power grid, known as "current connections." This includes information about the connections between substations and power plants connected by transmission lines and power consumption equipment, describing the grid's real-time transmission paths. In addition to current connections, the topology coordination table also records potential changes in connections under different operating conditions, known as "potentially switchable connections." For example, when a transmission line fails or requires maintenance, the grid maintains power by switching to alternative backup lines. The connections of these backup lines are recorded in the topology coordination table. Furthermore, under certain rules, the topology coordination table can support autonomous connection switching. This means that the grid can automatically adjust connections according to preset rules based on real-time operating data (such as load changes and device status) to achieve optimized operation. Essentially, a topology coordination table is a combined database of "connection relationships and connection variability," enabling the grid to better adapt to complex operating conditions and changing demands.

[0029] In step S200 , the edge node performs real-time collection of power supply data to establish a real-time power supply data set, where the real-time power supply data set includes demand data, load data, and energy storage data.

[0030] Preferably, edge nodes are used to collect power supply data in real time to obtain a real-time power supply data set, wherein the real-time power supply data set includes demand data, load data, and energy storage data. Specifically, demand data refers to the real-time demand information for electricity of each power-consuming terminal in the power system, reflecting the user's usage and usage pattern of electricity at different times. It is collected through power monitoring equipment (such as smart meters) at the user end, including real-time power consumption, power consumption, power consumption time distribution, etc. of different types of users (such as residential, commercial, industrial users, etc.). For example, the power used by residential users for air conditioning, television, lighting, etc. during the peak period of electricity consumption in the evening, and the real-time power consumption of each production line of industrial users during the production process.

[0031] Preferably, load data refers to the power load borne by each node or line in the power grid, reflecting the actual load status of the power grid during operation. It is collected through sensors and measuring devices (such as current transformers, voltage transformers, and power transmitters) on substations, transmission lines, and other key power equipment. It includes parameters such as the node's active power, reactive power, apparent power, current, and voltage, as well as information such as the load rate and losses of the transmission line. For example, the real-time current level and power factor at the outlet of a substation, as well as the load percentage of a transmission line at a specific moment, are examples. Energy storage data refers to information related to energy storage devices in the power system, used to understand the status and charge and discharge capacity of the energy storage devices. It is collected by the energy storage devices' own monitoring and measuring devices to monitor the operating status of the energy storage devices in real time. This includes parameters such as the energy storage device type (such as battery storage, flywheel storage, pumped hydro storage, etc.), capacity, remaining capacity (or state of charge (SOC)), charge and discharge power, and charge and discharge efficiency.

[0032] Step S300: activating the computing layer of the edge node, calculating the local coordination capability value according to the real-time power supply data set, and updating the local coordination capability value to the initial tree structure diagram.

[0033] Preferably, the computing layer of the edge node is activated, wherein the edge node is a computing node close to the data source or user end, and then the real-time power supply data is input into the computing layer of the edge node, and the computing resources of the edge node (such as processor, memory, etc.) are used to analyze and process the real-time power supply data, and the local coordination capability value is calculated from multiple dimensions, wherein the local coordination capability value measures the ability of the node to independently respond to changes in the current power supply status, specifically including the ability to regulate its own load, the response capability of local energy storage (release or absorption), the ability to adapt to fluctuations of local power generation (such as photovoltaics), the robustness to sudden load disturbances, and the ability to maintain stable operation without relying on external coordination.

[0034] Preferably, load data, such as real-time parameters like active power, reactive power, current, and voltage, are extracted from real-time power supply datasets. By analyzing the time-varying curves of these parameters, the magnitude of load fluctuations is calculated. If a node can automatically adjust the operating status of connected devices (e.g., adjusting the power of non-critical equipment or starting and stopping some loads) to maintain a stable load range during load fluctuations, a score is assigned based on the timeliness and effectiveness of these adjustments, and the corresponding node's load regulation capability is calculated. Based on the state of charge (SOC) and charge / discharge power in the energy storage data, different power supply scenarios are simulated. When the grid load is too high, the energy storage device is evaluated to determine whether it can quickly release energy within a short period of time (e.g., within 10 seconds) and meet a certain power release ratio (e.g., 80% of the load shortfall). When the grid power is excessive, the energy storage device is evaluated to determine whether it can quickly absorb excess energy with high efficiency (e.g., achieving 90% of the rated charge / discharge efficiency). The response capability of the local energy storage is calculated based on the energy storage device's response speed, power availability, and efficiency in different scenarios.

[0035] Preferably, for nodes associated with local power generation (e.g., photovoltaics), power generation data is combined with real-time environmental data (e.g., light intensity and temperature) to analyze power fluctuation patterns. Specifically, by calculating the maximum power fluctuation amplitude and frequency per unit time, the node is evaluated for its ability to mitigate the impact of power generation fluctuations on the grid through adaptive control strategies (e.g., adjusting inverter parameters and optimizing power output curves). For example, if a photovoltaic node can control power generation fluctuations within ±15% of the rated power when light intensity fluctuates dramatically, and can quickly track light intensity changes and adjust output, it demonstrates good fluctuation adaptation capability and is assigned a corresponding score. Sudden load disturbances (e.g., large motor startup causing instantaneous power surges) are identified in the real-time power supply dataset. The node's voltage and frequency fluctuations after the disturbance are observed, and the degree of deviation from the normal range and the recovery time are calculated. If, after a sudden load disturbance, the node's voltage fluctuations can be controlled within ±10% of the rated voltage within 1 second, the frequency fluctuations within 0.2 Hz, and stability can be restored within 5 seconds, the robustness against sudden load disturbances is calculated based on these parameters.

[0036] Optimally, the capability values ​​of each node are combined with the node's operational data for a certain period of time (e.g., one hour) without external coordination to analyze its voltage and frequency stability, equipment failure rate, and other factors. If the node's voltage and frequency remain within the standard range and no critical equipment failures occur during this period, it demonstrates strong independent and stable operational capabilities and is assigned a higher score. Finally, the local coordination capability values ​​of each dimension are combined into a feature vector and updated to the initial tree structure. This means that an attribute field is added to each node in the tree structure to store the local coordination capability value, facilitating edge fusion analysis and power supply coordinated control decisions.

