A method and related device for identifying key inducing factors of typical disasters in power systems with integrated sky, ground, communication, conduction and sensing.

Abstract

The present invention provides a method and related device for identifying key inducing factors of typical disasters in power systems that integrates sky, ground, communication, conduction and sensing, and relates to the technical field of power systems. Images, videos and text data of typical disasters in power systems are obtained through sky-ground network detection modules to generate multi-source data, and a multimodal data fusion system is established based on the multi-source data; a typical disaster event database is constructed in the multimodal data fusion system; continuous features and discrete features are obtained in the typical disaster event database to generate a parallel evaluation method; a key inducing factor identification mechanism is constructed based on probability distribution and fuzzy set theory; and a scoring result is determined based on the parallel evaluation method combined with the key inducing factor identification mechanism to determine whether there are high-risk factors. By mining high-risk and low-probability elements, a comprehensive consideration of key inducing factors of typical disasters is achieved, and the disaster warning capability of the power system is improved.

Description

Technical Field

[0001] The present application relates to the technical field of power systems, and in particular to a method and related device for identifying key inducing factors of typical disasters in power systems that integrates air, ground, communication, conduction and sensing. Background Art

[0002] With the widespread access to new energy sources, the new power system is facing more and more disaster threats in all links of source, grid, load and storage. The source-grid-load-storage links of the new power system may face various disaster threats, such as natural disasters, technical failures, etc. Therefore, it is particularly important to identify and analyze the key inducing factors of these disasters. In this context, satellite Internet technology is regarded as an effective tool to strengthen the identification of key inducing factors of disasters in new power systems because of its wide-area coverage and homogeneous service characteristics.

[0003] The new power system enabled by satellite Internet covers a multi-mode three-dimensional monitoring network integration system that integrates space, ground and space. It uses space-based sensing modules, air-based sensing modules and ground-based sensing modules to conduct comprehensive monitoring of the power system source-grid-load-storage links in the sensing area. At the same time, it has a multi-functional sensing system coordination mechanism that integrates communication, navigation and sensing. Combined with the potential connection between communication, navigation and sensing functions, it provides comprehensive communication, perception and navigation positioning enhancement services to the sensing area. The new power system source-grid-load-storage full-process risk situation perception architecture that integrates space, ground and space, communication, navigation and sensing provides massive multi-source heterogeneous data support for the risk perception of typical disasters in the new power system, improving the speed and accuracy of disaster perception. However, the existing methods for identifying key inducing factors of typical disasters in the new power system under the fusion of satellite and ground data still have the following defects: (1) Under the integration of space, ground and space, communication, navigation and sensing, the data collected for the new power system usually covers multi-modal data such as video, image and text. However, the current system analyzes a single type of data, such as image or numerical data, and does not pay enough attention to the comprehensive analysis of multi-modal data such as video, image and text. (2) The disaster monitoring data of the new power system with integrated sky, ground, communication, guidance and sensing contains both discrete and continuous data characteristics. Directly discretizing the continuous characteristics will lead to greater subjectivity and randomness, and at the same time cause a large amount of information waste. (3) For high-risk and low-probability events in the new power system with integrated sky, ground, communication, guidance and sensing, such as equipment performance under extreme weather conditions, it is often difficult to conduct effective analysis due to scarce data. However, the impact of these rare and high-risk events on the safe operation of the power system cannot be ignored.

[0004] Therefore, how to achieve a comprehensive consideration of the key inducing factors of typical disasters and improve the disaster warning capability of the power system has become a technical problem that needs to be solved urgently. Summary of the invention

[0005] In order to achieve a comprehensive consideration of the key inducing factors of typical disasters and improve the disaster warning capability of the power system, the present application provides a method and related devices for identifying the key inducing factors of typical disasters in the power system that integrates air, ground, communication, guidance and sensing.

