Power system global information sensing method and device using 5g technology

Abstract

The present application relates to the technical field of electric power digital information transmission, and particularly relates to a power system global information sensing method and device using 5G technology. The technical method of the present application obtains multi-dimensional monitoring data of each power transmission node in different regions of a power system based on a 5G communication mode, analyzes at least one monitoring data associated with the regional environment in the multi-dimensional monitoring data of each power transmission node in each region to identify whether there is an abnormal node in each power transmission node in each region under the exclusion of environmental interference factors; obtains an abnormal degree coefficient of the abnormal node according to an actual change trend represented by the multi-dimensional monitoring data of the abnormal node and the correlation between different dimensional monitoring data; determines the processing priority of the abnormal node according to the abnormal degree coefficient, and sequentially processes each abnormal node based on the processing priority. The technical solution can timely respond to the sensing and disposal of abnormal power transmission nodes in local regions of the power system.

Description

Technical Field

[0001] This invention relates to the technical field of power digital information transmission, and specifically to a method and device for full-domain information sensing of a power system using 5G technology. Background Technology

[0002] With the continuous construction of power grids, transmission networks have gradually developed into giant power grids with large capacity, long-distance transmission, and multi-regional interconnection, making the safe and stable operation of the power grid particularly important. Simultaneously, with the rapid development of internet technology, power monitoring data from sensors in local areas of the power grid is transmitted via network to the power system data management area. This enables comprehensive information perception across the entire power system and allows for timely adjustments to address internal power system changes, such as abnormal power transmission within the region, ensuring regional power safety.

[0003] In practice, when collecting monitoring data for local areas of the power system, anomalies can occur in the monitoring data of transmission nodes due to equipment aging or sudden malfunctions. This can lead to regional monitoring data anomalies. When the monitoring system platform processes data from all regional nodes, the large volume of data makes it difficult to identify and process critical node data in a timely and effective manner when processing data from different nodes in a fixed order. This results in the accumulation of anomalies in local areas of the power grid, leading to equipment failures. Summary of the Invention

[0004] To address the technical challenges of timely detection and handling of abnormal transmission nodes in localized power system areas, this invention aims to provide a method and device for comprehensive power system information sensing utilizing 5G technology. The specific technical solution adopted is as follows:

[0005] In a first aspect, embodiments of the present invention provide a method for comprehensive information perception of a power system using 5G technology, the method comprising:

[0006] Multi-dimensional monitoring data of each transmission node in different regions of the power system is obtained based on 5G communication. The monitoring data of each dimension includes a time data sequence of the corresponding region's transmission node based on preset monitoring items over the monitoring time.

[0007] Analyze at least one monitoring data related to the regional environment from the multidimensional monitoring data of each transmission node in each region, so as to identify whether there are abnormal nodes in each transmission node in each region after excluding environmental interference factors;

[0008] When an abnormal node is detected at a transmission node in the corresponding region, the abnormality coefficient of the abnormal node is obtained based on the actual change trend represented by the multidimensional monitoring data of the abnormal node and the correlation between monitoring data of different dimensions.

[0009] The processing priority of abnormal nodes is determined based on the abnormality degree coefficient, and each abnormal node is processed sequentially based on the processing priority.

[0010] In one optional embodiment, at least one monitoring data point related to the regional environment is parsed from the multidimensional monitoring data of each transmission node in each region to identify whether there are abnormal nodes in each transmission node of each region after excluding environmental interference factors, including:

[0011] Extract the first temperature data representing the equipment temperature and the second temperature data representing the ambient temperature from the multidimensional monitoring data of each transmission node in each region;

[0012] Based on the average change of the first and second temperature data over the monitoring period, the proportional coefficient of the correlation between equipment temperature and ambient temperature is obtained.

[0013] The multidimensional monitoring data is corrected based on the proportional coefficient, and the anomaly index of each transmission node in the corresponding area is determined based on the corrected monitoring data.

[0014] When the abnormal index is greater than or equal to the preset index threshold, the transmission node in the corresponding area is identified as an abnormal node.

[0015] In one optional embodiment, a proportionality coefficient is obtained based on the average change of the first temperature data and the second temperature data over the monitoring period, including:

[0016] Based on the ratio of the first temperature data to the second temperature data at the same monitoring time in each monitoring cycle, the influence of the equipment temperature on the ambient temperature at the corresponding monitoring time is obtained.

[0017] Based on the average impact magnitude at each monitoring time point across all monitoring cycles, the proportional coefficient of the correlation between equipment temperature and ambient temperature is obtained.

[0018] In one optional embodiment, the multidimensional monitoring data is corrected according to a scaling factor, and the anomaly index of each transmission node in the corresponding area is determined based on the corrected monitoring data, including:

[0019] The monitoring data for each dimension is corrected based on the proportional coefficient to obtain the actual monitoring data for the corresponding monitoring item;

[0020] Based on the actual monitoring data and historical monitoring average of each monitoring item, the percentage of mutation degree representing the deviation of the corresponding monitoring item data and the mutation state parameter representing the overall balance of mutation in the monitoring data are obtained.

[0021] The anomaly index of the corresponding transmission node is obtained based on the percentage of the maximum mutation degree and the mutation state parameters of all monitoring items at each transmission node.

[0022] In one optional embodiment, the anomalousness coefficient of the anomalous node is obtained based on the actual change trend represented by the multidimensional monitoring data of the anomalous node and the correlation between monitoring data of different dimensions, including:

[0023] Based on the proportional coefficient characterizing the correlation between equipment temperature and ambient temperature, the multidimensional monitoring data of the abnormal node is corrected to obtain the actual monitoring data of each monitoring item of the abnormal node.

[0024] Based on the fluctuation characteristics of the actual monitoring data of each monitoring item of the abnormal node during the abnormal period, the month-on-month change of the monitoring item corresponding to the abnormal node is obtained.