[0037] Furthermore, step S300 also includes step S310, establishing an evaluation feature set of local collaborative capability, wherein the evaluation feature set includes load flexibility features, energy storage support features, and local energy balance features; and step S320, using the evaluation feature set to calculate the local collaborative capability value of the real-time power supply data set.

[0038] Preferably, load flexibility characteristics, energy storage support characteristics, and local energy balance characteristics are used as evaluation characteristics of local collaborative capabilities to establish an evaluation feature set. Specifically, load flexibility characteristics are used to measure the ability of edge nodes to flexibly adjust loads. In actual power grid operation, different electrical equipment and users have diverse and uncertain demands for electricity, which may include load adjustment range (the upper and lower limits of the load that the node can adjust), load adjustment speed (how fast the node adjusts the load), and load adjustment accuracy (how accurately the node adjusts the load); energy storage support characteristics are used to evaluate the support capabilities of local energy storage equipment in the power supply process, which may include energy storage Capacity (the amount of electricity that the energy storage device can store), energy storage charging and discharging efficiency (the energy conversion efficiency of the energy storage device during the charging and discharging process), and energy storage response time (the time interval from the energy storage device receiving the charging and discharging instruction to the actual charging and discharging). Local energy balance characteristics are used to evaluate the balance between the node's own power generation and power consumption and the ability to effectively utilize local energy. They may include the matching degree of local power generation and load (the degree of fit between local photovoltaic, wind power and other distributed energy sources and local load demand), energy self-sufficiency rate (the proportion of the node relying on local energy to meet its own power demand), and energy complementarity (the complementary relationship between different types of local energy).

[0039] Preferably, the local collaborative capability value of the real-time power supply data set is calculated based on the evaluation feature set. Specifically, data related to the evaluation feature set is extracted from the real-time power supply data set, including extracting real-time power, change trend and other information of the load from the load data for calculating the load flexibility characteristics; the current power, charge and discharge status, etc. of the energy storage device are obtained from the energy storage data for evaluating the energy storage support characteristics, and the extracted data is preprocessed, including data cleaning, normalization and other operations to eliminate the influence of the data dimension and outliers; then, according to the evaluation features of the evaluation feature set, the specific values ​​of each feature are calculated respectively, for example, according to the adjustment amplitude and speed of the load over a period of time, the load flexibility feature values ​​such as the load adjustment range and the load adjustment speed are calculated; according to the charge and discharge power and efficiency of the energy storage device, the energy storage support feature values ​​such as the energy storage charge and discharge efficiency are calculated; finally, the weighted sum of each feature value is performed to obtain the final local collaborative capability value, thereby more comprehensively and accurately evaluating the collaborative capability of the edge node under the real-time power supply state.

[0040] Furthermore, step S320 also includes step S321, performing demand stability prediction on the demand data to generate a demand stability influencing factor; step S322, using the demand stability influencing factor to identify the risk of local energy balance characteristics in the evaluation feature set, and using the risk identification result to update the local collaborative capability value.

[0041] Preferably, the demand data of the power system is affected by multiple factors, such as time (season, day and night, etc.), weather, economic activities, user behavior, etc. Through time series analysis, machine learning (such as neural networks, support vector machines, etc.), combined with historical power grid data, a prediction model is established to predict power demand in the future period, judge its changing trend and degree of fluctuation, and then evaluate the stability of power grid demand, and generate a demand stability influencing factor to characterize the impact of demand stability on power system operation. For example, if the prediction results show that demand fluctuations are small and the changing trend is relatively stable, the demand stability influencing factor may take a higher value, indicating that demand stability is better and is more beneficial to the stable operation of the power system; conversely, if it is predicted that demand will fluctuate significantly or the uncertainty is strong, the demand stability influencing factor will take a lower value.

[0042] Preferably, the demand stability impact factor is used to identify the risks of the local energy balance characteristics in the evaluation feature set, that is, to analyze the potential impact of demand stability on the local energy balance. Specifically, if the demand stability impact factor is low, it means that the demand fluctuates greatly, which may lead to a decrease in the matching degree between local power generation and load, and the power generation equipment is difficult to quickly follow the large changes in demand, thereby increasing the risk of local energy balance; the instability of demand may cause the proportion of nodes relying on local energy to meet their own electricity needs to change. If the demand fluctuation exceeds the regulation capacity of local power generation and energy storage, more electricity will be obtained from the external power grid, thereby reducing the energy self-sufficiency rate and increasing dependence on the external power grid. Local energy faces certain risks; demand instability affects the complementarity between different types of local energy. For example, energy storage equipment originally cooperates with local power generation for charging and discharging according to a relatively stable demand pattern to achieve energy complementarity. However, when demand fluctuates significantly, the charging and discharging strategy of the energy storage equipment may need to be adjusted frequently. If the adjustment is not timely or unreasonable, it will affect the effect of energy complementarity and thus affect the local energy balance.

[0043] Preferably, the local coordination capability value is adjusted and updated according to the risk identification results of the local energy balance characteristics. Specifically, if the risk identification shows that the local energy balance faces a greater risk, for example, due to unstable demand, the local power generation and load matching degree decreases, the energy self-sufficiency rate decreases, and the energy complementarity is destroyed, the node's ability to independently respond to changes in power supply status is weakened, especially the ability to maintain stable operation without relying on external coordination, thereby reducing the local coordination capability value; if the risk identification results show that the demand stability is good and the impact on the local energy balance characteristics is small, the node can better utilize local energy resources under a relatively stable demand environment to achieve local energy balance and stable supply, thereby having stronger local coordination capability, and the local coordination capability value remains unchanged or appropriately increased. The risk of local energy balance characteristics is dynamically evaluated through the stability of demand data, and the local coordination capability value is updated in a timely manner to facilitate more accurate and real-time operation management of the power grid, thereby improving the stability and reliability of power grid operation.