[0006] In the first aspect, the present application provides a method for identifying key inducing factors of typical disasters in power systems that integrates air, ground, communication, conduction and sensing, using the following technical solutions:

[0007] A method for identifying key inducing factors of typical disasters in power systems that integrates air, ground, communication, conduction and sensing, including:

[0008] Acquire images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establish a multi-modal data fusion system based on the multi-source data;

[0009] In the multimodal data fusion system, a typical disaster event database is constructed through comprehensive perception of the target;

[0010] Acquire continuous features and discrete features in the typical disaster event database to generate a parallel evaluation method;

[0011] Construct the key inducing factor identification mechanism based on probability distribution and fuzzy set theory;

[0012] The scoring result is determined according to the parallel evaluation method in combination with the key predisposing factor identification mechanism to determine whether there are high-risk factors.

[0013] Optionally, the step of acquiring images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establishing a multi-modal data fusion system based on the multi-source data includes:

[0014] Obtain image information, video information and text data information of typical disasters in the power system through the sky-ground network detection module;

[0015] The image information of typical disasters in power system is homogeneously fused through weighted average formula;

[0016] Heterogeneously fusing the video information and text data to extract multimodal key features and comprehensively perceive the target;

[0017] Generate multi-source data according to the homogeneous fusion result and the heterogeneous fusion result;

[0018] A preset feature screening mechanism is obtained and a multi-source data is used to establish a multi-modal data fusion system.

[0019] Optionally, the step of performing isomorphic fusion on the image information of typical disasters in the power system by using a weighted average formula includes:

[0020] Pixel-level fusion of image information of typical disasters in power system through weighted average formula

[0021]

[0022] in, is the weight coefficient, which is adjusted according to the reliability and relevance of the data source. and are two image sources to be fused, is the fused image.

[0023] Optionally, the step of constructing a typical disaster event database by comprehensively perceiving the target in the multimodal data fusion system includes:

[0024] Obtaining disaster event records and corresponding feature information in the multimodal data fusion system and performing unified feature standardization to form a structured database;

[0025] Disaster event records, characteristic values ​​and target factors are stored in the structured database in a matrix form to construct a structured database.

[0026] Optionally, the step of acquiring continuous features and discrete features in the typical disaster event database to generate a parallel evaluation method includes:

[0027] By continuous feature analysis in the disaster event database, a probability distribution curve is drawn, the value interval is divided into three categories: common, possible and rare, and a membership function is constructed;

[0028] Conduct discrete feature analysis and define the state thresholds of feature components based on fuzzy set theory;

[0029] Risk assessment was performed on continuous and discrete feature analysis results separately to generate a parallel assessment method.

[0030] Optionally, the step of constructing a key inducing factor identification mechanism based on probability distribution and fuzzy set theory includes:

[0031] Constructing support thresholds of feature-fuzzy sets to calculate the occurrence frequency of rare features in historical disaster records;

[0032] Construct the certainty threshold of feature-fuzzy set, and verify the deterministic association results between rare factors and disaster outcomes by combining fuzzy weights and membership.

[0033] Construct the correlation threshold of feature-fuzzy set and quantify the correlation strength between rare factors and disasters through expected value and probability distribution;

[0034] A key inducing factor identification mechanism is constructed according to the support threshold of the feature-fuzzy set, the certainty threshold of the feature-fuzzy set and the correlation threshold of the feature-fuzzy set.

[0035] Optionally, the support threshold of the feature-fuzzy set is expressed as:

[0036]

[0037] in, represents a rare data set, t i is the i-th fault record in the evaluation space B, x i,j Is a feature in the fault record The components of and Indicates the total amount of record B and The factor, card() means the total number of qualified records, Th s express The support threshold of The ratio of the number of disaster records of the characteristic to the total number is determined t i ( x i,j )represent x i,j exist t i The value of or Describes the fuzzy set The membership degree of is the corresponding fuzzy weight;

[0038] The certainty threshold of the feature-fuzzy set can be expressed as:

[0039]

[0040] In the formula and represent rare factor sets and related fuzzy sets respectively, represent The corresponding fuzzy weights, express The certainty threshold, or Describes the fuzzy set The membership degree of , and the condition for the threshold formula of the certainty of the feature-fuzzy set to be established is:

[0041]

[0042] The correlation threshold of the feature-fuzzy set can be expressed as:

[0043]

[0044] Where Z=X Y,L=P Q,ME[…] stands for expectation, , , , and , respectively expressed as:

[0045]

[0046]

[0047]

[0048]

[0049]

[0050] in,

[0051] in, Th co express The correlation threshold.