[0025] Similarity calculations are performed on the actual monitoring data of each monitoring item of the abnormal node to obtain the correlation coefficient of the data changes of the abnormal node during the abnormal period.

[0026] The degree of abnormality of the abnormal node is obtained by comparing the month-on-month change and correlation coefficient of the monitoring items corresponding to the abnormal node.

[0027] In one optional embodiment, the month-on-month change of the monitoring item corresponding to the abnormal node is obtained based on the fluctuation characteristics of the actual monitoring data of each monitoring item of the abnormal node during the abnormal period, including:

[0028] The actual monitoring data of each monitoring item of the abnormal node are input into the preset first calculation model to obtain the monitoring difference and average difference of each monitoring item at adjacent times during the abnormal period.

[0029] Based on the difference between the monitoring difference and the average difference, the month-on-month change of the monitoring item corresponding to the abnormal node is obtained.

[0030] In one optional embodiment, similarity calculation is performed based on the actual monitoring data of each monitoring item of the abnormal node to obtain the correlation coefficient of the data changes of the abnormal node during the abnormal period, including:

[0031] The actual monitoring data of each monitoring item of the abnormal node are input into the preset second calculation model to obtain the mean similarity of the changing trends between each monitoring item of the abnormal node, and the extreme value difference ratio of the similarity between each monitoring data item.

[0032] The correlation coefficient of the abnormal nodes is obtained by multiplying the mean similarity and the ratio of extreme value differences.

[0033] In one optional embodiment, the anomalousness coefficient of the anomalous node is obtained based on the month-on-month change and correlation coefficient of the monitoring item corresponding to the anomalous node, including:

[0034] The cumulative change of all monitoring items for the abnormal node is summed to obtain the cumulative change of the abnormal node.

[0035] The abnormality coefficient of the abnormal node is obtained by inversely proportional to the cumulative change and the correlation coefficient.

[0036] In one optional embodiment, determining the processing priority of abnormal nodes based on an anomaly degree coefficient includes:

[0037] The first weight of the current abnormal node is obtained based on the proportion of the abnormality degree coefficient of the current abnormal node to the total degree coefficients of all abnormal nodes.

[0038] The processing priority of the abnormal node is obtained based on the first weight of the current abnormal node, the second weight of the power equipment to which it belongs, and the third weight of the corresponding region in the power system.

[0039] Secondly, embodiments of the present invention also provide a power system global information sensing device utilizing 5G technology, for implementing any of the sensing methods in the first aspect, the device comprising:

[0040] The data acquisition terminal is used to acquire multi-dimensional monitoring data of each transmission node in different regions of the power system based on 5G communication. The monitoring data of each dimension includes a time data sequence of the corresponding region's transmission node based on preset monitoring items over the monitoring time.

[0041] The processing terminal is connected to the acquisition terminal. The processing terminal is used to analyze at least one monitoring data related to the regional environment in the multi-dimensional monitoring data of each transmission node in each region, so as to identify whether there are abnormal nodes in each transmission node in each region after excluding environmental interference factors.

[0042] When an abnormal node is detected at a transmission node in the corresponding region, the abnormality coefficient of the abnormal node is obtained based on the actual change trend represented by the multidimensional monitoring data of the abnormal node and the correlation between monitoring data of different dimensions.

[0043] The processing priority of abnormal nodes is determined based on the abnormality degree coefficient, and each abnormal node is processed sequentially based on the processing priority.

[0044] The present invention has the following beneficial effects:

[0045] The technical method of this invention acquires multi-dimensional monitoring data of transmission nodes in different regions of a power system based on 5G communication. Since the monitoring data of each dimension includes a time data sequence constructed by the corresponding regional transmission node based on preset monitoring items over monitoring time, it can accurately analyze at least one monitoring data related to the regional environment in the multi-dimensional monitoring data of each transmission node in each region, so as to identify whether there are abnormal nodes in each region's transmission nodes after excluding environmental interference factors. When an abnormal node is detected in the corresponding region's transmission nodes, the power system can further analyze the actual change trend represented by the multi-dimensional monitoring data of the abnormal node and the correlation between monitoring data of different dimensions to obtain the abnormality degree coefficient of the abnormal node. The processing priority of the abnormal node is determined according to the abnormality degree coefficient, and each abnormal node is processed sequentially based on the processing priority. This technical solution avoids the problem of traditional isolated forest algorithms relying solely on the dispersion of data when performing anomaly analysis on regional nodes within a power system. This avoids the lag in handling low-progression hazards such as aging, which can lead to delays in the processing of serious hazards. Instead, it addresses the impact of untimely data processing due to the lack of comprehensive information perception across the power system. This solution prioritizes responses to sudden power system faults and ensures timely handling of high-risk nodes, achieving the goal of timely detection and handling of abnormal transmission nodes in local power system areas. Attached Figure Description

[0046] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0047] Figure 1 A flowchart illustrating a method for full-domain information perception of a power system using 5G technology, as provided in an embodiment of the present invention;

[0048] Figure 2 This is a flowchart illustrating the identification process of abnormal nodes according to an embodiment of the present invention.

[0049] Figure 3 A flowchart illustrating the calculation of the anomaly degree coefficient provided in one embodiment of the present invention;

[0050] Figure 4 This is a schematic diagram of the structure of a power system global information sensing device utilizing 5G technology, provided as an embodiment of the present invention. Detailed Implementation

[0051] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a power system full-domain information sensing method and device utilizing 5G technology proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0053] For the monitoring and handling of power system-wide information perception, existing methods mostly employ the isolated forest algorithm to detect the entire power system, which cannot handle monitoring anomalies in a timely manner. For example, localized aging of equipment (such as loose connections or insulation degradation) can cause continuous abnormal fluctuations in monitoring data (such as temperature and current). However, the isolated forest algorithm cannot distinguish the abnormal service attributes and will treat it as a stability fault such as a short circuit, thus assigning an excessively high anomaly weight to that area. This leads to misjudging the severity of such faults, resulting in a mismatch of data processing priorities. Consequently, the overall power system fault identification resources are heavily consumed by "aging anomalies," delaying the identification and handling of real faults. The following will describe in detail, with reference to the accompanying drawings, a specific solution for a power system-wide information perception method and equipment utilizing 5G technology provided by this invention.