[0044] Furthermore, step S300 also includes step S330, performing a computational test on the computing layer of the edge node to generate a test error; step S340, locating the abnormal edge node based on the test error, and matching the neighborhood assistance node according to the positioning result; step S350, using the matched neighborhood assistance node to perform data sharing processing of the abnormal edge node.

[0045] Preferably, the computing layer of the edge node is subjected to a computational test. Specifically, various simulated data or actually collected data samples are input for calculation during the test process. If there is a software fault (such as an algorithm error, a program vulnerability), a hardware fault (such as a processor fault, a memory error) or other abnormal situation (such as a data format mismatch, insufficient resources) in the computing layer, the calculation result will be wrong or the computing task cannot be completed normally, and a test error message will be generated, which usually includes information about the error type, the location where the error occurred (such as a specific line of code or module) and a related error description. Based on the generated test error message, combined with the operation log, system configuration and other information of the edge node, the edge node where the abnormality occurs is accurately found, and then the topology of the power supply network and the connection between the nodes are used to determine the edge node where the abnormality occurs. The system searches for neighboring nodes that are adjacent to the abnormal edge node and have sufficient computing resources, storage resources or data processing capabilities as assisting nodes to handle the related tasks of the abnormal edge node; finally, the neighboring assisting nodes are used to share and process the data of the abnormal edge node. For example, the neighboring nodes back up the shared data to prevent data loss, perform preliminary analysis and processing on the data, such as data cleaning and format conversion, and then send the processed data to other related nodes for further processing, which can make up for the functional defects of the abnormal edge node to a certain extent and ensure the data integrity of the power supply network and the normal operation of the system; at the same time, after the abnormal edge node is repaired, the shared data can be retransmitted back to the node to restore it to normal data processing and operation.

[0046] Furthermore, step S340 also includes step S341, using the positioning result to obtain all neighborhood nodes; step S342, performing independent adaptation analysis on all neighborhood nodes to establish a first adaptation analysis result; step S343, performing adaptation analysis on all neighborhood node combinations to establish a second adaptation analysis result; step S344, if the highest adaptation result in the second adaptation analysis result is higher than the highest adaptation result in the first adaptation analysis result, and meets the combination breakthrough threshold, then the neighborhood node combination corresponding to the highest adaptation result in the second adaptation analysis result is output as a neighborhood assistance node.

[0047] Preferably, based on the positioning results of the test error and the topological structure information of the power supply network, all neighboring nodes that are directly connected to the abnormal edge node or adjacent to it within a certain range are determined, and then an independent adaptation analysis is performed on each obtained neighboring node, that is, the degree of adaptation of each neighboring node when independently processing tasks related to the abnormal edge node (such as data processing, calculation, etc.) is evaluated. Specifically, the computing power (such as processor performance, computing speed, etc.) of the neighboring node, the storage capacity (capacity that can be used to store data), the compatibility with the data format of the abnormal edge node (whether the data of the abnormal edge node can be correctly processed and understood), the current load situation (whether there are sufficient resources to undertake additional tasks), etc. are evaluated, and then an adaptation degree evaluation result is calculated for each neighboring node, and the collection constitutes the first adaptation analysis result. Then analyze the adaptability of the neighborhood node combination, that is, combine all neighborhood nodes in different ways (such as two-node combination, three-node combination, etc.), and perform adaptation analysis on each combination, that is, consider the collaborative cooperation ability between nodes (such as data transmission speed, mutual communication efficiency, rationality of task allocation, etc.), evaluate its overall adaptability when processing tasks related to abnormal edge nodes, and then calculate an adaptation degree evaluation for each combination as the second adaptation analysis result.

[0048] Preferably, the highest adaptation evaluation in the first adaptation analysis result and the second adaptation analysis result are compared. If the highest adaptation result in the second adaptation analysis result (i.e., the evaluation of the combination with the highest degree of adaptation among all neighborhood node combinations) is higher than the highest adaptation result in the first adaptation analysis result (i.e., the evaluation of the node with the highest degree of adaptation when all neighborhood nodes are calculated independently), it indicates that there are some neighborhood node combinations that are more capable of processing tasks than a single node when processing independently. At the same time, it is judged whether this highest adaptation result meets the combination breakthrough threshold, wherein the combination breakthrough threshold is a pre-set standard used to measure whether the degree of adaptation of the neighborhood node combination is high enough. If the combination breakthrough threshold is met, it means that the neighborhood node combination is not only better than the adaptability of a single node, but also reaches a certain level of excellence. Then, the neighborhood node combination corresponding to the highest adaptation result in the second adaptation analysis result is determined as a neighborhood assistance node and outputted for tasks such as data sharing processing of abnormal edge nodes, so as to better solve the problem of abnormal edge nodes and ensure the normal operation of the power supply network.

[0049] Step S400 : performing edge fusion analysis based on the local collaborative capability value by utilizing the region division within the initial tree structure diagram to establish a first collaborative control scheme.

[0050] Preferably, the initial tree structure diagram divides the power grid system into different regions or node sets, and performs edge fusion analysis based on local coordination capability values. That is, for each region divided by the initial tree structure diagram, the local coordination capability values ​​of each edge node in the region are comprehensively analyzed. This may include statistical analysis of these values, such as calculating the average value, standard deviation, etc., to understand the overall coordination capability level in the region; it may also analyze the differences and distribution of local coordination capability values ​​between different nodes to find the dominant nodes and relatively weak nodes in the region; and analyze how edge nodes interact and coordinate with nodes in adjacent regions. For example, analyze the impact of the local coordination capability value of edge nodes at the junction of two regions on power transmission and coordinated operation between the two regions. Through edge fusion analysis, a comprehensive understanding of the coordination capability status of the entire power supply system at different regional levels can be obtained, and potential problems and optimization space can be discovered. Then, based on the results of the edge fusion analysis, the goals of collaborative control are determined, such as improving the stability of the power system, optimizing power distribution to reduce losses, and enhancing the power system's ability to respond to emergencies. Then, based on the analysis results and goals, specific control strategies and measures are formulated, which may include adjusting the control instructions of edge nodes in different areas, such as reasonably allocating load regulation tasks according to local collaborative capability values, determining the charging and discharging strategies of energy storage equipment, and optimizing the operating parameters of local power generation equipment. These control strategies and measures are used to establish the first collaborative control scheme, thereby achieving more accurate and effective management and control of the power system and improving the overall performance and reliability of the power system.

