A method and system for classifying power grid equipment disconnection based on multi-dimensional data analysis

Abstract

The present invention discloses a method and system for classifying power grid equipment disconnection based on multi-dimensional data analysis, which relates to the technical field of power grid data processing. It includes monitoring whether disconnection occurs in the target power grid through topological structure information and determining the disconnection type, laying a solid foundation for subsequent sub-network islanding risk assessment and ensuring the reliability of subsequent evaluation and analysis. The islanding risk of each sub-network is evaluated according to the topological structure diagram of the sub-network, the type of disconnection, and the data of the power grid. Then, similarity indicators for classification criteria are generated by combining the stability indicators of the sub-network and the cooperation indicators between devices. The similarity indicators are defined based on the islanding risk and equipment conditions of the sub-network, improving the adaptability and reliability of sub-network classification. A resource allocation strategy is generated for the classified sub-networks, and different resource allocation strategies are formulated for sub-networks in different situations, thereby ensuring the reliability of the operation of sub-network equipment after disconnection.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid data processing, and particularly to a power grid equipment disconnection classification method and system based on multi-dimensional data analysis. Background Art

[0002] The disconnection of power grid equipment, usually simply referred to as "disconnection or splitting", means that when the power system is disturbed and its stability is damaged, and the synchronization between the generator and other parts of the power system, or between a part of the system and other parts of the system is lost and cannot be restored, the electrical connection between them is cut off and decomposed into independent and non-connected parts to prevent the accident from expanding and causing serious consequences. Disconnection is an important means to ensure the safe and stable operation of the power system.

[0003] In the prior art, after the disconnection of power grid equipment, many subnets may be generated. Because the specific equipment conditions of each subnet are relatively complex, it is impossible to accurately and adaptively classify the subnets, resulting in poor operation stability of the power grid equipment in the subsequent subnets.

[0004] Therefore, how to improve the accuracy and adaptability of the classification between subnets after disconnection is a technical problem to be solved at present. Summary of the Invention

[0005] The purpose of the present invention is to solve the problem of low accuracy and adaptability in the classification between subnets after disconnection in the prior art, and to propose a power grid equipment disconnection classification method based on multi-dimensional data analysis, which includes:

[0006] Obtain the topological structure information of the target power grid, monitor whether a disconnection occurs in the target power grid through the topological structure information. After a disconnection occurs in the target power grid, obtain the operation records before and after the disconnection of the target power grid, and determine the type of disconnection according to the operation records before and after the disconnection of the target power grid;

[0007] Obtain the basic equipment information of the target power grid, determine the disconnection point location according to the operation records before and after the disconnection of the target power grid and the type of disconnection, rely on the topological structure information, basic equipment information and disconnection point location of the target power grid to construct the topological structure diagrams of all subnets after the disconnection of the target power grid, and evaluate the islanding risk of each subnet by analyzing the topological structure diagrams of each subnet according to the type of disconnection;

[0008] Obtain the equipment operation information of each equipment before and after the disconnection in the subnet topological structure diagram, and analyze the equipment operation information of each equipment to evaluate the stability index of each equipment and the cooperation index between the equipment. Determine the stability index of the subnet through the stability index of the equipment, and generate the similarity index of the subnet by combining the stability index of the subnet, the cooperation index between the equipment and the islanding risk;

[0009] Classify all sub - grids in the target power grid according to the similarity index of the sub - grids, generate a resource allocation strategy for the classified sub - grids, and optimize the stable operation of the equipment in the target power grid.

[0010] In some embodiments of the present application, monitor whether a disconnection occurs in the target power grid through topology structure information, including,

[0011] Construct the standard topology structure of the target power grid according to the topology structure information of the target power grid, and traverse the nodes, lines, and equipment in the target power grid at fixed intervals to construct the real - time topology structure of the target power grid. Compare the differences between the standard topology structure and the real - time topology structure of the target power grid to determine the topology difference degree;

[0012] Obtain the real - time power grid key parameters of each node on the standard topology structure of the target power grid, and determine the disconnection matching degree through the real - time power grid key parameters of each node on the standard topology structure and the preset disconnection characteristic pattern;

[0013] Comprehensively consider the topology difference degree and the disconnection matching degree to monitor whether a disconnection occurs in the target power grid.

[0014] In some embodiments of the present application, determine the type of disconnection according to the operation records before and after the disconnection of the target power grid, including,

[0015] Determine the disconnection time node, and convert the operation records within a period of time before and after the disconnection time node into an operation timeline. The operation timeline is a horizontal axis listed in chronological order, and all operation events are marked on this horizontal axis;

[0016] The operation timeline includes two types of timelines: the protection operation timeline and the non - protection operation timeline, and the time scales of the protection operation timeline and the non - protection operation timeline are aligned. Compare and comprehensively consider the protection operation timeline and the non - protection operation timeline to determine the type of disconnection.