[0054] Please see Figure 1 , Figure 1 This is a flowchart illustrating a method for comprehensive information sensing of a power system using 5G technology, provided as an embodiment of the present invention. This method can operate based on sensing devices within the power system. These sensing devices can be computer equipment or server equipment, as long as they are capable of running the sensing method; no specific limitations are imposed here. The sensing method includes:

[0055] S11. Based on 5G communication, acquire multi-dimensional monitoring data of each transmission node in different regions of the power system. The monitoring data of each dimension includes the time data sequence of the corresponding regional transmission node based on preset monitoring items over the monitoring time.

[0056] Specifically, a power system is a complex power transmission network composed of generation, transmission, transformation, and distribution. The operation of a power system involves numerous parameters requiring real-time monitoring, such as current, voltage, temperature, and load quantity. Each dimension of the multi-dimensional monitoring data corresponds to a pre-defined monitoring item, such as current monitoring or temperature monitoring. 5G communication, with its high bandwidth and low latency, can support real-time data acquisition and transmission from tens of thousands of transmission nodes in the power system. Each node has corresponding operating equipment, and 5G communication allows for the synchronous updating of the operating status, environmental parameters, and anomaly information of each piece of equipment. 5G communication enables comprehensive information perception of the power system, significantly improving the monitoring and response speed of the power grid and providing fundamental monitoring data for power equipment condition assessment, fault prediction, and emergency dispatch.

[0057] It is understandable that the communication advantages of 5G technology can enable the monitoring of equipment operating at different nodes within a region, and provide data transmission support for the monitored equipment based on different communication base stations within the region. For a single region, the power system receives electrical energy from power plants and transmits it through high-voltage transmission to substations in different areas of the region, thereby distributing power resources to different users. The power supply data from different substations within a region is considered as a regional node, and monitoring equipment or sensors are used to collect and monitor the operational data of the equipment within that node.

[0058] Furthermore, noise reduction processing can be performed on the monitoring data within the region based on the median filtering function. The monitoring data from different nodes within the region is processed by the operating equipment. Each preset data item for monitoring is a dimension of data, and each dimension of data is used to construct a time data sequence based on the data monitored for the corresponding monitoring item over time.

[0059] At this point, multi-dimensional monitoring data of each transmission node in different regions of the power system have been obtained, and we proceed to step S12.

[0060] S12. Analyze at least one monitoring data related to the regional environment from the multi-dimensional monitoring data of each transmission node in each region, so as to identify whether there are abnormal nodes in each transmission node in each region after excluding environmental interference factors.

[0061] Specifically, the operation of various equipment in a power system is affected by the environment. For example, the operating state of transmission equipment needs to follow thermal equilibrium. During operation, the equipment continuously generates heat due to the Joule effect of current, which is then dissipated to the surrounding environment through conduction and convection. When the ambient temperature is high, it affects the heat dissipation effect of the power system equipment, leading to abnormal fluctuations in the power equipment's operating data due to environmental influences. Therefore, when analyzing the operating data of power system equipment in the environment, it is necessary to remove the interference of environmental factors on the monitoring results. Taking the temperature parameter during the operation of power equipment as an example, at least one monitoring data point related to the regional environment can be extracted from the multi-dimensional monitoring data to determine the degree of interference of this data point under environmental factors, and then analyze whether there are any anomalies at each transmission node.

[0062] This invention will take the change in the relationship between equipment temperature and ambient temperature as an example to analyze the degree to which power system equipment is affected by ambient temperature, and then obtain the non-interference performance coefficient of the monitoring data during operation, that is, the anomaly index represented by each node, and determine whether there is an anomaly at the corresponding node based on the anomaly index. For an example, please refer to [link to example]. Figure 2 , Figure 2 The flowchart for identifying abnormal nodes is as follows. Step S12 includes sub-steps S12-1 to S12-4, which are described in detail below:

[0063] S12-1. Extract the first temperature data representing equipment temperature and the second temperature data representing ambient temperature from the multi-dimensional monitoring data of each transmission node in each region. Due to daily climate changes, the temperature data of the power coefficient operating equipment at each node of the monitoring cycle is recorded on a daily basis (0-24 hours), and a monitoring data-monitoring time series is constructed. The first temperature data represents the corresponding temperature value obtained over the monitoring time when the equipment is being monitored; the second temperature data represents the corresponding temperature value obtained over the monitoring time when the environment in which the equipment is located is being monitored.

[0064] S12-2. Based on the average changes of the first and second temperature data over the monitoring period, obtain the proportionality coefficient of the correlation between equipment temperature and ambient temperature. The proportionality coefficient characterizes the degree of correlation between equipment temperature and ambient temperature, and can be derived from the ratio of the first and second temperature data.

[0065] It is understandable that during the operation of power system equipment, the temperature change is positively correlated with the ambient temperature, in addition to being affected by the equipment's own power consumption. That is, the higher the ambient temperature, the worse the heat dissipation of the power equipment, and the higher the monitored temperature; conversely, the monitored temperature is relatively lower. Therefore, for the changes in monitoring data in a single cycle, by comparing the average changes in monitoring data over M adjacent monitoring cycles, i.e., the average comparison between power equipment and temperature changes, a proportionality coefficient can be obtained.