[0051] Step S500: Perform regional load analysis on the first collaborative control scheme to generate a load anomaly indicator.

[0052] Preferably, the load conditions of each area in the power supply system are analyzed according to the first collaborative control scheme to identify possible abnormal load conditions. Specifically, the load data in each area is collected, including but not limited to the real-time power consumption of each edge node, the power consumption over a period of time, the load change trend, etc.; different areas have different functions (such as commercial areas, residential areas, industrial areas, etc.), and their load characteristics will be significantly different. The characteristics of the load in each area are analyzed, such as the time and size of the peak and valley values ​​of the load, the frequency and amplitude of the load fluctuation, etc., to understand the load change trend of each area under normal conditions; the load data is then compared with the load distribution strategy and expected targets in the first collaborative control scheme to check whether the actual load is distributed and adjusted in accordance with the requirements of the control scheme, for example, whether the load in some areas exceeds the range set by the scheme, or the load adjustment in some areas fails to achieve the expected effect; at the same time, the impact of the control scheme on the load during implementation in different areas is analyzed to determine whether the scheme effectively achieves the optimal distribution and balance of the load. Then, based on the historical data of regional load, load characteristics and the goals of the first collaborative control scheme, load abnormality thresholds are set for each region, which may include power thresholds, load change rate thresholds, etc.; finally, the actual monitored load data is compared with the set thresholds. If the load data of a certain area exceeds the corresponding threshold, it is determined that a load abnormality has occurred in the area, and a load abnormality mark is generated for it to clearly indicate that there is a load abnormality in the area so that corresponding measures can be taken for adjustment and optimization to ensure the safe and stable operation of the power supply system.

[0053] Step S600: Use the load anomaly identifier and the topology coordination table to perform structural update optimization of the initial tree structure diagram, establish a second coordinated control scheme, and output a final coordinated control scheme after distinguishing the first coordinated control scheme and the second coordinated control scheme.

[0054] Preferably, based on the load anomaly identification and combined with the topology coordination table, the impact of the abnormal load on the topology structure of the entire power system is determined, and the structure of the initial tree structure diagram is updated and optimized, which may include adjusting the division of the area, changing the connection relationship between the nodes, or redistributing the roles and functions of each area in the system. For example, if the load anomaly in a certain area is caused by poor coordination with the adjacent areas, consider optimizing the connection method between them or adjusting the coordination strategy to better balance the load, so as to more effectively deal with the load anomaly and improve the overall stability and reliability; then according to the new structure and system operation status, that is, combined with the updated area division, node connection relationship and load anomaly, redesign the load distribution strategy, energy storage equipment control strategy, local power generation regulation strategy, etc., to clarify how to better coordinate power, support each other and coordinate operation between each area and node under different load conditions, and finally obtain a second coordinated control scheme to ensure the stable operation of the power system.

[0055] Preferably, the first collaborative control scheme and the second collaborative control scheme are comprehensively evaluated with system stability, load balance, energy efficiency, and the ability to cope with emergencies as evaluation indicators. Specifically, by simulating different operating scenarios, the performance of the two schemes in various situations is compared; then, considering the results of various evaluation indicators comprehensively, the two schemes are judged to determine which scheme can better meet the operating requirements of the power supply system as a whole and can more effectively solve current problems, such as load anomalies and poor coordination. For example, during the simulated load peak period, evaluate which scheme can better maintain system voltage stability and avoid overload; when simulating sudden failures, evaluate which scheme can restore the normal operation of the system faster; finally, based on the judgment results, select the scheme with better performance as the final collaborative control scheme, thereby achieving efficient, stable and reliable operation of the power system.

[0056] Furthermore, step S600 also includes step S610, using the load anomaly identifier to perform inversion positioning of the abnormal area; step S620, querying the topology collaboration table, and constructing a candidate sub-topology set based on the inversion positioning of the abnormal area and the query results; step S630, establishing a candidate tree structure diagram with all candidate sub-topology sets, and performing control fitting simulation on each candidate tree structure diagram; step S640, using a multi-index scoring matrix to output the optimal control scheme for each candidate tree structure diagram, and screening all the optimal control schemes, and outputting the screening results as the second collaborative control scheme.

[0057] Preferably, the load anomaly identification is used, combined with the regional geographical location of the power supply system, the power network connection relationship, etc., to reversely deduce and accurately determine the specific regional location and scope of the anomaly, and clarify which specific substations, lines or power consumption areas have loads exceeding expectations, abnormal fluctuations, etc., and then query the topology coordination table to obtain all connection information related to the abnormal area, including other areas connected to it, the transmission capacity of the lines, the functions of the nodes, etc., and construct a candidate sub-topology set based on the results of the inversion positioning of the abnormal area and the query information of the topology coordination table, that is, from the power system topology structure, extract the parts related to the abnormal area and combine them into different subsets, each subset representing a local topology structure, for example, the abnormal area and its directly connected areas are combined into a candidate sub-topology set, or the abnormal area and its upstream and downstream key nodes and lines are combined into another candidate sub-topology set to explore different optimization schemes.