[0052] In a second aspect, the present application provides a system for identifying key factors causing typical disasters in a power system that integrates air, ground, communication, conduction and sensing. The system for identifying key factors causing typical disasters in a power system that integrates air, ground, communication, conduction and sensing includes:

[0053] A data acquisition module, used to acquire images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establish a multi-modal data fusion system based on the multi-source data;

[0054] A database construction module, used to construct a typical disaster event database through comprehensive perception of the target in the multimodal data fusion system;

[0055] A parallel evaluation module, used for acquiring continuous features and discrete features in the typical disaster event database to generate a parallel evaluation method;

[0056] Identification mechanism module, used to construct key inducing factor identification mechanism based on probability distribution and fuzzy set theory;

[0057] A judgment module is used to determine the score result according to the parallel evaluation method combined with the key inducing factor identification mechanism to determine whether there is a high-risk factor.

[0058] In a third aspect, the present application provides a computer device, comprising: a memory and a processor, wherein the processor executes the method described above when running computer instructions stored in the memory.

[0059] In a fourth aspect, the present application provides a computer-readable storage medium, comprising instructions, which, when executed on a computer, enable the computer to execute the method described above.

[0060] In summary, the present application includes the following beneficial technical effects:

[0061] This application obtains images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establishes a multimodal data fusion system based on the multi-source data; constructs a typical disaster event database in the multimodal data fusion system; obtains continuous features and discrete features in the typical disaster event database to generate a parallel evaluation method; constructs a key inducing factor identification mechanism based on probability distribution and fuzzy set theory; determines the score results based on the parallel evaluation method combined with the key inducing factor identification mechanism to determine whether there are high-risk factors. By mining high-risk and low-probability elements, a comprehensive consideration of the key inducing factors of typical disasters can be achieved, and the disaster warning capability of the power system can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application.

[0063] Figure 2 It is a flow chart of an embodiment of a method for identifying key inducing factors of typical disasters in an electric power system that integrates air, ground, communication, conduction and sensing in the present application.

[0064] Figure 3 This is a schematic diagram of the identification mechanism of key inducing factors of typical disasters in the new type of power system with integrated sky-ground, communication-conducting-sensing and sensing in this application;

[0065] Figure 4 This is a membership function diagram of the present application taking ambient temperature as an example;

[0066] Figure 5 It is a structural block diagram of an embodiment of a system for identifying key inducing factors of typical disasters in power systems that integrates air, ground, communication, conduction and sensing in the present application. DETAILED DESCRIPTION

[0067] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below through the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0068] Reference Figure 1 , Figure 1 A schematic diagram of the computer device structure of the hardware operating environment involved in the embodiment of the present application.

[0069] like Figure 1 As shown, the computer device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components. The user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard), and the optional user interface 1003 may also include a standard wired interface and a wireless interface. The network interface 1004 may optionally include a standard wired interface and a wireless interface (such as a wireless fidelity (Wireless-Fidelity, Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (Random Access Memory, RAM), or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk storage. The memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0070] Those skilled in the art will understand that Figure 1 The structure shown in the figure does not constitute a limitation on the computer device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.

[0071] like Figure 1 As shown, the memory 1005 as a storage medium may include an operating system, a network communication module, a user interface module, and a program for identifying key inducing factors of typical disasters in a power system that integrates air, ground, communication, conduction, and sensing.

[0072] exist Figure 1In the computer device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in the present application can be set in the computer device, and the computer device calls the key inducing factor identification program of the typical disaster in the power system with integrated sky-ground and integrated communication, conduction and sensing stored in the memory 1005 through the processor 1001, and executes the key inducing factor identification method of the typical disaster in the power system with integrated sky-ground and integrated communication, conduction and sensing provided in the embodiment of the present application.

[0073] The present application embodiment provides a method for identifying key inducing factors of typical disasters in a power system that integrates air, ground, communication, conduction and sensing. Figure 2 , Figure 2 This is a flow chart of an embodiment of a method for identifying key inducing factors of typical disasters in a power system that integrates air, ground, communication, conduction and sensing in this application.