[0017] In some embodiments of the present application, determine the disconnection point location according to the operation records before and after the disconnection of the target power grid and the type of disconnection, including,

[0018] The type of disconnection includes active disconnection and passive disconnection;

[0019] For active disconnection, determine the disconnection point location according to the non - protection operation timeline of the target power grid;

[0020] For passive disconnection, conduct topology analysis according to the difference between the standard topology structure and the real - time topology structure of the target power grid and the protection operation timeline, so as to screen out the set of disconnection point locations. Analyze the changes in the key power grid parameters of each disconnection point in the set of disconnection point locations before and after the disconnection to determine the disconnection point location.

[0021] In some embodiments of the present application, the islanding risk of each subnet is evaluated by analyzing the topological structure diagram of each subnet according to the type of disconnection, including

[0022] identifying key nodes and key lines in the topological structure diagram of each subnet, evaluating the scale index, load index, type of power supply, and number of power supplies of the subnet through the key nodes and key lines, and evaluating the power redundancy of the subnet in combination with the type and number of power supplies;

[0023] evaluating the islanding risk level of the subnet according to the scale index, load index, power redundancy, and type of disconnection of the subnet, and describing the islanding risk situation of the subnet through the islanding risk level;

[0024] ;

[0025] wherein, is the islanding risk level of the th subnet, is the risk conversion coefficient of the load index of the th subnet, is the load index of the th subnet, is the risk conversion coefficient of the power redundancy of the th subnet, is the power redundancy of the th subnet, is the scale index of the th subnet, is a constant determined by the type of disconnection, is a preset constant, and [] is the rounding symbol.

[0026] In some embodiments of the present application, the device operation information of each device is analyzed to evaluate the stability index of each device and the cooperation index between devices, including

[0027] dividing the device operation information before and after the disconnection of each device in the subnet topological structure diagram into the device operation information in the complete period before and after the disconnection and the device operation information in the period after the disconnection according to the time node;

[0028] calculating the fluctuation amount of each operation index of the device according to the device operation information in the complete period before and after the disconnection, calculating the real-time amount of each operation index of the device according to the device operation information in the period after the disconnection, obtaining the failure rate of each device, and determining the stability index of the device by comprehensively considering the operation index fluctuation amount, operation index real-time amount, and failure rate of the device;

[0029] ;

[0030] wherein, is the stability index of the th device, is the number of operation indexes in the th device, is the stable weight of the th operation index of the th device, is the real-time quantity of the th operation index of the th device, is the fluctuation quantity of the th operation index of the th device, is a preset constant, is the failure rate of the th device, is the failure stability influence coefficient of the th device, is the failure stability influence coefficient mapped from the failure rate of the th device;

[0031] Screen out the coordination information between devices from the device operation information before and after device disconnection, and determine the cooperation index between devices according to the coordination information.

[0032] In some embodiments of the present application, determining the cooperation index between devices according to the coordination information includes,

[0033] Calculate the coefficient of variation of each cooperation index, and integrate the coefficient of variation according to multiple devices and multiple cooperation index categories, and describe the cooperation index between devices in the subnet through the integrated coefficient of variation.

[0034] In some embodiments of the present application, combining the stability index of the subnet, the cooperation index between devices and the island risk to generate a similarity index of the subnet includes,

[0035] Plan subnets with the same level into the same subnet set according to the island risk level, calculate the difference in stability index and the difference in cooperation index between devices in the subnet set, and generate a similarity index according to the difference in stability index and the difference in cooperation index between devices;

[0036] ;

[0037] Among them, is the similarity index between the th subnet and the th subnet in the same subnet set, is the island risk coefficient corresponding to the island risk level of the subnet set, is the similarity weight of the difference in the stability index of the subnet, is the th subnet and the is the similarity weight of the difference in the cooperation index between the devices of the subnet, is the th subnet and the

[0038] Correspondingly, the present application also provides a power grid equipment disconnection classification system based on multi-dimensional data analysis, including,

[0039] A monitoring module, configured to obtain the topological structure information of the target power grid, monitor whether a disconnection occurs in the target power grid through the topological structure information, and when a disconnection occurs in the target power grid, obtain the operation records before and after the disconnection of the target power grid, and determine the type of disconnection according to the operation records before and after the disconnection of the target power grid;

[0040] An evaluation module, configured to obtain the basic information of the equipment of the target power grid, determine the disconnection point location according to the operation records before and after the disconnection of the target power grid and the type of disconnection, and construct the topological structure diagrams of all subnets after the disconnection of the target power grid relying on the topological structure information, equipment basic information and disconnection point location of the target power grid, and evaluate the islanding risk of each subnet by analyzing the topological structure diagrams of each subnet according to the type of disconnection;

[0041] A generation module, configured to obtain the equipment operation information of each device before and after disconnection in the subnet topological structure diagram, and analyze the equipment operation information of each device to evaluate the stability index of each device and the cooperation index between devices, determine the stability index of the subnet through the stability index of the device, and generate the similarity index of the subnet by combining the stability index of the subnet, the cooperation index between devices and the islanding risk;

[0042] A classification module, configured to classify all subnets in the target power grid according to the similarity index of the subnets, generate a resource allocation strategy for the classified subnets, and thereby optimize the stable operation of the equipment in the target power grid.

