A power grid security monitoring system and method based on multimodal sensing technology

By building the security link and risk assessment of power grid equipment, dynamically dividing inspection areas, solving the problem of low efficiency of power grid security supervision systems in the existing technology, and achieving more accurate and efficient inspection and early warning of power grid equipment.

CN118627910BActive Publication Date: 2024-11-19NANJING QIZHENG INFORMATION TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411084696.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-08
Publication Date
2024-11-19
Estimated Expiration
2044-08-08

AI Technical Summary

Technical Problem

The existing power grid security supervision system based on multimodal perception technology is inefficient and time-consuming during the inspection process, and cannot fully and accurately obtain the inspection data of power grid equipment, resulting in inaccurate secondary inspection requirements and early warning information.

Method used

By building a security link for power grid equipment, combining historical data and multiple sensor monitoring results, the potential risk assessment value and chaotic intervention value of patrol are calculated, the machine inspection and human inspection areas are dynamically divided, early warning results are generated, and the inspection path is optimized. .

Benefits of technology

It improves the accuracy and efficiency of power grid equipment inspection, reduces the secondary inspection risks of staff, ensures the timeliness and accuracy of inspection results, and realizes effective supervision of power grid security data.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118627910B_ABST
    Figure CN118627910B_ABST
Patent Text Reader

Abstract

The present invention relates to the field of power grid security supervision technology, specifically a power grid security supervision system and method based on multimodal sensing technology, the system includes a risk assessment and analysis module, the risk assessment and analysis module obtains the complexity of the surrounding environment of each power grid device and the abnormal mutation characteristics of the corresponding power grid device monitoring results, and calculates the potential risk assessment value of each power grid device. In addition to sensing the operating status of the power grid equipment through a multimodal sensing sensor, the present invention also takes into account the complexity of the operating environment to which the power grid equipment belongs, and combines the abnormal data of the power grid equipment in the historical data to predict the abnormal risk probability of each power grid equipment, and then dynamically divides the machine inspection area and the manual inspection area in the area to which the power grid to be tested belongs, reducing the risk of secondary inspection of the power grid equipment by the staff, while ensuring the timeliness and accuracy of the inspection results, and realizing effective supervision of the power grid security data.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of power grid security supervision, and in particular to a power grid security supervision system and method based on multimodal sensing technology. Background Art

[0002] With the rapid development of smart grids, the operation and service model of power grids have undergone profound changes. By integrating advanced information and communication technologies, smart grids have realized intelligent management of the entire process of power generation, transmission, transformation, distribution and consumption, thereby improving the reliability and safety of the power grid. As the core of smart grids, the power monitoring system is responsible for real-time monitoring and control of equipment in the power grid to ensure the normal operation of the power grid and provide strong guarantees for the safe operation of the power grid.

[0003] Multimodal sensing technology refers to the use of multiple sensors and data sources to achieve comprehensive and accurate perception and understanding of the target object by integrating information from multiple modes. In the field of power grid security supervision, multimodal sensing technology can significantly improve the monitoring capabilities of power grid equipment, environment and potential threats.

[0004] Specifically, multimodal sensing technology can achieve multi-dimensional and all-round monitoring of power grid equipment by combining data from multiple sensors such as video, audio, infrared, and radar. These different modal data can complement each other to improve the accuracy and reliability of monitoring. At the same time, multimodal sensing technology can also use artificial intelligence algorithms to perform intelligent analysis of monitoring data to promptly discover and warn of potential safety hazards.

[0005] In the existing power grid security supervision system based on multimodal sensing technology, the operating status of the power grid equipment is often sensed through multimodal sensing devices (sensors), and the operating status of the power grid equipment is judged by analyzing the sensing data corresponding to the power grid equipment, and then the inspection path of the inspection equipment is planned; however, in actual work, due to the complexity of the operating environment of the power grid equipment, it is impossible to obtain the corresponding inspection data comprehensively and accurately by relying solely on the inspection equipment to perform inspection operations on the corresponding power grid equipment. It is often necessary for staff to conduct secondary inspections on some power grid equipment, and it is impossible to accurately provide early warning information for the area of ​​the secondary inspection. Therefore, the existing power grid security supervision system based on multimodal sensing technology has low supervision efficiency and is time-consuming and labor-intensive, and thus the existing technology has major defects. Summary of the invention

[0006] The purpose of the present invention is to provide a power grid security supervision system and method based on multimodal sensing technology to solve the problems raised in the above background technology.