[0066] For example, one can first obtain the influence of the equipment temperature relative to the ambient temperature at the same monitoring time by comparing the ratio of the first temperature data to the second temperature data at the same monitoring time in each monitoring cycle. Then, based on the average of the influence magnitudes at each monitoring time across all monitoring cycles, one can obtain the proportional coefficient of the correlation between equipment temperature and ambient temperature.

[0067] The proportionality coefficient is denoted as Through the formula:

[0068]

[0069] Calculate the proportionality coefficient In the formula, i, j, M, and N are all natural numbers greater than 1, i is the number of monitoring moments in a single cycle; N is the total number of moments in a single cycle; j is the cycle number; and M is the number of cycles that have been monitored. The monitored temperature of the equipment operating at time i in the j-th cycle is the first temperature data. The ambient temperature of the equipment during the j-th cycle is the second temperature data. It should be noted that the ambient temperature may reach 0 degrees Celsius; therefore, to ensure the calculation is meaningful, when encountering... When the temperature is 0 degrees Celsius, it is analyzed as 1 degree Celsius, and the impact on the overall calculation results is within the allowable error range. This represents the characteristic relationship between equipment temperature and ambient temperature, used to indicate the influence of ambient temperature on equipment operating temperature at a certain moment, and characterizes the magnitude of the influence of equipment temperature on ambient temperature at the corresponding monitoring moment. This represents the average value comparing the equipment temperature and the ambient temperature. The function normalizes the calculation results.

[0070] S12-3. Correct the multidimensional monitoring data according to the proportionality coefficient, and determine the anomaly index of each transmission node in the corresponding area based on the corrected monitoring data. The proportionality coefficient is the proportion of the influence of ambient temperature on the monitoring data during equipment operation. The collected monitoring data can be corrected based on the proportionality coefficient to obtain the actual monitoring data during equipment operation. The actual monitoring data can be recorded as a non-environmental influence parameter. Taking the correction of the monitoring temperature of operating equipment as an example, it can be done using the formula... Obtain the actual monitored temperature of the operating equipment at time i. , This is the first temperature data collected under the influence of ambient temperature.

[0071] It should be noted that there is a correlation between different monitoring items and ambient temperature. Based on this correlation, corresponding proportionality coefficients can be derived, and then the actual monitoring data for each item can be calculated. Based on the above steps, the non-environmental factors affecting various monitoring data during equipment operation can be detected. This refers to the actual monitoring data for each monitoring item. Once the actual monitoring data is obtained, the data changes at each transmission node can be accurately analyzed, and the corresponding node's anomaly index can be derived.

[0072] During the operation of power system equipment, under normal conditions without external interference, monitoring data should remain stable or experience only minor fluctuations. However, when sudden conditions occur that cause significant anomalies in the monitoring data, such as excessively high ambient temperatures leading to poor heat dissipation and resulting in high resistance at equipment connections and abnormal current data monitoring, it becomes crucial to consider the fluctuations in monitoring data after environmental interference. High data stability indicates normal equipment operation; conversely, abnormal data suggests malfunction. Therefore, it is necessary to compare monitoring data under specific environmental parameters and analyze the anomaly index of transmission nodes.

[0073] For example, sub-step S12-3 includes:

[0074] The first step is to correct the monitoring data for each dimension based on the proportional coefficient to obtain the actual monitoring data for the corresponding monitoring item. A calculation model can be set up based on the relationship between the proportional coefficient and the monitoring data of each dimension to achieve the corresponding correction of the monitoring data of each dimension, and the actual monitoring data of the corresponding monitoring item can be obtained based on the correction results.

[0075] The second step involves obtaining the percentage of abrupt changes, representing the degree of deviation from the data for each monitoring item, and a mutation state parameter, representing the overall balance of abrupt changes in the monitoring data, based on the actual monitoring data and the historical average for each item. The percentage of abrupt changes reflects the degree of deviation between the current value and the historical average of the corresponding monitoring item. If, among temperature, current, and voltage, temperature fluctuates the most, then the percentage of abrupt changes for the temperature monitoring item is the highest, indicating that this monitoring item is most likely the source of the anomaly. Similarly, the mutation state parameter measures the balance of the current overall monitoring state. For example, when the fluctuations of multiple monitoring items are similar, it indicates that the power system is generally stable; conversely, if individual monitoring items deviate too much, the mutation state parameter increases, indicating a risk of system imbalance.

[0076] The third step is to obtain the anomaly index of the corresponding transmission node based on the percentage of the maximum mutation degree and the mutation state parameters of all monitored items at each transmission node. The calculation result can be obtained by multiplying the percentage of the maximum mutation degree and the mutation state parameters, and then normalizing the calculation result to bring it within a preset range to obtain the anomaly index of the transmission node.

[0077] Specifically, it can be based on the formula:

[0078]

[0079] Calculate the anomaly index In the formula, k represents the monitoring item for the corresponding operating equipment in the power system; K represents the total number of all monitoring items. This represents the actual monitoring data for the k-th monitoring item. The historical monitoring average of the k-th monitoring item; This represents the absolute value of the difference between the actual monitoring data of the k-th monitoring item and the historical monitoring mean. This represents the percentage of mutation degree for the k-th monitoring item. The maximum percentage of mutation degree is selected from all monitoring items for calculation, and max() is the maximum percentage of mutation degree among all mutation degree percentages. The percentage of mutations in the overall monitoring data is compared with the overall mean, serving as a mutation status parameter; anomaly index. This indicates that, based on the proportion of mutation monitoring items relative to the overall situation, a higher proportion of mutation monitoring items suggests poorer monitoring balance of power equipment under the current environment and a higher probability of abnormal states. It is understandable that this application analyzes the difference between actual and historical monitoring data. A value of 0 indicates that there is no difference between actual and historical monitoring, and the overall calculation is meaningless. Therefore, in this embodiment of the invention, it is limited to... Not zero.