[0058] Preferably, all candidate sub-topology sets are converted into candidate tree structure diagrams, that is, a node is selected as the root node, and other nodes and edges are connected to form a tree structure to more clearly show the hierarchical relationship and dependency relationship between the elements in the candidate sub-topology set; then each candidate tree structure diagram is subjected to control fitting simulation, including simulating the operation of the power system under the candidate tree structure under a given control strategy, and calculating multiple operating indicators such as voltage distribution, current size, power loss, etc. through fitting simulation to evaluate the performance of the candidate tree structure diagram in actual operation. For example, by changing the output of the generator, adjusting the tap of the transformer, controlling the input of reactive compensation equipment and other control operations, the operating status of the system under different control parameters is observed, and the control parameter combination that best suits the candidate tree structure diagram is found to achieve optimal or better operation.

[0059] Preferably, a multi-index scoring matrix is ​​established using candidate tree structure diagrams as matrix rows and evaluation indicators (such as system stability, voltage qualification rate, power loss, load balance, etc.) as matrix columns to evaluate the control fitting simulation results of each candidate tree structure diagram. The score of each candidate tree structure diagram under the corresponding evaluation indicator is used as a matrix element, that is, by analyzing and calculating the control fitting simulation results, each candidate tree structure diagram is scored on each indicator, thereby comprehensively evaluating the performance of each candidate tree structure diagram; then, based on the multi-index scoring matrix, the optimal control scheme is determined for each candidate tree structure diagram, for example, by comprehensively considering the weights of each evaluation indicator, the comprehensive score of each candidate tree structure diagram is calculated, and then the control scheme with the highest comprehensive score is selected as the optimal control scheme for the candidate tree structure diagram; finally, the optimal control schemes of all candidate tree structure diagrams are screened, and schemes that perform poorly on certain key indicators are excluded, or the schemes are adjusted according to actual operating conditions and limitations, thereby obtaining a second coordinated control scheme for adjusting the system operation mode, optimizing load distribution, improving system stability and reliability, etc., and ensuring efficient and coordinated operation of the power system.

[0060] Furthermore, step S600 also includes step S650, using the first collaborative control scheme to perform control fitting of the power supply network, and using the first control fitting result to establish a first control score; step S660, executing the control fitting of the second collaborative control scheme, and using the second control fitting result to establish a second control score; step S670, obtaining an updated tree structure diagram mapped by the second collaborative control scheme, and establishing a structural change cost based on the updated tree structure diagram and the initial tree structure diagram; step S680, using the first control score, the second control score, and the structural change cost to select and judge the first collaborative control scheme and the second collaborative control scheme.

[0061] Preferably, the first collaborative control scheme is applied to the power supply network model for simulation operation. By adjusting various control parameters in the power supply network, such as generator output, transformer tap position, and the input amount of reactive compensation devices, the operating state of the power supply network is made as close as possible to the target state set by the first collaborative control scheme. Then, based on various operating indicators and data obtained during the control fitting process, such as voltage stability, frequency deviation, power loss, load balance, etc., the implementation effect of the first collaborative control scheme is quantitatively evaluated, thereby obtaining a first control score, which is used to measure the performance of the first collaborative control scheme in power supply network control. Similarly, the second collaborative control scheme is controlled and fitted on the power supply network model. By adjusting the corresponding control parameters, the power supply network is operated in accordance with the requirements of the second collaborative control scheme. Then, based on the operating data and indicators obtained during the second control fitting process, the performance of voltage, frequency, power loss, etc. is considered to establish a second control score to quantitatively evaluate the effect of the second collaborative control scheme in power supply network control.

[0062] Preferably, the power grid topology is mapped according to the second coordinated control scheme to form an updated tree structure diagram that is different from the initial tree structure diagram, reflecting the structural changes of the power supply network under the second coordinated control scheme, including the connection relationship between nodes, the open and closed status of lines, etc. By comparing the updated tree structure diagram with the initial tree structure diagram, the changes in the power grid structure are analyzed, such as which lines have been added or removed, which node connection methods have been changed, etc., and the costs of these structural changes are evaluated, which may include equipment modification costs, line construction costs, losses caused by power outages, etc., and then a quantitative indicator that can reflect the economic and operational costs of the structural changes is established, namely the structural change cost. Finally, the first control score, the second control score, and the structural change cost are comprehensively compared and evaluated between the first coordinated control scheme and the second coordinated control scheme. Specifically, if the first control score and the second control score are similar, but the structural change cost of the second coordinated control scheme is too high, the first coordinated control scheme is preferred; conversely, if the second control score is significantly better than the first control score and the structural change cost is within an acceptable range, the second coordinated control scheme is selected to obtain better power supply network control effect, thereby selecting the coordinated control scheme that is most suitable for power supply network operation.

[0063] Furthermore, step S650 also includes step S651, sending the first control fitting result to the evaluation channel connected to the initial tree structure diagram; step S652, calling the dynamic prediction sub-channel in the evaluation channel, using the dynamic prediction sub-channel to predict the trend change control result performance of energy storage and load, and generating a dynamic prediction score; step S653, calling the local coordination sub-channel in the evaluation channel, using the local coordination sub-channel to evaluate the power coordination degree under the first control fitting result, and establish a coordination score; step S654, calling the control evaluation sub-channel in the evaluation channel, using the control evaluation sub-channel to evaluate the delay, energy consumption, reliability, and load matching degree under the first control fitting result, and establish a control evaluation score; step S655, outputting the first control score according to the dynamic prediction score, coordination score, and control evaluation score.

[0064] Preferably, the first control fitting result is various data and operating indicators obtained by the first collaborative control scheme during the power supply network control fitting process, including but not limited to real-time operating data such as voltage, current, power, frequency, and adjustment of control parameters such as generator output, transformer tap position, and reactive compensation device status. It is sent to an evaluation channel connected to the initial tree structure diagram, wherein the evaluation channel is a module for multi-dimensional evaluation of the power supply network control scheme, which can obtain information related to the power supply network topology structure and conduct a more comprehensive and accurate evaluation of the control fitting results.