[0007] In order to solve the above technical problems, the present invention provides the following technical solutions: a power grid security supervision method based on multimodal sensing technology, the method comprising the following steps:

[0008] S1. Obtain the connection relationship between the power grid equipment in the area to which the power grid to be tested belongs and the various sensing sensors bound to each power grid equipment, obtain the monitoring results of each sensing sensor in real time, and build a security association chain for the power grid equipment by combining the monitoring results between the sensing sensors in the historical data;

[0009] S2. Obtain the complexity of the surrounding environment of each power grid device and the abnormal mutation characteristics of the corresponding power grid device monitoring results, and calculate the potential risk assessment value of each power grid device;

[0010] S3. Based on the security association chain of power grid equipment and the potential risk assessment value of each power grid equipment, calculate the security inspection chaos intervention value of each power grid equipment; combine the abnormal data of power grid equipment in historical data to predict the abnormal risk probability of each power grid equipment;

[0011] S4. Based on the abnormal risk probability prediction value of each power grid equipment, dynamically divide the machine inspection area in the area where the power grid to be tested belongs, and generate security warning results for feedback.

[0012] Furthermore, the monitoring results obtained by different sensing sensors are different. The jth sensing sensor bound to the i-th power grid device in the area to be tested is recorded as Ai j ;

[0013] When building the security association chain of power grid equipment, obtain the abnormal monitoring data corresponding to the sensor in the historical data and extract Ai j Each time the corresponding monitoring data is abnormal, the remaining sensor sensors with abnormal monitoring results are combined to obtain Ai j The corresponding set of abnormal sensors;

[0014] Get Ai j The corresponding union of each abnormal sensor set is obtained, and the abnormal frequency ratio of each element in the obtained union is calculated respectively. The power grid equipment to which each element in the obtained union belongs whose abnormal frequency ratio is greater than or equal to the preset value is taken as Ai j A security-related node in the security-related chain corresponding to the power grid equipment; Ai j The security association chain corresponding to the power grid equipment is composed of Ai j Corresponding to each security-related node and Ai j Corresponding power grid equipment composition;

[0015] The abnormal frequency ratio of each element in the obtained union is equal to the corresponding element in Ai jThe number of occurrences of each abnormal sensor set and Ai j The ratio of the number of corresponding abnormal sensor sets;

[0016] When Ai j When the corresponding abnormal sensor set is an empty set, it is determined that Ai j The security association chain corresponding to the power grid equipment only contains Ai j Corresponding power grid equipment.

[0017] The present invention constructs a security association chain for power grid equipment, the purpose of which is to analyze the operating status of other power grid equipment when the power grid equipment is abnormal, and to lock the association relationship between the abnormal operating status of different power grid equipment from the perspective of big data, thereby providing data support for calculating the security inspection chaos intervention value of each power grid equipment in the subsequent steps.

[0018] Furthermore, in the process of calculating the potential risk assessment value of each power grid device in S2,

[0019] Obtain the complexity of the surrounding environment of each power grid device and the abnormal mutation characteristics of the corresponding power grid device monitoring results, and record the complexity of the surrounding environment of the i-th power grid device as EC i , the abnormal mutation feature of the monitoring result of the i-th power grid equipment is recorded as AMF i ;

[0020] in, ,

[0021] Ni represents the number of perception sensors bound to the i-th power grid equipment; Indicates the average number of sensing sensors bound to each power grid device in the area to which the power grid to be tested belongs;

[0022] in, ,

[0023] MRi k Represents the monitoring result of the sensor bound to the i-th power grid device; NMi k LNi represents the midpoint of the normal monitoring interval preset in the database for the kth sensor bound to the i-th power grid device; k represents the interval length of the normal monitoring interval preset in the database by the kth perception sensor bound to the i-th power grid device;

[0024] ,

[0025] ,

[0026] min{} represents a function that finds the minimum value; max{} represents a function that finds the maximum value; average{} represents a function that finds the average value;

[0027] The potential risk assessment value of the i-th power grid equipment is recorded as Rav i ,

[0028] ,

[0029] Among them, β represents the risk assessment conversion coefficient preset in the database.