[0043] Compared with the prior art, the beneficial effects of the present invention are:

[0044] 1. Monitor whether a disconnection occurs in the target power grid through the topological structure information and determine the type of disconnection, laying a solid foundation for the subsequent islanding risk assessment of the subnet and ensuring the reliability of subsequent evaluation and analysis.

[0045] 2. Evaluate the islanding risk of each sub-network based on the topological structure diagram of the sub-network, the type of islanding, and the data of the power grid. Then, combine the stability index of the sub-network and the cooperation index between devices to generate a similarity index for the classification standard. Define the similarity index based on the islanding risk and device conditions of the sub-network, which improves the adaptability and reliability of sub-network classification. Generate a resource allocation strategy for the classified sub-networks, and formulate different resource allocation strategies for sub-networks in different situations, so as to ensure the reliability of the operation of sub-network devices after islanding. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 FIG. is a schematic flow chart of a power grid device islanding classification method based on multi-dimensional data analysis proposed by the present invention;

[0047] Figure 2 FIG. is a schematic structural diagram of a power grid device islanding classification system based on multi-dimensional data analysis proposed by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0048] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0049] Refer to Figure 1 , a power grid device islanding classification method based on multi-dimensional data analysis, includes the following steps:

[0050] Step S101, obtain the topological structure information of the target power grid, monitor whether islanding occurs in the target power grid through the topological structure information. After islanding occurs in the target power grid, obtain the operation records before and after the islanding of the target power grid, and determine the type of islanding according to the operation records before and after the islanding of the target power grid.

[0051] In this embodiment, the topological structure information of the target power grid includes power grid nodes, lines, equipment types, capacities, etc. There are two types of power grid disconnection, namely active disconnection and passive disconnection. The reasons for passive disconnection are as follows: accidents occur in power grid tie lines, tie transformers or busbars. When a key tie line, tie transformer or busbar in the power grid fails, in order to prevent the spread of the fault, the system will automatically disconnect the affected part of the power grid from other parts. Overload tripping: When some parts of the power grid trip due to overload, it may lead to power grid disconnection. Protection misoperation tripping: Due to the misoperation of the protection device, it may cause unnecessary tripping of power grid equipment, thus triggering power grid disconnection. The reasons for active disconnection are as follows: to relieve power grid oscillation. When the power grid oscillates, in order to stabilize the operation of the power grid, the operator may manually disconnect the power grid. The low-frequency and low-voltage disconnection device automatically disconnects the power grid. When the frequency or voltage of the power grid drops to a certain level, the low-frequency and low-voltage disconnection device will automatically act to disconnect the power grid to prevent system collapse.

[0052] In some embodiments of the present application, the topological structure information is used to monitor whether disconnection occurs in the target power grid, including

[0053] Construct the standard topological structure of the target power grid according to the topological structure information of the target power grid, and traverse the nodes, lines and equipment in the target power grid at fixed intervals to construct the real-time topological structure of the target power grid, and compare the differences between the standard topological structure and the real-time topological structure of the target power grid to determine the topological difference degree;

[0054] Obtain the real-time power grid key parameters of each node on the standard topological structure of the target power grid, and determine the disconnection matching degree through the real-time power grid key parameters of each node on the standard topological structure and the preset disconnection characteristic pattern;

[0055] Comprehensively consider the topological difference degree and the disconnection matching degree to monitor whether disconnection occurs in the target power grid.

[0056] In this embodiment, topological search applies the search technology of artificial intelligence to analyze and identify the power grid structure in the power grid field. It traverses and searches each node, line and equipment in the power grid to construct a real-time topological structure diagram of the power grid. Compare the currently constructed real-time topological structure diagram with historical data or a preset normal topological structure (standard topology). If significant differences are found between the two, quantify these differences, such as changes in the connection methods of certain nodes or lines.

[0057] The preset disconnection characteristic pattern is trained and constructed according to the power grid key parameters during power grid disconnection in history. In the case of disconnection, the power grid key parameters may show the following situations:

[0058] Frequency deviation: The frequencies in different parts of the power grid begin to show significant differences, which may be caused by load changes, improper adjustment of generator output, or changes in the power grid structure. Frequency deviation is an early warning signal for islanding.