[0080] S12-4. When the anomaly index is greater than or equal to a preset index threshold, the corresponding transmission node in the area is identified as an anomaly node. The index threshold can be set based on the experience of technical personnel or through calibration experiments, ensuring that it can accurately determine the presence of anomalies in the transmission node. For example, setting the index threshold to 0.6, when the anomaly index... If the monitoring data of the monitored power system equipment is abnormal, it indicates that the power equipment may be malfunctioning; conversely, if the abnormality index is normal, it indicates that the power system equipment may be malfunctioning. If the signal is normal, it indicates that the power equipment is in normal operating condition and no abnormal response or handling is required.

[0081] At this point, based on the above analysis, it has been determined whether each transmission node is an abnormal node, and we proceed to step S13.

[0082] S13. When an abnormal node is detected at a transmission node in the corresponding area, the abnormality degree coefficient of the abnormal node is obtained based on the actual change trend represented by the multidimensional monitoring data of the abnormal node and the correlation between monitoring data of different dimensions.

[0083] Specifically, the degree of abnormality at an abnormal node can be determined using an anomaly severity coefficient. A higher coefficient indicates a higher requirement for timely response to the abnormal node, thus determining the priority for handling the anomaly. The aforementioned steps correct the monitoring data based on ambient temperature, yielding the actual monitoring data for each item. Based on this actual data, the actual trend of change can be determined, identifying the data variations for each item and eliminating the influence of environmental changes on the monitoring results. Furthermore, by analyzing the correlation between monitoring data from different dimensions, it can be determined whether a serious abnormal incident exists at the abnormal node, thus deriving an accurate anomaly severity coefficient.

[0084] It should be noted that monitoring equipment anomalies mainly fall into two categories: one is sudden anomalies, such as equipment short circuits, which can cause drastic or abrupt changes in monitoring data such as current and voltage at a certain moment; the other is gradual anomalies caused by equipment aging, such as increased contact resistance due to oxidation or loosening of connectors, according to Joule's law (…). When the same current flows through it, the heat generated will be much higher than that of normal equipment, resulting in a persistently high or slowly increasing trend in temperature monitoring data. In power systems, equipment aging is a common phenomenon; the focus is on controlling the degree of aging, marking aging equipment, and replacing it promptly. Sudden anomalies, however, are more dangerous; if not addressed promptly in their early stages, they can lead to serious losses. Therefore, in one specific implementation, when processing abnormal data from power system equipment, the causes of the abnormal state should be analyzed, and the processing order of abnormal nodes in different areas should be dynamically adjusted according to the type and severity of the anomaly to prioritize the handling of severe anomalies. Please refer to [link to relevant documentation]. Figure 3 , Figure 3 The flowchart for calculating the anomaly degree coefficient is shown below. Step S13 includes sub-steps S13-1 to S13-4, which are described in detail below:

[0085] S13-1. Based on the proportionality coefficient characterizing the correlation between equipment temperature and ambient temperature, the multidimensional monitoring data of the abnormal node is corrected to obtain the actual monitoring data of each monitoring item of the abnormal node. The proportionality coefficient can determine the degree of influence of ambient temperature on the accuracy of monitoring data collection. Therefore, the monitoring data of each dimension can be corrected based on the determined proportionality coefficient, thereby obtaining the actual monitoring data of each monitoring item.

[0086] It is understandable that, based on the above steps, power equipment with abnormal nodes in the power system can be identified. However, at the monitoring time of the abnormal state, since the detected abnormal state is only in its initial stage, its development trend is not yet apparent or obvious. Therefore, it is necessary to analyze the staged anomalies of the equipment by combining monitoring data before and after the abnormal state monitoring time. This can be achieved by obtaining the time average T from the time the abnormal trend appears (i.e., the abnormal index reaches the index threshold) from historical monitoring data, as well as the preceding period before the abnormal state. Pre-launch period This information can be obtained based on DL / T 596-2021 "Preventive Testing Procedures for Power Equipment". Generally, the early warning analysis period is 3 days before the occurrence of an abnormal state. In this embodiment, the early warning period is configured as 1 day. The duration is 24 hours, which is used as the abnormal period for equipment monitoring. And monitor and analyze the degree of abnormality of power equipment during this period.

[0087] S13-2. Based on the fluctuation characteristics of the actual monitoring data of each monitoring item at the abnormal node during the abnormal period, obtain the month-on-month change of the corresponding monitoring item at the abnormal node. By analyzing the fluctuation characteristics of the actual monitoring data, the incremental degree of abnormality at the abnormal node can be determined and characterized by the month-on-month change.

[0088] Specifically, this includes: inputting the actual monitoring data of each monitoring item at the abnormal node into a preset first calculation model to obtain the monitoring difference and average difference of each monitoring item at adjacent times during the abnormal period; the first calculation model can be constructed based on the calculation formulas for the monitoring difference and average difference. Since the monitoring difference and average difference characterize the fluctuation characteristics of the data, the month-on-month change of the monitoring item corresponding to the abnormal node can be obtained based on the difference between the monitoring difference and the average difference.

[0089] Furthermore, most sudden failures are the result of the accumulation of gradual anomalies. That is, sudden anomalies are often accompanied by gradual anomalies caused by equipment aging, such as increased contact resistance due to prolonged oxidation of electrical connectors. Therefore, when analyzing power system data, the month-on-month changes in monitoring data over a period can be obtained by analyzing the performance of each monitoring data item. .