[0030] In the present invention express When The maximum value in ;

[0031] express When The minimum value in ;

[0032] Furthermore, when calculating the security inspection chaos intervention value of each power grid device in S3, the security inspection chaos intervention value of the i-th power grid device is recorded as PC i ,

[0033] ,

[0034] Among them, Ti represents the interval between the time point of the most recent abnormal state of the i-th power grid equipment and the current time;

[0035] Represents the average value of the interval between any two adjacent time points corresponding to abnormal status of the i-th power grid equipment in the historical data;

[0036] TP represents the average time interval between the last abnormal state of each power grid device in the area to be tested and the current time;

[0037] It represents the average value of the interval between two consecutive abnormal status time points of any power grid equipment in the area to which the tested power grid belongs; and ,TP and All are not 0;

[0038] G i represents the number of chain nodes in the security association chain containing the i-th power grid equipment; r represents the conversion coefficient preset in the database;

[0039] When predicting the abnormal risk probability of each power grid device in S3, the maximum security patrol chaos intervention values ​​corresponding to the abnormality of the i-th power grid device in the historical data are counted, and the statistical results are divided into the preset security patrol chaos intervention value interval corresponding to the i-th power grid device, and a risk assessment reference data pair corresponding to each security patrol chaos intervention value interval is constructed, wherein the first value in the risk assessment reference data pair is the midpoint of the corresponding security patrol chaos intervention value interval; the second value in the risk assessment reference data pair is the quotient of the number of division results of the statistical security patrol chaos intervention values ​​belonging to the corresponding security patrol chaos intervention value interval divided by the product of the number of elements in the statistical results and the interval length of the preset security patrol chaos intervention value interval; the interval lengths of the preset security patrol chaos intervention value intervals are all equal and the intersection between different security patrol chaos intervention value intervals is empty;

[0040] In the function fitting software, each risk assessment reference data pair corresponding to the constructed i-th power grid equipment is fitted according to the normal distribution function model to obtain the abnormal risk fitting function of the i-th power grid equipment;

[0041] The abnormal risk probability of the i-th power grid equipment is recorded as ARP i ,

[0042] ,

[0043] Among them, Hi{} represents the abnormal risk fitting function of the i-th power grid equipment; Hi{g} represents the function value corresponding to Hi{} when the security inspection chaos intervention value is g.

[0044] The function fitting software in the present invention includes matlab software or Origin software.

[0045] Furthermore, when dynamically dividing the machine inspection area in the area to which the power grid to be tested belongs, all power grid devices in the area to which the power grid to be tested belongs, whose corresponding abnormal risk probability is greater than a second preset value, are screened, and the set of the screened power grid devices is recorded as the first initial candidate set;

[0046] The position points of each element in the first initial candidate set in the area to which the power grid to be tested belongs are marked respectively, and the largest area enclosed by connecting the marked position points is recorded as the human inspection area; the remaining area in the area to which the power grid to be tested belongs except the human inspection area is recorded as the machine inspection area.

[0047] Furthermore, the security warning result in S4 includes a recommended route for human inspection and a recommended route for machine inspection.

[0048] When obtaining the recommended routes for manual inspection and machine inspection, the machine inspection area or the manual inspection area is used as the marking area, the positions corresponding to each power grid equipment in the marking area are obtained, and the power grid equipment in the marking area is arranged and combined to obtain different arrangement and combination results; the splicing results of the path fragments corresponding to the position points of any two adjacent elements in each arrangement and combination result in the preset form of the database are used as the inspection planning route corresponding to the corresponding arrangement and combination; the inspection planning route with the shortest distance among the inspection planning routes corresponding to the marked area is used as the best recommended route,

[0049] The best recommended route corresponding to the marked area when it is a human inspection area is recorded as the human inspection recommended route; the best recommended route corresponding to the marked area when it is a machine inspection area is recorded as the machine inspection recommended route.