[0059] Increasing phase difference: In the power grid, the voltages and currents at different nodes should be in phase or have a very small phase difference. If the phase difference begins to increase, it may mean that some parts of the power grid are losing synchronization, which is an important feature of islanding.

[0060] Voltage fluctuation: The stability of the power grid voltage is the basis for the normal operation of the power grid. If the voltage begins to fluctuate significantly, it may be caused by sudden load changes, generator failures, or power grid structure problems. Voltage fluctuation is also an indicator of the islanding risk.

[0061] Abnormal power flow distribution: The power flow (i.e., the direction and magnitude of power flow) in the power grid should follow a predetermined path and distribution. If the power flow distribution is abnormal, such as some lines being overloaded while others are lightly loaded, it may mean that there are problems with the power grid structure or operation mode, which is also an indication of the islanding risk.

[0062] It should be noted that the key parameters of the power grid are different from the operating parameters of the equipment under the subsequent power grid. The power grid is a complex system composed of multiple devices and lines, and the key parameters of the power grid usually refer to the parameters that can reflect the overall operating state and characteristics of the power grid, such as voltage, current, frequency, and phase angle. These parameters are the basis for the stable operation of the power grid and are also important bases for power grid state monitoring and fault judgment. The parameters of each device under the power grid, on the other hand, focus more on describing the operating state and characteristics of specific devices in the power grid, such as the voltage, current, and power factor of the device. Although these parameters are also important information in the operation of the power grid, they more reflect the operating state of a single device rather than the overall state of the power grid.

[0063] In some embodiments of the present application, the type of islanding is determined according to the operation records before and after the islanding of the target power grid, including

[0064] Determine the islanding time node, and convert the operation records within a period of time before and after the islanding time node into an operation timeline. The operation timeline is a horizontal axis listed in chronological order, and all operation events are marked on this horizontal axis;

[0065] The operation timeline includes two types of timelines: the protection operation timeline and the non - protection operation timeline, and the time scales of the protection operation timeline and the non - protection operation timeline are aligned. Compare and comprehensively analyze the protection operation timeline and the non - protection operation timeline to determine the type of islanding.

[0066] In this embodiment, the protection operation timeline is the action record of protection devices (such as circuit breakers, relays, etc.) in the power grid, and the non - protection operation timeline is the operation record of the power grid including but not limited to switch operations, equipment commissioning and decommissioning, load adjustment, etc. By combining the operation records on the protection operation timeline and the non - protection operation timeline, the type of disconnection is judged, whether it is an active disconnection or a passive disconnection.

[0067] Step S102: Obtain the basic information of the devices in the target power grid. Determine the disconnection point location according to the operation records and the disconnection type before and after the disconnection of the target power grid. Rely on the topological structure information, basic device information, and disconnection point location of the target power grid to construct the topological structure diagrams of all sub - grids after the disconnection of the target power grid. Analyze the topological structure diagrams of each sub - grid through the disconnection type to evaluate the islanding risk of each sub - grid.

[0068] In this embodiment, for active disconnection, the disconnection point is usually known. Active disconnection is a disconnection measure taken actively by power grid dispatching personnel according to the system operation status and prediction to prevent large - area power outages that may occur in the system. For passive disconnection, the disconnection point is usually unknown. Passive disconnection is caused by power grid faults or abnormalities, resulting in automatic or forced disconnection of the system.

[0069] In this embodiment, the islanding risk refers to the risk brought by the islanding phenomenon. The islanding (sub - grid) phenomenon refers to the situation where when the power supply of the power grid is interrupted due to faults or other external factors, the energy storage system, distributed power source, or a certain sub - grid forms an independent power supply area spontaneously with the local load. In this isolated "island" operation, the power source continues to supply power to the load, but is disconnected from the main power grid. The risk lies in the stability or unpredictability of the self - powered area that cannot be controlled by the main power grid. Generally speaking, the islanding risk corresponding to active disconnection is less than that corresponding to passive disconnection. The islanding risk may cause the following hazards:

[0070] 1. Deterioration of power quality: The voltage and frequency within the island may get out of control, leading to a serious decline in power quality, which in turn affects the normal operation of load equipment.

[0071] 2. Impairment of power grid stability: Island operation may also cause problems such as current surges and phase asynchronization when the power grid is re - connected, damaging power grid equipment, and thus affecting the overall stability and security of the power grid.

[0072] 3. Equipment damage and economic losses: The islanding effect may cause abnormal voltage and frequency in the system, damaging device equipment. At the same time, island operation will lead to energy waste and load imbalance, resulting in economic losses.