[0074] In this embodiment, the method for identifying key inducing factors of typical disasters in a power system with integrated air-ground and communication-conducting-sensing integration includes the following steps:

[0075] Step S10: Obtain images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establish a multi-modal data fusion system based on the multi-source data.

[0076] It is understandable that if Figure 3 As shown, this embodiment includes the following steps: collecting multi-source data such as images, videos, texts, etc. of typical disasters in the new power system under the sky-ground network monitoring module, establishing a multimodal data fusion system that fully covers sky-ground information, and extracting fault event records and corresponding features; designing typical disaster risk-related patterns in the power system, mining the correlation between the collected fault event records and corresponding features and their components, and normalizing them, building a new power system typical disaster event database, analyzing the data characteristics of typical disaster events in the new power system, and designing a typical disaster key inducing factor analysis method for complex data environments; fully considering the low-frequency environmental elements and less disaster periods in the database, building a typical disaster key inducing factor identification mechanism, mining high-risk and low-probability elements, and realizing a comprehensive consideration of the key inducing factors of typical disasters.

[0077] It should be noted that the steps of obtaining images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establishing a multimodal data fusion system based on the multi-source data include: obtaining image information, video information and text data information of typical disasters in the power system through the sky-ground network detection module; performing homogeneous fusion of image information of typical disasters in the power system through a weighted average formula; performing heterogeneous fusion of video information and text data to extract multimodal key features and comprehensively perceive targets; generating multi-source data based on homogeneous fusion results and heterogeneous fusion results; and obtaining a preset feature screening mechanism and multi-source data to establish a multimodal data fusion system.

[0078] It should be noted that the step of performing isomorphic fusion of image information of typical disasters in the power system by using a weighted average formula includes: performing pixel-level fusion of image information of typical disasters in the power system by using a weighted average formula

[0079]

[0080] in, is the weight coefficient, which is adjusted according to the reliability and relevance of the data source. and are two image sources to be fused, is the fused image.

[0081] In the specific implementation, a multimodal data fusion system covering all sky and ground information is established. Images collected by remote sensing satellites, drone inspections under navigation enhancement, and ground cameras are collected, and homogeneous fusion is performed based on pixel-level fusion, and heterogeneous fusion is performed based on feature-level and decision-level fusion to obtain sky-ground fusion images.

[0082] The sky-ground fusion images, surveillance videos, texts and other modal data in the disaster data set are selected to form a multimodal data set, and the corresponding labels are marked for the disaster event categories; the pre-trained models of various data sources or the feature extraction models of specific tasks are used to extract the key features in each modal data; the modal data features are fused and the fused features are further analyzed to obtain a comprehensive perception of the target.

[0083] Based on the comprehensive perception of the target and combined with the input feature screening mechanism, the disaster event information in the system is extracted and a disaster event database is constructed.

[0084] It should be noted that the corresponding input feature screening mechanism in the steps of extracting disaster event information in the system and building a disaster event database based on comprehensive perception of the target and combining the input feature screening mechanism is as follows: the input features must meet the following requirements: ① Related to most typical disasters in power systems such as wildfires and lightning; ② Available in the entire power system; ③ Available within the study period and statistically significant.

[0085] Step S20: Construct a typical disaster event database through comprehensive perception of the target in the multimodal data fusion system.

[0086] In the specific implementation, before the step of building a typical disaster event database through comprehensive perception of the target in the multimodal data fusion system, it includes: designing typical disaster risk-related patterns of the power system, and mining the correlation between the collected disaster event records and the corresponding characteristics and their components.

[0087] Among them, the typical disaster risk related model of the power system is designed, and the correlation between the collected disaster event records and the corresponding characteristics and their components is mined. Specifically, X = {x i,1 , x i,2 , …, x i,j , …, x i,n Disaster event t All corresponding factors in i,j Can be a feature Any component of Y ={ y 1, y 2, ..., y i , …, y m} is a set of target factors, one of which is a target component y i represents the result of a disaster event. Let P = {p 1,1 ,p 1,2 ,…,p i , j ,…,p m,n} and Q={q1,q2,…,q i ,…,q m} contains fuzzy sets related to X and Y. Then, the typical disaster risk correlation pattern of the power system has the following form:

[0088]

[0089] The above formula indicates that if X corresponds to P, then Y is related to Q. When the voting score of historical disaster records exceeds the voting threshold, the correlation of typical disasters in the power system can be identified.