[0050] A power grid security supervision system based on multimodal sensing technology, the system includes the following modules:

[0051] A security association analysis module, which obtains the connection relationship between the power grid equipment in the area to which the power grid to be tested belongs and the various sensing sensors bound to each power grid equipment, obtains the monitoring results of each sensing sensor in real time, and builds a security association chain of the power grid equipment in combination with the monitoring results between the sensing sensors in the historical data;

[0052] A risk assessment and analysis module, which obtains the complexity of the surrounding environment of each power grid device and the abnormal mutation characteristics of the corresponding power grid device monitoring results, and calculates the potential risk assessment value of each power grid device;

[0053] Inspection chaos state analysis module, which calculates the security inspection chaos intervention value of each power grid device based on the security association chain of the power grid device and the potential risk assessment value of each power grid device; combines the abnormal data of the power grid device in the historical data to predict the abnormal risk probability of each power grid device;

[0054] The security warning feedback module dynamically divides the machine inspection area in the area to which the power grid to be tested belongs based on the abnormal risk probability prediction value of each power grid device, and generates security warning results for feedback.

[0055] Furthermore, the inspection chaos state analysis module includes a security inspection chaos intervention value calculation unit and an abnormal risk probability prediction unit.

[0056] The security inspection chaos intervention value calculation unit calculates the security inspection chaos intervention value of each power grid device based on the security association chain of the power grid device and the potential risk assessment value of each power grid device;

[0057] The abnormal risk probability prediction unit predicts the abnormal risk probability of each power grid device in combination with the abnormal data of the power grid device in the historical data.

[0058] Compared with the prior art, the beneficial effect achieved by the present invention is that in the process of power grid security supervision based on multimodal sensing technology, in addition to sensing the operating status of power grid equipment through multimodal sensing sensors, the present invention also takes into account the complexity of the operating environment to which the power grid equipment belongs, and combines the abnormal data of the power grid equipment in the historical data to predict the abnormal risk probability of each power grid equipment, and then dynamically divides the machine inspection area and the manual inspection area in the area to which the power grid to be tested belongs, thereby reducing the risk of staff's secondary inspection of power grid equipment, while ensuring the timeliness and accuracy of the inspection results, and realizing effective supervision of power grid security data. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0060] Figure 1 It is a structural schematic diagram of a power grid security monitoring system based on multimodal sensing technology of the present invention;

[0061] Figure 2 It is a flow chart of a power grid security supervision method based on multimodal sensing technology of the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] See also Figure 1 The present invention provides a technical solution: a power grid security supervision system based on multimodal sensing technology, the system comprising the following modules:

[0064] A security association analysis module, which obtains the connection relationship between the power grid equipment in the area to which the power grid to be tested belongs and the various sensing sensors bound to each power grid equipment, obtains the monitoring results of each sensing sensor in real time, and builds a security association chain of the power grid equipment in combination with the monitoring results between the sensing sensors in the historical data;

[0065] A risk assessment and analysis module, which obtains the complexity of the surrounding environment of each power grid device and the abnormal mutation characteristics of the corresponding power grid device monitoring results, and calculates the potential risk assessment value of each power grid device;

[0066] Inspection chaos state analysis module, which calculates the security inspection chaos intervention value of each power grid device based on the security association chain of the power grid device and the potential risk assessment value of each power grid device; combines the abnormal data of the power grid device in the historical data to predict the abnormal risk probability of each power grid device;

[0067] The security warning feedback module dynamically divides the machine inspection area in the area to which the power grid to be tested belongs based on the abnormal risk probability prediction value of each power grid device, and generates security warning results for feedback.

[0068] The inspection chaos state analysis module includes a security inspection chaos intervention value calculation unit and an abnormal risk probability prediction unit.

[0069] The security inspection chaos intervention value calculation unit calculates the security inspection chaos intervention value of each power grid device based on the security association chain of the power grid device and the potential risk assessment value of each power grid device;

[0070] The abnormal risk probability prediction unit predicts the abnormal risk probability of each power grid device in combination with the abnormal data of the power grid device in the historical data.