[0065] Preferably, the dynamic prediction sub-channel is a unit in the evaluation channel for predicting the future operating trend of the power supply network. It uses relevant data in the first control fitting result, such as the current load level, the charging and discharging status of the energy storage device, etc., as well as the historical operating data and prediction model of the power supply network (such as that constructed based on time series analysis, machine learning algorithms, etc.) to predict the trend changes of energy storage and load in the future. For example, it predicts whether the load will peak or trough in the next few hours or days, whether the energy storage device can meet the needs of load changes, and how the control scheme performs in responding to these changes. Then, based on the prediction results, the performance of the first control fitting result in terms of energy storage and load trend changes is quantitatively scored to generate a dynamic prediction score to reflect the ability of the control scheme to adapt to future dynamic changes in energy storage and load.

[0066] Preferably, the local coordination sub-channel mainly evaluates the power coordination between local areas in the power supply network. Under the first control fitting result, different areas in the power supply network may have different power demands and supply conditions. The local coordination sub-channel is used to analyze the power transmission, distribution and coordination between various areas, and evaluate the degree of coordination between them, such as checking the power balance between generators, load centers, energy storage equipment, etc. in different areas, as well as the power loss and stability during transmission. If the various areas can effectively coordinate with each other, so that the power distribution is reasonable, the transmission is stable and the loss is small, it means that the first control fitting result performs well in local power coordination, and a higher coordination score is given; on the contrary, if there are problems such as power conflicts and unstable transmission between areas, the score will be lowered accordingly.

[0067] Preferably, the control evaluation sub-channel is a unit that evaluates the multi-faceted control performance of the first control fitting result, and evaluates the effectiveness of the control scheme from multiple angles such as delay, energy consumption, reliability and load matching. Specifically, it mainly considers the time delay from the issuance of the control signal to its actual effectiveness, including equipment response time, signal transmission delay, etc.; analyzes the energy consumption of the power supply network during operation under the first control fitting result, including the energy consumption of the power generation equipment, line transmission loss, etc.; evaluates the operational reliability of the power supply network under the control scheme, such as whether the system can operate stably under various working conditions, and whether it is prone to power outages, failures, etc.; evaluates the degree of matching between the power supply capacity of the power supply network and the actual load demand. If the power supply can accurately meet the load demand, neither over-powering nor under-powering, it means that the load matching is high; and then comprehensively evaluates the results to obtain a control evaluation score to comprehensively reflect the pros and cons of the first control fitting result in terms of control performance. Finally, the dynamic prediction score, coordination score and control evaluation score are comprehensively processed to obtain the first control score. For example, a corresponding weight is assigned to each score according to its importance, and then they are weighted and summed to obtain the final first control score. If the control evaluation score is considered to be the most important for measuring the pros and cons of the control plan, a higher weight is assigned to it, while the dynamic prediction score and coordination score are relatively secondary and are assigned lower weights.

[0068] Furthermore, step S600 also includes step S690, performing collaborative scheduling of the power grid based on the final collaborative control scheme and establishing a feedback verification data set; step S6100, using the feedback verification data set to evaluate the effect of collaborative scheduling, and performing collaborative scheduling optimization based on the effect evaluation feedback.

[0069] Preferably, the final collaborative control scheme is used to carry out collaborative dispatching of the power grid, including coordinated control and operation of power sources, energy storage devices, transmission lines, load nodes, etc. in the power grid according to the strategies and parameters in the scheme. For example, according to the load demand of different regions, the power generation of the power source is reasonably allocated, the charging and discharging of the energy storage device is controlled, and the power transmission of the transmission line is optimized to achieve stable and efficient operation of the power grid; then, the real-time operating data such as voltage, current, power, etc. of each node in the power grid, as well as the output of the power source, the status of the energy storage device, the change of the load, etc. are collected to establish a feedback verification data set to reflect the actual operation of the collaborative dispatching of the power grid.

[0070] Preferably, the effect of collaborative scheduling is evaluated using the established feedback verification data set. Specifically, by calculating the power supply reliability indicators of the power grid, such as power outage time and number of power outages, the impact of collaborative scheduling on ensuring the stability of power supply is evaluated; by analyzing the energy consumption indicators of the power grid, such as line loss and equipment energy consumption, the effect of collaborative scheduling on energy saving is evaluated; the degree of matching between the actual load and the scheduling plan can also be evaluated to measure the ability of collaborative scheduling to meet user electricity demand. Based on the results of the effect evaluation, the problems and shortcomings in the collaborative scheduling process are analyzed. For example, if the load in a certain area is often overloaded, the power distribution strategy of the area is adjusted or the transmission capacity of the transmission line is optimized; if the energy consumption indicator is too high, the charging and discharging control of the energy storage device is further optimized to reduce line loss; then, based on the analysis results, the collaborative scheduling plan is targeted for optimization and adjustment, such as modifying the control parameters, adjusting the scheduling strategy, etc., so that the collaborative scheduling of the power grid can be continuously improved, and the performance and effect of the collaborative scheduling of the power grid can be continuously improved through feedback optimization.

[0071] In the above, refer to Figure 1 The power supply collaborative control method based on sensor edge computing according to an embodiment of the present invention is described in detail. Figure 2 The present invention describes a power supply collaborative control system based on sensor edge computing according to an embodiment of the present invention.

[0072] The power supply collaborative control system based on sensor edge computing according to the embodiment of the present invention is used to solve the technical problems existing in the prior art, such as high data transmission delay, inflexible centralized control response, high risk of regional load imbalance, and unreasonable grid resource allocation, and achieves the technical effects of reducing data transmission delay, improving the real-time performance and flexibility of power supply control, and realizing accurate and efficient configuration of grid resources. Figure 2 As shown, the power supply collaborative control system based on sensor edge computing includes: a power grid data set acquisition module 10, a real-time power supply data set establishment module 20, a local collaborative capability value calculation module 30, an edge fusion analysis module 40, a regional load analysis module 50, and a collaborative control scheme output module 60.