[0090] It should be noted that the steps of constructing a typical disaster event database through comprehensive perception of the target in the multimodal data fusion system include: obtaining disaster event records and corresponding feature information in the multimodal data fusion system and standardizing the features to form a structured database; storing disaster event records, feature values ​​and target factors in the structured database in a matrix form to construct a structured database.

[0091] In the specific implementation, a typical disaster event database of a new power system is constructed. Specifically, assume that B = {t1, t2, ..., ti, ..., tm} is a set of total m historical records of disaster events (transactions), where i = 1, 2, ..., m represents an included record. In B, assume that F = { , ,…, ,…, ,Y} is a set of features = 1,2,…,n represents one of the total n functional features, and Y represents the predicted target feature. In each feature, there are several components to specify different states, among which ={ , ,…, ,…, }∈F. Therefore, the typical disaster event database of the new power system can be written as:

[0092] =

[0093] The features in the new power system typical disaster event database are divided into continuous features c and discrete features d, which provide a basis for subsequent mining of high-risk and low-probability features in continuous and discrete feature data. The improved new power system typical disaster event database can be written as:

[0094]

[0095] Step S30: Acquire continuous features and discrete features in a typical disaster event database to generate a parallel evaluation method.

[0096] It should be noted that the steps of obtaining continuous features and discrete features in a typical disaster event database to generate a parallel evaluation method include: analyzing continuous features in the disaster event database, drawing a probability distribution curve, dividing the numerical range into three categories: common, possible and rare, and constructing a membership function; performing discrete feature analysis, and defining the state thresholds of feature components based on fuzzy set theory; and performing risk assessment on the continuous feature analysis results and the discrete feature analysis results respectively to generate a parallel evaluation method.

[0097] In the specific implementation, for each continuous feature, according to the occurrence frequency of the continuous feature input in the database, a probability distribution curve is drawn and a membership function is constructed, and the construction of different feature fuzzy sets is realized based on the same form to describe the membership of different fuzzy sets.

[0098] It should be noted that the membership function construction method is as follows: first draw a probability distribution curve, and divide the characteristic value of each continuous feature into three numerical intervals based on historical data and expert experience, such as the ambient temperature (℃) is divided into ~15, 15~20, 20~, and these three numerical intervals are defined as common (O), possible (P), and rare (R); then select the body shape function as the shape of the membership function, and the horizontal and vertical coordinates are the boundaries and memberships between each fuzzy set, respectively, so as to construct the membership function. Figure 4 As shown, Figure 4 EMI1.1 is a membership function in this embodiment, taking temperature as an example.

[0099] Step S40: constructing a key inducing factor identification mechanism based on probability distribution and fuzzy set theory.

[0100] It can be understood that the support threshold of the feature-fuzzy set is expressed as:

[0101]

[0102] in, represents a rare data set, t i is the i-th fault record in the evaluation space B, x i,j Is a feature in the fault record The components of and Indicates the total amount of record B and The factor, card() means the total number of qualified records, Th s express The support threshold of The ratio of the number of disaster records of the characteristic to the total number is determined t i ( x i,j )represent x i,j exist t i The value of or Describes the fuzzy set The membership degree of is the corresponding fuzzy weight;

[0103] The certainty threshold of the feature-fuzzy set can be expressed as:

[0104]

[0105] In the formula and represent rare factor sets and related fuzzy sets respectively, represent The corresponding fuzzy weights, express The certainty threshold, or Describes the fuzzy set The membership degree of , and the condition for the threshold formula of the certainty of the feature-fuzzy set to be established is:

[0106]

[0107] The feature-fuzzy set correlation threshold can be expressed as:

[0108]

[0109] Where Z=X Y,L=P Q,ME[…] stands for expectation, , , , and , respectively expressed as:

[0110]

[0111]

[0112]

[0113]

[0114]

[0115] in,

[0116] in, express The correlation threshold.