[0090]

[0091] In the formula, The time number within the abnormal period. The duration of the abnormal period. For monitoring items At any moment Actual monitoring data; This is the monitoring difference between actual monitoring data at adjacent times. The larger this value is, the higher the degree of change in the monitoring data under non-environmental influence at adjacent times, and the more serious the abnormal state of the power equipment. The average difference of monitoring data under non-environmental influences within the abnormal period; month-on-month change. For the current moment The difference between the mean and the first-order difference is the value of the difference. The larger the value, the greater the degree of abnormality. Here, norm represents the normalization function, which is used to normalize the data to eliminate the influence of dimensions.

[0092] In practical applications, analyzing fluctuations in abnormal states prior to a fault to determine whether a device exhibits a sudden and serious hazard remains difficult to accurately identify. Furthermore, different electrical equipment faults manifest differently when no abnormal state occurs. For example, aging increases the resistance of electrical equipment contacts, leading to corresponding changes in current and voltage. Sudden faults, such as short circuits, disrupt stable electrical connections, causing abnormal current balance in the three phases. This indicates that zero-sequence current has been generated in the three-phase power line, with voltage decreasing and current increasing. Therefore, based on the abnormal periods of the power system obtained through the above steps, it is also necessary to conduct correlation analysis by combining the various monitoring data items mentioned above.

[0093] S13-3. Similarity calculations are performed on the actual monitoring data of each monitoring item at the abnormal node to obtain the correlation coefficient of the data changes of the abnormal node during the abnormal period. The monitoring data used for similarity calculation can be all actual monitoring data during the abnormal period; alternatively, it can be actual monitoring data closer to the abnormal moment, such as the actual monitoring data from the two hours prior to when the abnormal index reaches the index threshold. The similarity calculation model can be used to determine whether there is similarity in the abnormal changes of each monitoring item, thereby deriving the correlation coefficient.

[0094] For example, the actual monitoring data of each monitoring item of the abnormal node is input into the preset second calculation model to obtain the mean similarity value of the changing trend between each monitoring item of the abnormal node, and the extreme value difference ratio of the similarity between each monitoring data item; the correlation coefficient of the abnormal node is obtained by multiplying the mean similarity value and the extreme value difference ratio.

[0095] Specifically, the changing trends of parameters can be analyzed based on the volatility of the data, combined with the performance relationship of the monitoring data. For example, when the resistance of power equipment increases, the voltage at the node needs to be increased to ensure current stability. Therefore, similarity analysis can be performed on the change curves of each monitoring data point during abnormal periods using the DTW (Dynamic Time Warping) algorithm. The negative value of the DTW is then normalized and used as the similarity metric. Obtain the correlation coefficient .

[0096]

[0097] In the formula, K represents the total number of all monitored items; For the similarity of k monitoring items; The average value of the similarity of the changing trends among the various monitoring items represents the overall similarity among the monitoring items. The larger the value, the more balanced the operation of the power equipment. The maximum value among all similarities; It is the minimum value of all similarities; The extreme difference represents the similarity between the monitoring data items. The larger this value is, the more balanced and worse the operation of the power equipment. The natural exponential function exp() is then used to perform inverse proportional normalization on the absolute value of the extreme difference, so that the larger the ratio of extreme differences is, the smaller the normalized value is, indicating that the operation of the power equipment is more balanced.

[0098] S13-4. Based on the month-on-month change and correlation coefficient of the monitoring items corresponding to the abnormal nodes, obtain the abnormality degree coefficient of the abnormal nodes. The abnormality degree coefficient is used to characterize the severity of the abnormal nodes during the abnormal period, and can be calculated based on the relationship between the month-on-month change, the correlation coefficient, and the abnormality degree coefficient.

[0099] For example, the cumulative change of all monitoring items of an abnormal node is summed to obtain the cumulative change of the abnormal node. The cumulative change can be used to determine the overall cumulative change of the abnormal node. Then, based on the inverse proportionality between the cumulative change and the correlation coefficient, the abnormality degree coefficient of the abnormal node can be obtained.

[0100] Furthermore, regarding weak equilibrium anomalies, the greater the month-on-month increase in monitoring data, the higher the fluctuations in the monitoring data (the more obvious the trend). Therefore, the greater the possibility of sudden anomalies in power equipment within the power system area, the more likely these anomalies are. This, combined with the month-on-month growth of monitoring data during the abnormal period, serves as the anomaly severity coefficient for the abnormal state. .

[0101]

[0102] In the formula, This is the inverse proportional value of the correlation coefficient during abnormal periods, indicating the imbalance between monitoring data; For cumulative changes; using The () function normalizes the calculation results. The larger the value, the more severe the abnormal state of the monitoring data during that period.

[0103] At this point, the abnormality coefficient of the abnormal node has been obtained, and we proceed to step S14.

[0104] S14. Determine the processing priority of abnormal nodes based on the abnormality degree coefficient, and process each abnormal node in sequence based on the processing priority.

[0105] Specifically, the anomaly degree coefficient can determine whether the corresponding abnormal node needs to be handled urgently, and then the handling priority of the abnormal node can be obtained. All handling priority levels can be divided into multiple levels based on actual needs, such as 10 levels. When the handling priority is higher, it means that the handling time is more urgent and the sudden failure needs to be responded to and dealt with first; conversely, when the handling priority is lower, the abnormal node is handled later.

[0106] For example, determining the processing priority of anomaly nodes based on anomaly severity coefficients includes: obtaining the first weight of the current anomaly node based on the proportion of its anomaly severity coefficient to the total severity coefficients of all anomaly nodes; and obtaining the processing priority of the anomaly node based on its first weight, the second weight of its associated power equipment, and the third weight of its corresponding region within the power system. Since the weights of different dimensions also reflect the priority level of anomaly handling, the processing priority calculated based on these three weights is more accurate.