[0073] The power grid data set acquisition module 10 is used to acquire power grid data sets and establish an initial tree structure diagram and a topology coordination table. The power grid data set includes power grid physical topology, regional division, and node capability data. The real-time power supply data set establishment module 20 is used to perform real-time acquisition of power supply data by edge nodes and establish a real-time power supply data set. The real-time power supply data set includes demand data, load data, and energy storage data. The local coordination capability value calculation module 30 is used to activate the computing layer of the edge node, calculate the local coordination capability value based on the real-time power supply data set, and update the local coordination capability value to the initial tree structure diagram. The edge fusion analysis module 40 is used to perform edge fusion analysis based on the local coordination capability value using the regional division within the initial tree structure diagram to establish a first coordinated control scheme. The regional load analysis module 50 is used to perform regional load analysis on the first coordinated control scheme and generate a load anomaly identifier. The coordinated control scheme output module 60 is used to use the load anomaly identifier and the topology coordination table to perform structural optimization of the initial tree structure diagram, establish a second coordinated control scheme, and output a final coordinated control scheme after distinguishing the first coordinated control scheme and the second coordinated control scheme.

[0074] The specific configuration of the collaborative control scheme output module 60 will be described in detail below. The collaborative control scheme output module 60 further includes: performing control fitting of the power supply network using the first collaborative control scheme, and establishing a first control score using the first control fitting result; performing control fitting of the second collaborative control scheme, and establishing a second control score using the second control fitting result; obtaining an updated tree structure diagram of the second collaborative control scheme mapping, and establishing a structural change cost based on the updated tree structure diagram and the initial tree structure diagram; and selecting and discriminating between the first collaborative control scheme and the second collaborative control scheme using the first control score, the second control score, and the structural change cost.

[0075] The specific configuration of the collaborative control scheme output module 60 will be described in detail below. The collaborative control scheme output module 60 further includes: sending the first control fitting result to an evaluation channel connected to the initial tree structure diagram; calling the dynamic prediction sub-channel in the evaluation channel, using the dynamic prediction sub-channel to predict the trend change control result performance of energy storage and load, and generating a dynamic prediction score; calling the local collaborative sub-channel in the evaluation channel, using the local collaborative sub-channel to evaluate the power collaborative coordination under the first control fitting result, and establishing a collaborative coordination score; calling the control evaluation sub-channel in the evaluation channel, using the control evaluation sub-channel to evaluate the delay, energy consumption, reliability, and load matching under the first control fitting result, and establishing a control evaluation score; and outputting a first control score based on the dynamic prediction score, collaborative coordination score, and control evaluation score.

[0076] The specific configuration of the local collaborative capability value calculation module 30 will be described in detail below. The local collaborative capability value calculation module 30 further includes: establishing a local collaborative capability evaluation feature set, which includes load flexibility features, energy storage support features, and local energy balance features; and using this evaluation feature set to calculate the local collaborative capability value of a real-time power supply dataset.

[0077] The specific configuration of the local collaborative capability calculation module 30 will be described in detail below. The local collaborative capability calculation module 30 further includes: performing demand stability prediction on the demand data to generate a demand stability impact factor; utilizing the demand stability impact factor to identify risks associated with local energy balance characteristics within the evaluation feature set; and updating the local collaborative capability value using the risk identification results.

[0078] The specific configuration of the collaborative control solution output module 60 will be described in detail below. The collaborative control solution output module 60 further includes: using the load anomaly identifier to invert and locate the abnormal area; querying the topology collaborative table, and constructing a candidate sub-topology set based on the inverted location of the abnormal area and the query results; establishing a candidate tree structure diagram based on all candidate sub-topology sets, and performing a control fitting simulation on each candidate tree structure diagram; using a multi-index scoring matrix to output the optimal control solution for each candidate tree structure diagram, and screening all optimal control solutions, and outputting the screening results as the second collaborative control solution.

[0079] The specific configuration of the local collaborative capability value calculation module 30 will be described in detail below. The local collaborative capability value calculation module 30 further includes: performing computational testing on the edge node's computing layer to generate a test error; locating the abnormal edge node based on the test error; matching a neighboring assisting node based on the positioning result; and utilizing the matched neighboring assisting node to perform data sharing processing on the abnormal edge node.

[0080] The specific configuration of the local collaborative capability value calculation module 30 will be described in detail below. The local collaborative capability value calculation module 30 further includes: using the positioning results to obtain all neighboring nodes; performing an adaptation analysis on all neighboring nodes independently to establish a first adaptation analysis result; performing an adaptation analysis on all neighboring nodes in combination to establish a second adaptation analysis result; if the highest adaptation result in the second adaptation analysis result is higher than the highest adaptation result in the first adaptation analysis result and meets the combination breakthrough threshold, then outputting the neighboring node combination corresponding to the highest adaptation result in the second adaptation analysis result as a neighborhood assisting node.

[0081] The specific configuration of the coordinated control scheme output module 60 will be described in detail below. The coordinated control scheme output module 60 further includes: performing coordinated dispatch of the power grid based on the final coordinated control scheme and establishing a feedback verification dataset; evaluating the effectiveness of the coordinated dispatch using the feedback verification dataset; and optimizing the coordinated dispatch based on the feedback from the evaluation.

[0082] The power supply collaborative control system based on sensor edge computing provided in an embodiment of the present invention can execute the power supply collaborative control method based on sensor edge computing provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0083] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other and are not used to limit the scope of protection of the present invention.

[0084] The above specific embodiments do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application shall be included within the scope of protection of this application.

Claims

1. A power supply collaborative control method based on sensor edge computing, characterized in that: The method comprises: Collecting a power grid data set, and establishing an initial tree structure diagram and a topology coordination table, wherein the power grid data set includes power grid physical topology, regional division, and node capability data; The edge node performs real-time collection of power supply data to establish a real-time power supply data set, which includes demand data, load data, and energy storage data; activating a computing layer of an edge node, calculating a local collaborative capability value according to the real-time power supply data set, and updating the local collaborative capability value to the initial tree structure diagram; Performing edge fusion analysis based on the local collaborative capability value by utilizing the regional division within the initial tree structure diagram to establish a first collaborative control scheme; Performing regional load analysis on the first coordinated control scheme to generate a load anomaly indicator; The load anomaly identifier and the topology coordination table are used to perform structural update optimization of the initial tree structure diagram, establish a second coordinated control scheme, and output a final coordinated control scheme after distinguishing the first coordinated control scheme and the second coordinated control scheme.