[0071] like Figure 2 As shown, a power grid security supervision method based on multimodal sensing technology includes the following steps:

[0072] S1. Obtain the connection relationship between the power grid equipment in the area to which the power grid to be tested belongs and the various sensing sensors bound to each power grid equipment, obtain the monitoring results of each sensing sensor in real time, and build a security association chain for the power grid equipment by combining the monitoring results between the sensing sensors in the historical data;

[0073] There are differences in the monitoring results obtained by different perception sensors. The jth perception sensor bound to the i-th power grid device in the area to be tested is recorded as Ai j ;

[0074] When building the security association chain of power grid equipment, obtain the abnormal monitoring data corresponding to the sensor in the historical data and extract Ai j Each time the corresponding monitoring data is abnormal, the remaining sensor sensors with abnormal monitoring results are combined to obtain Ai j The corresponding set of abnormal sensors;

[0075] Get Aij The corresponding union of each abnormal sensor set is obtained, and the abnormal frequency ratio of each element in the obtained union is calculated respectively. The power grid equipment to which each element in the obtained union belongs whose abnormal frequency ratio is greater than or equal to the preset value is taken as Ai j A security-related node in the security-related chain corresponding to the power grid equipment; Ai j The security association chain corresponding to the power grid equipment is composed of Ai j Corresponding to each security-related node and Ai j Corresponding power grid equipment composition;

[0076] The abnormal frequency ratio of each element in the obtained union is equal to the corresponding element in Ai j The number of occurrences of each abnormal sensor set and Ai j The ratio of the number of corresponding abnormal sensor sets;

[0077] When Ai j When the corresponding abnormal sensor set is an empty set, it is determined that Ai j The security association chain corresponding to the power grid equipment only contains Ai j Corresponding power grid equipment.

[0078] S2. Obtain the complexity of the surrounding environment of each power grid device and the abnormal mutation characteristics of the corresponding power grid device monitoring results, and calculate the potential risk assessment value of each power grid device;

[0079] In the process of calculating the potential risk assessment value of each power grid device in S2,

[0080] Obtain the complexity of the surrounding environment of each power grid device and the abnormal mutation characteristics of the corresponding power grid device monitoring results, and record the complexity of the surrounding environment of the i-th power grid device as EC i , the abnormal mutation feature of the monitoring result of the i-th power grid equipment is recorded as AMF i ;

[0081] in, ,

[0082] Ni represents the number of perception sensors bound to the i-th power grid equipment; Indicates the average number of sensing sensors bound to each power grid device in the area to which the power grid to be tested belongs;

[0083] in, ,

[0084] MRi k Represents the monitoring result of the sensor bound to the i-th power grid device; NMi k LNi represents the midpoint of the normal monitoring interval preset in the database for the kth sensor bound to the i-th power grid device;k represents the interval length of the normal monitoring interval preset in the database by the kth perception sensor bound to the i-th power grid device;

[0085] ,

[0086] ,

[0087] min{} represents a function that finds the minimum value; max{} represents a function that finds the maximum value; average{} represents a function that finds the average value;

[0088] The potential risk assessment value of the i-th power grid equipment is recorded as Rav i ,

[0089] ,

[0090] Among them, β represents the risk assessment conversion coefficient preset in the database.

[0091] S3. Based on the security association chain of power grid equipment and the potential risk assessment value of each power grid equipment, calculate the security inspection chaos intervention value of each power grid equipment; combine the abnormal data of power grid equipment in historical data to predict the abnormal risk probability of each power grid equipment;

[0092] When calculating the security inspection chaos intervention value of each power grid device in S3, the security inspection chaos intervention value of the i-th power grid device is recorded as PC i ,

[0093] ,

[0094] Among them, Ti represents the interval between the time point of the most recent abnormal state of the i-th power grid equipment and the current time;

[0095] Represents the average value of the interval between any two adjacent time points corresponding to abnormal status of the i-th power grid equipment in the historical data;

[0096] TP represents the average time interval between the last abnormal state of each power grid device in the area to be tested and the current time;

[0097] It represents the average value of the interval between two consecutive abnormal status time points of any power grid equipment in the area to which the tested power grid belongs; and ,TP and All are not 0;

[0098] G irepresents the number of chain nodes in the security association chain containing the i-th power grid equipment; r represents the conversion coefficient preset in the database;