[0073] In some embodiments of this application, determining the disconnection point location according to the operation records and the disconnection type before and after the disconnection of the target power grid includes

[0074] The types of islanding include active islanding and passive islanding;

[0075] For active islanding, the islanding point location is determined according to the non-protection operation timeline of the target power grid;

[0076] For passive islanding, topological analysis is performed based on the difference between the standard topological structure and the real-time topological structure of the target power grid and the protection operation timeline, so as to screen out the set of islanding point locations, and the change of key power grid parameters of each islanding point in the set of islanding point locations before and after islanding is analyzed to determine the islanding point location.

[0077] In this embodiment, for active islanding, the specific islanding point is obtained according to the algorithms and strategies preset by the system and is known. For passive islanding, combined with the topological structure of the power grid and the configuration of the protection device, the range of the islanding point is narrowed, and the possible islanding points form a set of islanding point locations. The islanding point location is comprehensively judged by combining the records on the protection operation timeline and the change of key power grid parameters before and after islanding. The change of key power grid parameters (such as frequency deviation change, phase difference change, voltage fluctuation change, etc. can all reflect the islanding situation).

[0078] Step S103, obtain the device operation information of each device in the subnet topological structure diagram before and after islanding, and analyze the device operation information of each device to evaluate the stability index of each device and the cooperation index between devices. The stability index of the subnet is determined by the stability index of the device, and the similarity index of the subnet is generated by combining the stability index of the subnet, the cooperation index between devices and the islanding risk.

[0079] In this embodiment, the device operation information includes voltage, current, temperature, etc. The cooperation index between a single device and multiple devices under the subnet is determined according to the operation information of the device, and the two together constitute the similarity index.

[0080] In some embodiments of the present application, the islanding risk of each subnet is evaluated by analyzing the topological structure diagram of each subnet, including,

[0081] Identify key nodes and key lines in the topological structure diagram of each subnet, evaluate the scale index, load index, type of power source and number of power sources of the subnet through the key nodes and key lines, and evaluate the power redundancy of the subnet by combining the type of power source and the number of power sources;

[0082] Evaluate the islanding risk level of the subnet according to the scale index, load index, power redundancy of the subnet and the type of islanding, and describe the islanding risk situation of the subnet through the islanding risk level;

[0083] ;

[0084] Among them, is the risk level of islanding for the th subnet, is the risk conversion coefficient of the load index for the th subnet, is the load index for the th subnet, is the risk conversion coefficient of the power redundancy for the th subnet, is the power redundancy for the th subnet, is the scale index for the th subnet, is a constant determined by the type of islanding splitting, is a preset constant, and [] is the rounding symbol.

[0085] In this embodiment, the scale index, load index, type of power supply, and number of power supplies of the subnet are evaluated through key nodes and key lines. In the topological structure diagram of the subnet, key nodes (such as power supply points, load centers, etc.) and key lines (such as transmission corridors, important tie lines, etc.) are identified. The scale of the subnet directly affects its ability to operate in islanding mode. Smaller subnets may be more likely to maintain stable operation after islanding splitting, while larger subnets may face greater challenges. The load index is the load stability situation obtained by comprehensively considering the type, distribution, and variation law of the load, and analyzing the type of power supply in the subnet, such as traditional thermal power, hydropower, nuclear power, wind power, photovoltaic power, etc. Different types of power supplies have different stabilities and reliabilities during islanding operation. The power redundancy of the subnet is evaluated by combining the type and number of power supplies, that is, the ability of the subnet to maintain stable operation after losing some power supplies. The higher the power redundancy, the lower the risk of the subnet operating in islanding mode.

[0086] In this embodiment, is a constant determined by the type of islanding splitting. The constant corresponding to active islanding splitting is greater than the constant corresponding to passive islanding splitting, so that the value is smaller. represents the correction of the demand for the scale index of the subnet to the sum of the load and power redundancy.

[0087] In some embodiments of the present application, the operation information of each device is analyzed to evaluate the stability index of each device and the cooperation index between devices, including

[0088] Dividing the operation information of each device before and after islanding splitting in the subnet topological structure diagram into the operation information of the device during the complete period before and after islanding splitting and the operation information of the device during the period after islanding splitting according to time nodes;

[0089] Calculate the fluctuation amount of each operation index of the device based on the device operation information in the complete time period before and after the disconnection, calculate the real-time amount of each operation index of the device based on the device operation information in the time period after the disconnection, obtain the failure rate of each device, and determine the stability index of the device by synthesizing the operation index fluctuation amount, the operation index real-time amount and the failure rate;

[0090] ;

[0091] Among them, is the stability index of the th device, is the number of operation indexes in the th device, is the stability weight of the th operation index of the th device, is the real-time amount of the th operation index of the th device, is the fluctuation amount of the th operation index of the th device, is a preset constant, is the failure rate of the th device, is the failure stability influence coefficient of the th device, is the failure stability influence coefficient mapped from the failure rate of the th device;

[0092] Screen out the coordination information between devices from the device operation information before and after the device disconnection, and determine the cooperation index between devices according to the coordination information.