[0117] It should be noted that the support measure scoring algorithm of the feature-fuzzy set, the certainty measure scoring algorithm of the feature-fuzzy set and the relevance measure scoring algorithm of the feature-fuzzy set can be expressed as follows.

[0118] The support measure scoring algorithm of feature-fuzzy set can be expressed as:

[0119]

[0120] The above formula indicates that when the number of votes for an event record exceeds zero, the event record satisfies (X, P). Deployed to compute The value of is converted to a membership degree through the membership function. The resulting membership degree should be greater than the support measure threshold so that low memberships are discarded, and the levels of the accepted components are aggregated to produce a support measure score.

[0121] The certainty measure scoring algorithm of feature-fuzzy set can be expressed as:

[0122]

[0123] From the above formula, we can see that when different target factors are detected, will change.

[0124] The feature-fuzzy set correlation measurement scoring algorithm can be expressed as:

[0125] in, , , , and can be written as:

[0126]

[0127]

[0128]

[0129]

[0130]

[0131] in, It can be expressed as:

[0132]

[0133] In specific implementation, the method for identifying key inducing factors of typical disasters in the new power system with integrated sky, ground, communication, conduction and sensing proposed in this embodiment establishes a multimodal data fusion system with full coverage of sky, ground information under the full-process risk situation awareness architecture of source-grid-load-storage in the new power system with integrated sky, ground, communication, conduction and sensing. It can analyze the risk factors affecting the safety status of the new power system in a more comprehensive and multi-dimensional manner, thereby giving reasonable suggestions for the safe production of the new power system, and has certain reference significance for building a safe production environment of the new power system with real-time data, intelligent management and control, and reasonable early warning.

[0134] It should be noted that the steps of constructing a key inducing factor identification mechanism based on probability distribution and fuzzy set theory include: constructing a support threshold of a feature-fuzzy set to calculate the frequency of occurrence of rare features in historical disaster records; constructing a certainty threshold of a feature-fuzzy set, and verifying the deterministic association results between rare factors and disaster outcomes in combination with fuzzy weights and membership; constructing a relevance threshold of a feature-fuzzy set and quantifying the strength of the association between rare factors and disasters through expected values ​​and probability distributions; constructing a key inducing factor identification mechanism based on the support threshold of a feature-fuzzy set, the certainty threshold of a feature-fuzzy set and the relevance threshold of a feature-fuzzy set.

[0135] Step S50: Determine the score result according to the parallel evaluation method combined with the key predisposing factor identification mechanism to determine whether there is a high-risk factor.

[0136] In the specific implementation, based on the state thresholds of the three types of feature-fuzzy sets, the numerical probability distribution of the rare factors covered is used as the driving force to describe the corresponding risk distribution characteristics; according to the risk distribution characteristics of the above-mentioned continuous and discrete features, the state fuzzy scoring algorithms of the three corresponding feature-fuzzy sets are designed respectively. Based on the above thresholds and scoring results, the rare high-risk feature factors in the rare feature factor set are further mined. The three state fuzzy scoring algorithms of feature-fuzzy sets are specifically: the support measure scoring algorithm of feature-fuzzy sets, the certainty measure scoring algorithm of feature-fuzzy sets, and the relevance measure scoring algorithm of feature-fuzzy sets.

[0137] This embodiment obtains images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establishes a multimodal data fusion system based on the multi-source data; constructs a typical disaster event database in the multimodal data fusion system; obtains continuous features and discrete features in the typical disaster event database to generate a parallel evaluation method; constructs a key inducing factor identification mechanism based on probability distribution and fuzzy set theory; determines the score results based on the parallel evaluation method combined with the key inducing factor identification mechanism to determine whether there are high-risk factors. By mining high-risk and low-probability elements, a comprehensive consideration of the key inducing factors of typical disasters is achieved, and the disaster warning capability of the power system is improved.