[0099] When predicting the abnormal risk probability of each power grid device in S3, the maximum security patrol chaos intervention values ​​corresponding to the abnormality of the i-th power grid device in the historical data are counted, and the statistical results are divided into the preset security patrol chaos intervention value interval corresponding to the i-th power grid device, and a risk assessment reference data pair corresponding to each security patrol chaos intervention value interval is constructed, wherein the first value in the risk assessment reference data pair is the midpoint of the corresponding security patrol chaos intervention value interval; the second value in the risk assessment reference data pair is the quotient of the number of division results of the statistical security patrol chaos intervention values ​​belonging to the corresponding security patrol chaos intervention value interval divided by the product of the number of elements in the statistical results and the interval length of the preset security patrol chaos intervention value interval; the interval lengths of the preset security patrol chaos intervention value intervals are all equal and the intersection between different security patrol chaos intervention value intervals is empty;

[0100] In the function fitting software, each risk assessment reference data pair corresponding to the constructed i-th power grid equipment is fitted according to the normal distribution function model to obtain the abnormal risk fitting function of the i-th power grid equipment;

[0101] The abnormal risk probability of the i-th power grid equipment is recorded as ARP i ,

[0102] ,

[0103] Among them, Hi{} represents the abnormal risk fitting function of the i-th power grid equipment; Hi{g} represents the function value corresponding to Hi{} when the security inspection chaos intervention value is g.

[0104] S4. Based on the abnormal risk probability prediction value of each power grid equipment, dynamically divide the machine inspection area in the area where the power grid to be tested belongs, and generate security warning results for feedback.

[0105] When dynamically dividing the machine inspection area in the area to which the power grid to be tested belongs, all power grid devices in the area to which the power grid to be tested belongs whose corresponding abnormal risk probability is greater than a second preset value are screened, and a set of the screened power grid devices is recorded as a first initial candidate set;

[0106] The position points of each element in the first initial candidate set in the area to which the power grid to be tested belongs are marked respectively, and the largest area enclosed by connecting the marked position points is recorded as the human inspection area; the remaining area in the area to which the power grid to be tested belongs except the human inspection area is recorded as the machine inspection area.

[0107] The security warning results in S4 include the recommended routes for human inspection and the recommended routes for machine inspection.

[0108] When obtaining the recommended routes for manual inspection and machine inspection, the machine inspection area or the manual inspection area is used as the marking area, the positions corresponding to each power grid equipment in the marking area are obtained, and the power grid equipment in the marking area is arranged and combined to obtain different arrangement and combination results; the splicing results of the path fragments corresponding to the position points of any two adjacent elements in each arrangement and combination result in the preset form of the database are used as the inspection planning route corresponding to the corresponding arrangement and combination; the inspection planning route with the shortest distance among the inspection planning routes corresponding to the marked area is used as the best recommended route,

[0109] The best recommended route corresponding to the marked area when it is a human inspection area is recorded as the human inspection recommended route; the best recommended route corresponding to the marked area when it is a machine inspection area is recorded as the machine inspection recommended route.

[0110] In this embodiment, if there are three power grid devices in the marked area and they are marked as A1, A2 and A3 respectively, then the permutation and combination results of the power grid devices in the marked area are: {A1, A2, A3}, {A1, A3, A2}, {A2, A1, A3}, {A2, A3, A1}, {A3, A1, A2} and {A3, A2, A1} respectively;

[0111] If the distance between A1 and A2 is L (A1,A2) , if the distance between A1 and A3 is L (A1,A3) , if the distance between A2 and A3 is L (A2,A3) ,

[0112] Then the distances of {A1, A2, A3} and {A3, A2, A1} are both L (A1,A2) +L (A2,A3) ;

[0113] The distances of {A1, A3, A2} and {A2, A3, A1} are both L (A1,A3) +L (A2,A3) ;

[0114] The distances of {A2, A1, A3} and {A3, A1, A2} are both L (A1,A2) +L (A2,A3) .

[0115] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device.