[0093] In this embodiment, the operation index fluctuation amount is a value describing the index fluctuation situation obtained according to the change rate in this period of time, and the operation index real-time amount is a relatively stable value determined according to the average value, the maximum value, the mode, etc. in this period of time, represents the correction of the operation index fluctuation amount to the operation index real-time amount. The stability index of the device is the product of the average value of the sum of the comprehensive influence amounts of the device operation indexes and the failure stability influence coefficient.

[0094] In some embodiments of the present application, determining the cooperation index between devices according to the coordination information includes,

[0095] Calculate the coefficient of variation of each cooperation index, and integrate the coefficient of variation according to multiple devices and multiple cooperation index categories, and describe the cooperation index between the devices in the subnet through the integrated coefficient of variation.

[0096] In this embodiment, the collaboration metrics include but are not limited to the following:

[0097] Communication latency: The latency time of information transmission between devices.

[0098] Synchronization accuracy: The degree of synchronization of the states or operations between devices.

[0099] Response speed: The response speed of one device to a request or instruction from another device.

[0100] Error rate: The error rate of information transmission or operations between devices.

[0101] Load balancing: The degree of balance of the workload between devices.

[0102] The coordination degree between devices is measured by calculating the coefficient of variation of each metric data. The smaller the coefficient of variation, the higher the coordination degree between devices. Multiple metrics and multiple devices are integrated to determine the collaboration metrics between the devices in the subnet.

[0103] In some embodiments of the present application, similarity metrics of the subnet are generated by combining the stability metrics of the subnet, the collaboration metrics between devices, and the island risk, including,

[0104] Subnets with the same level of island risk are planned into the same subnet set according to the island risk level. The differences in stability metrics and the differences in collaboration metrics between devices for each pair of subnets in the subnet set are calculated, and similarity metrics are generated based on the differences in stability metrics and the differences in collaboration metrics between devices;

[0105] ;

[0106] Among them, is the similarity metric between the th subnet and the th subnet in the same subnet set, is the island risk coefficient corresponding to the island risk level of this subnet set, is the similarity weight of the difference in stability metrics of the subnet, is the difference in stability metrics between the th subnet and the th subnet in the same subnet set, is the similarity weight of the difference in collaboration metrics between devices of the subnet, is the difference in collaboration metrics between devices between the th subnet and the th subnet in the same subnet set.

[0107] In this embodiment, subnets are classified according to similarity metrics in the same subnet set. is the island risk coefficient corresponding to the island risk level of the subnet set, and different island risk levels correspond to different island risk coefficients.

[0108] Step S104: Classify all subnets in the target power grid according to the similarity metrics of the subnets, and generate a resource configuration strategy for the classified subnets, so as to optimize the stable operation of the equipment in the target power grid.

[0109] In this embodiment, a resource configuration strategy is generated according to the specific characteristics of different types of subnets. The resource configuration strategy includes specific combinations in different directions such as equipment upgrade, backup power supply configuration, line transformation, and network security protection.

[0110] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0111] 1. Monitor whether a disconnection occurs in the target power grid through the topological structure information, and determine the disconnection type, laying a solid foundation for the subsequent island risk assessment of subnets and ensuring the reliability of subsequent evaluation and analysis.

[0112] 2. Evaluate the island risk of each subnet according to the topological structure diagram of the subnet, the type of disconnection, and the data of the power grid. Then, combine the stability index of the subnet and the cooperation index between devices to generate a similarity index for the classification standard. Define the similarity index through the island risk and equipment conditions of the subnet, improving the adaptability and reliability of subnet classification. Generate a resource configuration strategy for the classified subnets, and formulate different resource configuration strategies for different subnets to ensure the reliability of the operation of subnet equipment after disconnection.

[0113] Through the description of the above embodiments, those skilled in the art can clearly understand that the present invention can be implemented by hardware or by means of software plus a necessary general hardware platform. Based on such an understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (which can be a CD-ROM, a USB flash drive, a mobile hard disk, etc.), including several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various implementation scenarios of the present invention.