[0107] It is understandable that the higher the anomaly coefficient of the monitoring data at different nodes within a power system, the greater the likelihood of sudden or emergency failures in the power equipment at that node. To prevent further losses, these sudden failures need to be prioritized. Therefore, based on the monitoring data at different nodes, the anomaly coefficient of the corresponding node with anomalies is obtained through the steps described above. Based on the data obtained from each abnormal node The degree coefficient at all nodes within the power system Calculate the first weight Based on the weights at different nodes in a region, priority analysis of regional node data processing is achieved.

[0108]

[0109] In the formula, h is the node number; The first weight of the abnormal node h; This is the ratio of the analysis result of the abnormality degree of the power equipment at node h to the cumulative value of the abnormality degree ratio of each node in the system, that is, the cumulative sum of all degree coefficients. The larger this value is, the more serious the abnormality at node h is, and the higher the priority of processing the abnormal node.

[0110] Based on the above steps, the priority of monitoring data processing at different nodes is obtained. By analyzing the severity of abnormal states of power equipment at different nodes, potential faults such as sudden and gradual faults are analyzed, and different first weights are set according to the severity.

[0111] Because different electrical devices perform different functions within a power system, the substantial economic losses resulting from their failures vary. This study ranks the actual or estimated losses of each electrical device at different nodes based on historical failure data, and uses the proportion of these losses to the overall losses as the second weight for prioritizing device handling. The third weight can be configured as the maximum weight coefficient among the regional nodes, and this is used to calculate the processing priority of equipment when an abnormal state occurs in a single regional node across the entire power system. .

[0112]

[0113] In the formula, Regional abnormal nodes The first weight; The second weight of the regional anomaly node h; The equipment weight coefficient of the monitoring data is affected by the proportion of power equipment failure losses. The larger the value, the higher the priority of equipment failure in the region that needs to be dealt with. This is the sum of the weights of all devices at the regional node when an abnormal state occurs; The maximum weight coefficient in the regional node represents the severity of the anomaly of the local equipment. Here, it serves as a priority coefficient affecting equipment processing to avoid a situation where a single equipment in the equipment node region has a high degree of anomaly while the other equipment has a low degree of anomaly, resulting in a low overall degree of anomaly in the regional node and causing unnecessary losses due to untimely repair of the equipment in that region.

[0114] Obtain the processing priority of abnormal nodes. Going forward, based on processing priority, the power system's global information sensing system will process abnormal data at nodes in each region sequentially to improve the efficiency of power system anomaly handling and timely response, thereby avoiding resource waste.

[0115] Based on the same technical concept as the sensing method, this embodiment of the invention also provides a power system global information sensing device utilizing 5G technology, for implementing any of the aforementioned sensing methods. Please refer to... Figure 4 , Figure 4 This is a schematic diagram of the sensing device, which includes a data acquisition terminal 1 and a processing terminal 2.

[0116] The acquisition terminal 1 can be configured as an acquisition board composed of current sensors, voltage sensors, temperature sensors, etc., and a filtering circuit can also be configured on the acquisition board. The acquisition terminal 1 is used to acquire multi-dimensional monitoring data of each transmission node in different areas of the power system based on 5G communication. The monitoring data of each dimension includes a time data sequence of the corresponding transmission node based on preset monitoring items over the monitoring time.

[0117] The processing terminal 2 can be configured as a computer. The processing terminal 2 is connected to the acquisition terminal 1. The processing terminal 2 is used to analyze at least one monitoring data related to the regional environment in the multi-dimensional monitoring data of each transmission node in each region, so as to identify whether there are abnormal nodes in each transmission node in each region after excluding environmental interference factors. When an abnormal node is detected in the transmission node of the corresponding region, the abnormality degree coefficient of the abnormal node is obtained according to the actual change trend represented by the multi-dimensional monitoring data of the abnormal node and the correlation between the monitoring data of different dimensions. The abnormality degree coefficient is used to determine the processing priority of the abnormal node, and each abnormal node is processed sequentially based on the processing priority.

[0118] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0119] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A method for comprehensive information sensing of a power system using 5G technology, characterized in that, The method includes: Multi-dimensional monitoring data of each transmission node in different regions of the power system is obtained based on 5G communication. The monitoring data of each dimension includes a time data sequence of the corresponding region's transmission node based on preset monitoring items over the monitoring time. The system analyzes at least one monitoring data point related to the regional environment from the multidimensional monitoring data of each transmission node in each region, in order to identify whether there are abnormal nodes in each transmission node in each region after excluding environmental interference factors. When an abnormal node is detected at a power transmission node in the corresponding area, the abnormality coefficient of the abnormal node is obtained based on the actual change trend represented by the multidimensional monitoring data of the abnormal node and the correlation between monitoring data of different dimensions. The processing priority of the abnormal node is determined according to the abnormality degree coefficient, and each abnormal node is processed sequentially based on the processing priority. Methods for obtaining the anomaly degree coefficient of abnormal nodes include: Based on the proportionality coefficient characterizing the correlation between equipment temperature and ambient temperature, the multidimensional monitoring data of the abnormal node is corrected to obtain the actual monitoring data of each monitoring item of the abnormal node. Based on the fluctuation characteristics of the actual monitoring data of each monitoring item of the abnormal node during the abnormal period, the month-on-month change of the monitoring item corresponding to the abnormal node is obtained. Similarity calculations are performed on the actual monitoring data of each monitoring item of the abnormal node to obtain the correlation coefficient of the data change of the abnormal node during the abnormal period. The abnormality coefficient of the abnormal node is obtained based on the month-on-month change of the monitoring item corresponding to the abnormal node and the correlation coefficient.