[0116] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions described in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A power grid security supervision method based on multimodal sensing technology, It is characterized in that The method comprises the following steps: S1. Obtain the connection relationship between the power grid equipment in the area to which the power grid to be tested belongs and the various sensing sensors bound to each power grid equipment, obtain the monitoring results of each sensing sensor in real time, and build a security association chain for the power grid equipment by combining the monitoring results between the sensing sensors in the historical data; S2. Obtain the complexity of the surrounding environment of each power grid device and the abnormal mutation characteristics of the corresponding power grid device monitoring results, and calculate the potential risk assessment value of each power grid device; S3. Based on the security association chain of power grid equipment and the potential risk assessment value of each power grid equipment, calculate the security inspection chaos intervention value of each power grid equipment; combine the abnormal data of power grid equipment in historical data to predict the abnormal risk probability of each power grid equipment; S4. Based on the abnormal risk probability prediction value of each power grid device, dynamically divide the machine inspection area in the area where the power grid to be tested belongs, and generate security warning results for feedback; There are differences in the monitoring results obtained by different perception sensors. The jth perception sensor bound to the i-th power grid device in the area to which the power grid to be tested belongs is recorded as Ai j ; When building the security association chain of power grid equipment, obtain the abnormal monitoring data corresponding to the sensor in the historical data and extract the Ai j Each time the corresponding monitoring data is abnormal, the remaining sensor sensors with abnormal monitoring results are combined to obtain Ai j The corresponding set of abnormal sensors; Get Ai j The corresponding union of each abnormal sensor set is obtained, and the abnormal frequency ratio of each element in the obtained union is calculated respectively. The power grid equipment to which each element in the obtained union belongs whose abnormal frequency ratio is greater than or equal to the preset value is taken as Ai j A security-related node in the security-related chain corresponding to the power grid equipment; Ai j The security association chain corresponding to the power grid equipment is composed of Ai j Corresponding to each security-related node and Ai j Corresponding power grid equipment composition; The abnormal frequency ratio of each element in the obtained union is equal to the corresponding element in Ai j The number of occurrences of each abnormal sensor set and Ai j The ratio of the number of corresponding abnormal sensor sets; When Ai j When the corresponding abnormal sensor set is an empty set, it is determined that Ai j The security association chain corresponding to the power grid equipment only contains Ai j Corresponding power grid equipment; In the process of calculating the potential risk assessment value of each power grid device in S2, Obtain the complexity of the surrounding environment of each power grid device and the abnormal mutation characteristics of the corresponding power grid device monitoring results, and record the complexity of the surrounding environment of the i-th power grid device as EC i , the abnormal mutation feature of the monitoring result of the i-th power grid equipment is recorded as AMF i ; in, , Ni represents the number of perception sensors bound to the i-th power grid equipment; Indicates the average number of sensing sensors bound to each power grid device in the area to which the power grid to be tested belongs; in, , MRi k Represents the monitoring result of the sensor bound to the i-th power grid device; NMi k LNi represents the midpoint of the normal monitoring interval preset in the database for the kth sensor bound to the i-th power grid device; k represents the interval length of the normal monitoring interval preset in the database by the kth perception sensor bound to the i-th power grid device; , , min{} represents a function that finds the minimum value; max{} represents a function that finds the maximum value; average{} represents a function that finds the average value; The potential risk assessment value of the i-th power grid equipment is recorded as Rav i , , Among them, β represents the risk assessment conversion coefficient preset in the database; When calculating the security inspection chaos intervention value of each power grid device in S3, the security inspection chaos intervention value of the i-th power grid device is recorded as PC i , , Among them, Ti represents the interval between the time point of the most recent abnormal state of the i-th power grid equipment and the current time; Represents the average value of the interval between any two adjacent time points corresponding to abnormal status of the i-th power grid equipment in the historical data; TP represents the average time interval between the last abnormal state of each power grid device in the area to which the tested power grid belongs and the current time; It represents the average time interval between two consecutive abnormal status points of any power grid equipment in the area to which the tested power grid belongs; and ,TP and All are not 0; G i represents the number of chain nodes in the security association chain containing the i-th power grid equipment; r represents the conversion coefficient preset in the database; When predicting the abnormal risk probability of each power grid device in S3, the maximum security patrol chaos intervention values ​​corresponding to the abnormality of the i-th power grid device in the historical data are counted, and the statistical results are divided into the preset security patrol chaos intervention value interval corresponding to the i-th power grid device, and a risk assessment reference data pair corresponding to each security patrol chaos intervention value interval is constructed, wherein the first value in the risk assessment reference data pair is the midpoint of the corresponding security patrol chaos intervention value interval; the second value in the risk assessment reference data pair is the quotient of the number of division results of the statistical security patrol chaos intervention values ​​belonging to the corresponding security patrol chaos intervention value interval divided by the product of the number of elements in the statistical results and the interval length of the preset security patrol chaos intervention value interval; the interval lengths of the preset security patrol chaos intervention value intervals are all equal and the intersection between different security patrol chaos intervention value intervals is empty; In the function fitting software, each risk assessment reference data pair corresponding to the constructed i-th power grid equipment is fitted according to the normal distribution function model to obtain the abnormal risk fitting function of the i-th power grid equipment; The abnormal risk probability of the i-th power grid equipment is recorded as ARP i , , Among them, Hi{} represents the abnormal risk fitting function of the i-th power grid equipment; Hi{g} represents the function value corresponding to Hi{} when the security inspection chaos intervention value is g.