[0114] Correspondingly, the present application also provides a power grid equipment disconnection classification system based on multi-dimensional data analysis, as Figure 2 shown, including,

[0115] A monitoring module, which is used to obtain the topological structure information of the target power grid, monitor whether a disconnection occurs in the target power grid through the topological structure information, after a disconnection occurs in the target power grid, obtain the operation records before and after the disconnection of the target power grid, and determine the type of disconnection according to the operation records before and after the disconnection of the target power grid;

[0116] An evaluation module, which is used to obtain the basic device information of the target power grid, determine the disconnection point location according to the operation records and disconnection type before and after the disconnection of the target power grid, and construct the topological structure diagrams of all sub-grids after the disconnection of the target power grid relying on the topological structure information, basic device information and disconnection point location of the target power grid, and analyze the topological structure diagrams of each sub-grid through the disconnection type to evaluate the islanding risk of each sub-grid;

[0117] A generation module, which is used to obtain the device operation information before and after the disconnection of each device in the sub-grid topological structure diagram, and analyze the device operation information of each device to evaluate the stability index of each device and the coordination index between devices, determine the stability index of the sub-grid through the stability index of the device, and generate the similarity index of the sub-grid by combining the stability index of the sub-grid, the coordination index between devices and the islanding risk;

[0118] A classification module, which is used to classify all sub-grids in the target power grid according to the similarity index of the sub-grid, generate a resource allocation strategy for the classified sub-grids, so as to optimize the stable operation of the devices in the target power grid.

[0119] Those skilled in the art can understand that the drawings are only schematic diagrams of a preferred implementation scenario, and the modules or processes in the drawings are not necessarily essential for implementing the present invention.

[0120] Those skilled in the art can understand that the modules in the system in the implementation scenario can be distributed in the system of the implementation scenario according to the description of the implementation scenario, or can be changed accordingly and located in one or more systems different from this implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0121] The above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention, according to the technical solution and inventive concept of the present invention, makes equivalent substitutions or changes, and should be covered by the protection scope of the present invention.

Claims

1. A method for classifying power grid equipment based on multidimensional data analysis, characterized in that: include, Acquire the topological structure information of the target power grid, monitor whether a disconnection occurs in the target power grid through the topological structure information, and when a disconnection occurs in the target power grid, obtain the operation records before and after the disconnection of the target power grid, and determine the type of disconnection according to the operation records before and after the disconnection of the target power grid; Obtain basic equipment information of the target power grid, determine the location of the decoupling point based on the operation records and decoupling type before and after the decoupling of the target power grid, build the topological structure diagram of all subnets after the decoupling of the target power grid based on the topological structure information of the target power grid, basic equipment information and the location of the decoupling point, and analyze the topological structure diagram of each subnet by the decoupling type to evaluate the islanding risk of each subnet; Obtain the device operation information of each device before and after decoupling in the subnet topology diagram, and analyze the device operation information of each device to evaluate the stability index of each device and the coordination index between devices. Determine the stability index of the subnet based on the stability index of the device, and generate the similarity index of the subnet by combining the stability index of the subnet, the coordination index between devices and the island risk. Classify all subnets in the target power grid according to their similarity indicators, and generate resource allocation strategies for the classified subnets to optimize the stability of equipment operation in the target power grid; in, Assess the islanding risk of each subnet by analyzing the topology diagram of each subnet by type of decomposition, including: Identify key nodes and key lines in the topology diagram of each subnet, evaluate the scale index, load index, type and number of power sources of the subnet through key nodes and key lines, and evaluate the power redundancy of the subnet in combination with the type and number of power sources; The islanding risk level of the subnet is assessed based on the subnet's scale index, load index, power redundancy, and type of decoupling, and the islanding risk situation of the subnet is described by the islanding risk level; ； in, For the The island risk level of each subnet, For the The risk conversion coefficient of the load index of each subnet, For the The load index of each subnet, For the The risk conversion factor of the power redundancy of each subnet, For the Power redundancy of each subnet, For the The size index of each subnet, is a constant determined by the type of solution, is a preset constant, [] is a rounding symbol; Combine the stability index of the subnet, the coordination index between devices and the island risk to generate similar indicators of the subnet, including: According to the island risk level, subnets of the same level are planned into the same subnet set. The stability index difference between each subnet in the subnet set and the coordination index difference between devices are calculated. Similarity index is generated based on the stability index difference and the coordination index difference between devices. ； in, To the same subnet set subnet and Similarity index between subnets, is the island risk coefficient corresponding to the island risk level of the subnet set, is the similarity weight of the difference in stability index of the subnet, For the same subnet set subnet and The difference in stability index between subnets is is the similarity weight of the difference in the coordination index between the devices in the subnet, For the same subnet set subnet and The difference in coordination indicators between devices in different subnets.