2. The power supply collaborative control method based on sensor edge computing according to claim 1, characterized in that: After determining the first collaborative control scheme and the second collaborative control scheme, outputting a final collaborative control scheme includes: Performing control fitting of the power supply network using the first coordinated control scheme, and establishing a first control score using the first control fitting result; performing a control fit of the second coordinated control scheme, and establishing a second control score using a second control fit result; Obtaining an updated tree structure diagram of the second coordinated control scheme mapping, and establishing a structure change cost according to the updated tree structure diagram and the initial tree structure diagram; The first control score, the second control score, and the structural change cost are used to select and judge the first cooperative control scheme and the second cooperative control scheme.

3. The power supply collaborative control method based on sensor edge computing according to claim 2, characterized in that: The step of establishing a first control score using the first control fitting result includes: sending the first control fitting result to an evaluation channel connected to the initial tree structure graph; Invoking a dynamic prediction sub-channel in the evaluation channel, using the dynamic prediction sub-channel to predict the performance of energy storage and load trend change control results, and generating a dynamic prediction score; Invoking a local coordination sub-channel in the evaluation channel, using the local coordination sub-channel to evaluate the power coordination degree under the first control fitting result, and establishing a coordination score; Invoking a control evaluation subchannel in the evaluation channel, using the control evaluation subchannel to evaluate the delay, energy consumption, reliability, and load matching degree under the first control fitting result, and establishing a control evaluation score; A first control score is output according to the dynamic prediction score, the coordination score, and the control evaluation score.

4. The power supply collaborative control method based on sensor edge computing according to claim 1, characterized in that: The calculating the local coordination capability value according to the real-time power supply data set includes: Establishing a local collaborative capability evaluation feature set, the evaluation feature set including load flexibility features, energy storage support features, and local energy balance features; The evaluation feature set is used to calculate the local collaborative capability value of the real-time power supply data set.

5. The power supply collaborative control method based on sensor edge computing according to claim 4, characterized in that: The calculating of the local coordination capability value of the real-time power supply data set by using the evaluation feature set further includes: Performing demand stability prediction on the demand data to generate a demand stability influencing factor; The demand stability influencing factor is used to evaluate the risk identification of local energy balance characteristics in the feature set, and the risk identification result is used to update the local collaborative capability value.

6. The power supply collaborative control method based on sensor edge computing according to claim 1, characterized in that: The method of using the load anomaly identifier and the topology coordination table to perform structural update and optimization of the initial tree structure graph to establish a second coordinated control scheme includes: Using the load anomaly marker to perform inversion positioning of the abnormal area; Querying the topology coordination table, and constructing a candidate sub-topology set according to the inversion positioning of the abnormal area and the query result; Establishing a candidate tree structure graph with all candidate sub-topology sets, and performing a control fitting simulation on each candidate tree structure graph; The multi-index scoring matrix is ​​used to output the optimal control scheme for each candidate tree structure diagram, and all the optimal control schemes are screened, and the screening results are output as the second collaborative control scheme.

7. The power supply collaborative control method based on sensor edge computing according to claim 1, characterized in that: The computing layer of the activated edge node includes: Perform computational tests on the computing layer of edge nodes and generate test errors; Locating abnormal edge nodes based on the test error, and matching neighboring assisting nodes according to the positioning results; The matched neighborhood assisting nodes are used to share data of abnormal edge nodes.

8. The power supply collaborative control method based on sensor edge computing according to claim 7, characterized in that: Matching the neighborhood assisting node according to the positioning result includes: Obtain all neighboring nodes using the positioning result; Performing independent adaptation analysis on all neighboring nodes to establish a first adaptation analysis result; Performing adaptation analysis of all neighborhood node combinations to establish a second adaptation analysis result; If the highest adaptation result in the second adaptation analysis result is higher than the highest adaptation result in the first adaptation analysis result and meets the combination breakthrough threshold, the neighborhood node combination corresponding to the highest adaptation result in the second adaptation analysis result is output as a neighborhood assisting node.

9. The power supply collaborative control method based on sensor edge computing according to claim 1, characterized in that: After the final collaborative control solution is output, it includes: Performing coordinated dispatch of the power grid based on the final coordinated control scheme and establishing a feedback verification data set; The feedback verification data set is used to evaluate the effect of collaborative scheduling, and collaborative scheduling optimization is performed based on the effect evaluation feedback.

10. A power supply collaborative control system based on sensor edge computing, characterized in that: The system is used to implement the power supply collaborative control method based on sensor edge computing according to any one of claims 1 to 9, and the system includes: A power grid data set acquisition module, configured to acquire power grid data sets and establish an initial tree structure diagram and a topology coordination table. The power grid data sets include power grid physical topology, regional division, and node capability data. A real-time power supply data set establishment module is used to perform real-time collection of power supply data by edge nodes and establish a real-time power supply data set, wherein the real-time power supply data set includes demand data, load data, and energy storage data; A local collaborative capability value calculation module is used to activate the computing layer of the edge node, calculate the local collaborative capability value according to the real-time power supply data set, and update the local collaborative capability value to the initial tree structure diagram; an edge fusion analysis module, configured to perform edge fusion analysis based on the local collaborative capability value by utilizing the regional division within the initial tree structure diagram, and establish a first collaborative control scheme; a regional load analysis module, configured to perform regional load analysis on the first coordinated control scheme and generate a load anomaly indicator; The collaborative control scheme output module is used to use the load anomaly identifier and the topology collaborative table to perform structural update optimization of the initial tree structure diagram, establish a second collaborative control scheme, and output the final collaborative control scheme after judging the first collaborative control scheme and the second collaborative control scheme.

Images