2. According to claim 1, a power grid security supervision method based on multimodal sensing technology, Features: When dynamically dividing the machine inspection area in the area to which the power grid to be tested belongs, all power grid devices in the area to which the power grid to be tested belongs whose corresponding abnormal risk probability is greater than a second preset value are screened, and a set of the screened power grid devices is recorded as a first initial candidate set; Marking the location points of each element in the first initial candidate set in the area to which the power grid to be tested belongs, and recording the largest area enclosed by connecting the marked location points as the manual inspection area; The remaining area of ​​the area to be tested power grid except the manual inspection area is recorded as the machine inspection area.

3. According to claim 2, a power grid security supervision method based on multimodal sensing technology, Features: The security warning results in S4 include the recommended routes for human inspection and the recommended routes for machine inspection. When obtaining the recommended routes for manual inspection and machine inspection, the machine inspection area or the manual inspection area is used as the marking area, the positions corresponding to each power grid equipment in the marking area are obtained, and the power grid equipment in the marking area is arranged and combined to obtain different arrangement and combination results; the splicing results of the path fragments corresponding to the position points of any two adjacent elements in each arrangement and combination result in the preset form of the database are used as the inspection planning route corresponding to the corresponding arrangement and combination; the inspection planning route with the shortest distance among the inspection planning routes corresponding to the marked area is used as the best recommended route, The best recommended route corresponding to the marked area when it is a human inspection area is recorded as the human inspection recommended route; the best recommended route corresponding to the marked area when it is a machine inspection area is recorded as the machine inspection recommended route.

4. A power grid security supervision system based on multimodal sensing technology, applying a power grid security supervision method based on multimodal sensing technology as described in any one of claims 1 to 3, It is characterized in that The system includes the following modules: A security association analysis module, which obtains the connection relationship between the power grid equipment in the area to which the power grid to be tested belongs and the various sensing sensors bound to each power grid equipment, obtains the monitoring results of each sensing sensor in real time, and builds a security association chain of the power grid equipment in combination with the monitoring results between the sensing sensors in the historical data; A risk assessment and analysis module, which obtains the complexity of the surrounding environment of each power grid device and the abnormal mutation characteristics of the corresponding power grid device monitoring results, and calculates the potential risk assessment value of each power grid device; Inspection chaos state analysis module, which calculates the security inspection chaos intervention value of each power grid device based on the security association chain of the power grid device and the potential risk assessment value of each power grid device; combines the abnormal data of the power grid device in the historical data to predict the abnormal risk probability of each power grid device; The security warning feedback module dynamically divides the machine inspection area in the area to which the power grid to be tested belongs based on the abnormal risk probability prediction value of each power grid device, and generates security warning results for feedback.

5. According to claim 4, a power grid security monitoring system based on multimodal sensing technology, Features: The inspection chaos state analysis module includes a security inspection chaos intervention value calculation unit and an abnormal risk probability prediction unit. The security inspection chaos intervention value calculation unit calculates the security inspection chaos intervention value of each power grid device based on the security association chain of the power grid device and the potential risk assessment value of each power grid device; The abnormal risk probability prediction unit predicts the abnormal risk probability of each power grid device in combination with the abnormal data of the power grid device in the historical data.