[0138] In addition, an embodiment of the present application also proposes a computer-readable storage medium, on which is stored a program for identifying key inducing factors of typical disasters in an electric power system that integrates the sky and the ground and integrates communications, conductions, and senses. When the program for identifying key inducing factors of typical disasters in an electric power system that integrates the sky and the ground and integrates communications, conductions, and senses is executed by a processor, the steps of the method for identifying key inducing factors of typical disasters in an electric power system that integrates the sky and the ground and integrates communications, conductions, and senses as described above are implemented.

[0139] Reference Figure 5 , Figure 5 This is a structural block diagram of an embodiment of a system for identifying key inducing factors of typical disasters in power systems that integrates the sky, the ground, and the communication, the conduction, and the sensing.

[0140] like Figure 5 As shown, the sky-ground integrated, communication-conducting-sensing integrated power system typical disaster key inducing factor identification system proposed in the embodiment of the present application includes:

[0141] The data acquisition module 10 is used to acquire images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establish a multi-modal data fusion system based on the multi-source data;

[0142] A database construction module 20, used to construct a typical disaster event database through comprehensive perception of the target in a multimodal data fusion system;

[0143] A parallel evaluation module 30, for obtaining continuous features and discrete features in a typical disaster event database to generate a parallel evaluation method;

[0144] An identification mechanism module 40 is used to construct a key inducing factor identification mechanism based on probability distribution and fuzzy set theory;

[0145] The judgment module 50 is used to determine the score result according to the parallel evaluation method combined with the key predisposing factor identification mechanism to judge whether there is a high-risk factor.

[0146] It should be understood that the above is only an example and does not constitute any limitation on the technical solution of the present application. In specific applications, technicians in this field can make settings as needed, and the present application does not impose any limitation on this.

[0147] This embodiment obtains images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establishes a multimodal data fusion system based on the multi-source data; constructs a typical disaster event database in the multimodal data fusion system; obtains continuous features and discrete features in the typical disaster event database to generate a parallel evaluation method; constructs a key inducing factor identification mechanism based on probability distribution and fuzzy set theory; determines the score results based on the parallel evaluation method combined with the key inducing factor identification mechanism to determine whether there are high-risk factors. By mining high-risk and low-probability elements, a comprehensive consideration of the key inducing factors of typical disasters is achieved, and the disaster warning capability of the power system is improved.

[0148] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of the present application. In practical applications, technicians in this field can select part or all of it according to actual needs to achieve the purpose of the present embodiment, and no limitation is made here.

[0149] In addition, for technical details not described in detail in this embodiment, please refer to the method for identifying key inducing factors of typical disasters in the power system with integrated sky-ground and communication-conducting-sensing provided in any embodiment of the present application, which will not be repeated here.

[0150] In addition, it should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0151] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0152] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as a read-only memory (ROM) / RAM, a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of each embodiment of the present application.

[0153] The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for identifying key inducing factors of typical disasters in power systems that integrates air, ground, communication, conduction and sensing, characterized in that: include: Acquire images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establish a multi-modal data fusion system based on the multi-source data; In the multimodal data fusion system, a typical disaster event database is constructed through comprehensive perception of the target; Acquire continuous features and discrete features in the typical disaster event database to generate a parallel evaluation method; Construct the key inducing factor identification mechanism based on probability distribution and fuzzy set theory; Determine the score result according to the parallel evaluation method combined with the key predisposing factor identification mechanism to judge whether there is a high-risk factor; The step of obtaining continuous features and discrete features in the typical disaster event database to generate a parallel evaluation method includes: By continuous feature analysis in the disaster event database, a probability distribution curve is drawn, the value interval is divided into three categories: common, possible and rare, and a membership function is constructed; Conduct discrete feature analysis and define the state thresholds of feature components based on fuzzy set theory; Conduct risk assessment on continuous feature analysis results and discrete feature analysis results separately to generate a parallel assessment method; The steps of constructing a key inducing factor identification mechanism based on probability distribution and fuzzy set theory include: Constructing support thresholds of feature-fuzzy sets to calculate the occurrence frequency of rare features in historical disaster records; Construct the certainty threshold of feature-fuzzy set, and verify the deterministic association results between rare factors and disaster outcomes by combining fuzzy weights and membership. Construct the correlation threshold of feature-fuzzy set and quantify the correlation strength between rare factors and disasters through expected value and probability distribution; A key inducing factor identification mechanism is constructed according to the support threshold of the feature-fuzzy set, the certainty threshold of the feature-fuzzy set and the correlation threshold of the feature-fuzzy set.