2. The power system full-domain information perception method using 5G technology according to claim 1, characterized in that, The process of analyzing at least one monitoring data point related to the regional environment from the multidimensional monitoring data of each transmission node in each region, to identify whether there are abnormal nodes in each transmission node of each region after excluding environmental interference factors, includes: Extract the first temperature data representing the equipment temperature and the second temperature data representing the ambient temperature from the multidimensional monitoring data of each transmission node in each region; Based on the average change of the first temperature data and the second temperature data during the monitoring period, the proportional coefficient of the correlation between the equipment temperature and the ambient temperature is obtained. The multidimensional monitoring data is corrected according to the proportional coefficient, and the anomaly index of each transmission node in the corresponding area is determined based on the corrected monitoring data. When the abnormal index is greater than or equal to a preset index threshold, the power transmission node in the corresponding area is determined to be an abnormal node.

3. The power system full-domain information perception method using 5G technology according to claim 2, characterized in that, The step of obtaining the proportional coefficient of the correlation between the equipment temperature and the ambient temperature based on the average change of the first temperature data and the second temperature data over the monitoring period includes: Based on the ratio of the first temperature data to the second temperature data at the same monitoring time in each monitoring cycle, the influence of the equipment temperature on the ambient temperature at the corresponding monitoring time is obtained. Based on the average value of the influence amplitude at each monitoring moment in all monitoring cycles, the proportional coefficient of the correlation between the equipment temperature and the ambient temperature is obtained.

4. The power system full-domain information perception method using 5G technology according to claim 2, characterized in that, The step of correcting the multidimensional monitoring data according to the proportional coefficient and determining the anomaly index of each transmission node in the corresponding area based on the corrected monitoring data includes: The monitoring data for each dimension is corrected according to the aforementioned proportional coefficient to obtain the actual monitoring data for the corresponding monitoring item; Based on the actual monitoring data and historical monitoring average of each monitoring item, the percentage of mutation degree representing the deviation of the corresponding monitoring item data and the mutation state parameter representing the overall balance of mutation in the monitoring data are obtained. The anomaly index of the corresponding transmission node is obtained based on the maximum percentage of mutations in all monitored items at each transmission node and the mutation state parameters.

5. The power system full-domain information perception method using 5G technology according to claim 1, characterized in that, The step of obtaining the month-on-month change of the monitoring item corresponding to the abnormal node based on the fluctuation characteristics of the actual monitoring data of each monitoring item of the abnormal node during the abnormal period includes: The actual monitoring data of each monitoring item of the abnormal node are input into the preset first calculation model to obtain the monitoring difference and average difference of each monitoring item at adjacent times during the abnormal period. Based on the difference between the monitoring difference and the average difference, the month-on-month change of the monitoring item corresponding to the abnormal node is obtained.

6. The power system full-domain information perception method using 5G technology according to claim 1, characterized in that, The step of performing similarity calculations based on the actual monitoring data of each monitoring item of the abnormal node to obtain the correlation coefficient of the data changes of the abnormal node during the abnormal period includes: The actual monitoring data of each monitoring item of the abnormal node are input into the preset second calculation model to obtain the average similarity value of the changing trend between each monitoring item of the abnormal node, and the extreme value difference ratio of the similarity between each monitoring data item. The correlation coefficient of the abnormal node is obtained by multiplying the mean similarity value and the ratio of extreme value differences.

7. The power system full-domain information perception method using 5G technology according to claim 1, characterized in that, Based on the month-on-month change of the monitoring item corresponding to the abnormal node and the correlation coefficient, the abnormality degree coefficient of the abnormal node is obtained, including: The cumulative change of all monitoring items of the abnormal node is summed to obtain the cumulative change of the abnormal node. The abnormality coefficient of the abnormal node is obtained by inversely proportional to the cumulative change and the correlation coefficient.

8. The power system full-domain information perception method using 5G technology according to claim 1, characterized in that, The step of determining the processing priority of the abnormal node based on the anomaly degree coefficient includes: The first weight of the current abnormal node is obtained based on the proportion of the abnormality degree coefficient of the current abnormal node to the total degree coefficients of all abnormal nodes. The processing priority of the abnormal node is obtained based on the first weight of the current abnormal node, the second weight of the power equipment to which it belongs, and the third weight of the corresponding region in the power system.

9. A power system full-domain information sensing device utilizing 5G technology, characterized in that, The device is used to implement the steps of the power system full-domain information perception method using 5G technology as described in any one of claims 1-8, and the device includes: The data acquisition terminal is used to acquire multi-dimensional monitoring data of each transmission node in different regions of the power system based on 5G communication. The monitoring data of each dimension includes a time data sequence of the corresponding region's transmission node based on preset monitoring items over the monitoring time. A processing terminal is connected to the acquisition terminal. The processing terminal is used to analyze at least one monitoring data related to the regional environment in the multi-dimensional monitoring data of each transmission node in each region, so as to identify whether there are abnormal nodes in each transmission node in each region after excluding environmental interference factors. When an abnormal node is detected at a power transmission node in the corresponding area, the abnormality coefficient of the abnormal node is obtained based on the actual change trend represented by the multidimensional monitoring data of the abnormal node and the correlation between monitoring data of different dimensions. And determine the processing priority of the abnormal node based on the abnormality degree coefficient, and process each abnormal node sequentially based on the processing priority; Methods for obtaining the anomaly degree coefficient of abnormal nodes include: Based on the proportionality coefficient characterizing the correlation between equipment temperature and ambient temperature, the multidimensional monitoring data of the abnormal node is corrected to obtain the actual monitoring data of each monitoring item of the abnormal node. Based on the fluctuation characteristics of the actual monitoring data of each monitoring item of the abnormal node during the abnormal period, the month-on-month change of the monitoring item corresponding to the abnormal node is obtained. Similarity calculations are performed on the actual monitoring data of each monitoring item of the abnormal node to obtain the correlation coefficient of the data change of the abnormal node during the abnormal period. The abnormality coefficient of the abnormal node is obtained based on the month-on-month change of the monitoring item corresponding to the abnormal node and the correlation coefficient.

Images