2. The method for classifying power grid equipment based on multidimensional data analysis according to claim 1, characterized in that: Use topology information to monitor whether the target grid is disconnected, including: The standard topology of the target power grid is constructed according to the topology information of the target power grid, and the nodes, lines and devices in the target power grid are traversed at a fixed period to construct the real-time topology of the target power grid, and the difference between the standard topology and the real-time topology of the target power grid is compared to determine the topology difference degree; Acquire the real-time key grid parameters of each node on the standard topology of the target grid, and determine the splitting matching degree through the real-time key grid parameters of each node on the standard topology and the preset splitting characteristic mode; The topology difference and split matching degree are combined to monitor whether splitting occurs in the target power grid.

3. The method for classifying power grid equipment based on multidimensional data analysis according to claim 1, characterized in that: And determine the type of disconnection based on the operation records before and after the target power grid is disconnected, including: Determine the decoupling time node, and convert the operation records in a period of time before and after the decoupling time node into an operation timeline. The operation timeline is a horizontal axis listed in chronological order, and all operation events are marked on the horizontal axis; The operation timeline includes two timelines: a protection operation timeline and a non-protection operation timeline. The time scales of the protection operation timeline and the non-protection operation timeline are aligned. The protection operation timeline and the non-protection operation timeline are compared and integrated to determine the type of decoupling.

4. The method for classifying power grid equipment based on multidimensional data analysis according to claim 2 or 3, characterized in that: Determine the location of the disconnection point based on the operation records and disconnection type before and after the target grid is disconnected, including: The types of decoupling include active decoupling and passive decoupling; For active disconnection, the disconnection point is determined according to the non-protection operation timeline of the target power grid; For passive disconnection, topology analysis is performed based on the difference between the standard topology and the real-time topology of the target power grid and the protection operation timeline to screen out a set of disconnection point locations. The changes in key parameters of the power grid before and after disconnection at each disconnection point in the set of disconnection point locations are analyzed to determine the disconnection point location.

5. The method for classifying power grid equipment based on multidimensional data analysis according to claim 1, characterized in that: And analyze the equipment operation information of each device to evaluate the stability index of each device and the coordination index between devices, including, The device operation information of each device before and after the decoupling in the subnet topology structure diagram is divided into the device operation information of the complete period before and after the decoupling and the device operation information of the period after the decoupling according to the time node; Calculate the fluctuation of each operating index of the equipment based on the equipment operation information of the complete period before and after the decoupling, calculate the real-time quantity of each operating index of the equipment based on the equipment operation information of the period after the decoupling, obtain the failure rate of each equipment, and determine the stability index of the equipment by combining the fluctuation of the equipment operation index, the real-time quantity of the operation index and the failure rate; ； in, For the The stability index of each device, For the The number of running indicators in the device, For the Device The stable weight of the operating indicators, For the Device The real-time quantity of the operating indicators, For the Device The fluctuation of the operating indicators, is the preset constant, For the The failure rate of each device, For the The failure stability influence coefficient of each device, For the The failure stability impact coefficient obtained by mapping the failure rate of each device; The coordination information between the devices is screened out from the device operation information before and after the devices are decoupled, and the coordination indicators between the devices are determined based on the coordination information.

6. The method for classifying power grid equipment based on multidimensional data analysis according to claim 5, characterized in that: Determine the coordination indicators between devices based on the coordination information. include, The coefficient of variation of each collaborative indicator is calculated, and the coefficient of variation is integrated according to multiple devices and multiple collaborative indicator categories. The integrated coefficient of variation is used to describe the collaborative indicators between the devices in the subnet.

7. A network classification system for power grid equipment based on multidimensional data analysis, applied to the network classification method for power grid equipment based on multidimensional data analysis as claimed in any one of claims 1 to 6, characterized in that: The system further comprises, A monitoring module is used to obtain topological information of the target power grid, monitor whether a disconnection occurs in the target power grid through the topological information, obtain operation records before and after the disconnection of the target power grid when the disconnection occurs in the target power grid, and determine the type of disconnection according to the operation records before and after the disconnection of the target power grid; The evaluation module is used to obtain basic equipment information of the target power grid, determine the location of the decoupling point according to the operation records and decoupling type of the target power grid before and after decoupling, build the topological structure diagram of all subnets after decoupling of the target power grid based on the topological structure information of the target power grid, basic equipment information and the location of the decoupling point, and analyze the topological structure diagram of each subnet by the decoupling type to evaluate the islanding risk of each subnet; A generation module is used to obtain the device operation information of each device before and after the decoupling in the subnet topology diagram, and analyze the device operation information of each device to evaluate the stability index of each device and the coordination index between devices, determine the stability index of the subnet through the stability index of the device, and generate the similarity index of the subnet by combining the stability index of the subnet, the coordination index between devices and the island risk; The classification module is used to classify all subnets in the target power grid according to the similarity indicators of the subnets, and generate resource allocation strategies for the classified subnets to optimize the stability of equipment operation in the target power grid.

Images