2. According to claim 1, a method for identifying key inducing factors of typical disasters in power systems integrating air, ground, communication, conduction and sensing, characterized in that: The step of acquiring images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establishing a multi-modal data fusion system based on the multi-source data includes: Obtain image information, video information and text data information of typical disasters in the power system through the sky-ground network detection module; The image information of typical disasters in power system is homogeneously fused through weighted average formula; Heterogeneously fusing the video information and text data to extract multimodal key features and comprehensively perceive the target; Generate multi-source data according to the homogeneous fusion result and the heterogeneous fusion result; A preset feature screening mechanism is obtained and a multi-source data is used to establish a multi-modal data fusion system.

3. According to claim 2, a method for identifying key inducing factors of typical disasters in power systems integrating air, ground, communication, conduction and sensing is characterized by: The step of performing homogeneous fusion of image information of typical disasters in the power system by using a weighted average formula includes: Pixel-level fusion of image information of typical disasters in power system through weighted average formula in, is the weight coefficient, which is adjusted according to the reliability and relevance of the data source. and are two image sources to be fused, is the fused image.

4. According to claim 1, a method for identifying key inducing factors of typical disasters in power systems integrating air, ground, communication, conduction and sensing is characterized by: The step of constructing a typical disaster event database by comprehensively perceiving the target in the multimodal data fusion system includes: Obtaining disaster event records and corresponding feature information in the multimodal data fusion system and performing unified feature standardization to form a structured database; Disaster event records, characteristic values ​​and target factors are stored in the structured database in a matrix form to construct a structured database.

5. The method for identifying key inducing factors of typical disasters in power systems integrating air, ground, communication, conduction and sensing according to claim 1 is characterized in that: The support threshold of the feature-fuzzy set is expressed as: in, represents a rare data set, t i is the i-th fault record in the evaluation space B, x i,j Is a feature in the fault record The components of and Indicates the total amount of record B and The factor, card() means the total number of qualified records, Th s express The support threshold of The ratio of the number of disaster records of the characteristic to the total number is determined t i (x i,j )represent x i,j exist t i The value of or Describes the fuzzy set The membership degree of is the corresponding fuzzy weight; The certainty threshold of the feature-fuzzy set can be expressed as: In the formula and represent rare factor sets and related fuzzy sets respectively, represent The corresponding fuzzy weights, express The certainty threshold, or Describes the fuzzy set The membership degree of , and the condition for the threshold formula of the certainty of the feature-fuzzy set to be established is: The correlation threshold of the feature-fuzzy set can be expressed as: Where Z=X Y,L=P Q,ME[…] stands for expectation, , , , and , respectively expressed as: in, in, Th co express The correlation threshold.

6. A system for identifying key inducing factors of typical disasters in power systems that integrates air, ground, communication, conduction and sensing, characterized by: The method according to claim 1 is performed, wherein the sky-ground integrated, communication-conducting-sensing integrated power system typical disaster key inducing factor identification system comprises: A data acquisition module, used to acquire images, videos and text data of typical disasters in the power system through the sky-ground network detection module to generate multi-source data, and establish a multi-modal data fusion system based on the multi-source data; A database construction module, used to construct a typical disaster event database through comprehensive perception of the target in the multimodal data fusion system; A parallel evaluation module, used for acquiring continuous features and discrete features in the typical disaster event database to generate a parallel evaluation method; Identification mechanism module, used to construct key inducing factor identification mechanism based on probability distribution and fuzzy set theory; A judgment module is used to determine the score result according to the parallel evaluation method combined with the key inducing factor identification mechanism to determine whether there is a high-risk factor.

7. A computer device, characterized in that: The device comprises: a memory and a processor, wherein the processor executes the method according to any one of claims 1 to 5 when running computer instructions stored in the memory.

8. A computer-readable storage medium, characterized in that: The method comprises instructions, which, when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 5.

Images