Substation geological disaster risk prediction method based on satellite-ground data fusion

By constructing an integrated air-ground substation geological disaster monitoring system, and combining satellite internet and multi-level intelligent integration methods, the problems of accuracy and computational complexity in substation geological disaster risk prediction have been solved, enabling comprehensive and accurate prediction and management of various disasters.

CN118917560AInactive Publication Date: 2024-11-08HUNAN UNIV
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202411412624.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-10-11
Publication Date
2024-11-08
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

Existing methods for predicting geological disaster risks in substations suffer from low accuracy and complex calculation processes. Ground sensors have limited monitoring range, making it difficult to comprehensively collect large-area features, and the complexity of calculations is increased by the presence of multiple disaster impact features.

Method used

Construct an integrated air-ground substation geological disaster monitoring system, combining space-based, air-based, and ground-based platforms for comprehensive perception, utilizing satellite internet to assist data transmission, employing a multi-level adaptive intelligent integration method, and enhancing risk prediction through fuzzy logic reasoning modules.

Benefits of technology

It improves the comprehensiveness and applicability of disaster prediction, enhances the substation's ability to respond to various potential disasters, ensures data coverage and accuracy, and provides an innovative solution for the accurate prediction and management of geological disasters in substations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118917560A_ABST
    Figure CN118917560A_ABST
Patent Text Reader

Abstract

The invention discloses a substation geological disaster risk prediction method based on satellite-ground data fusion. The substation geological disaster risk prediction method comprises the following steps: S1, constructing a sky-ground integrated substation geological disaster monitoring system; a space-based platform, an air-based platform and a foundation platform are adopted to comprehensively perceive the geological environment of the transformer substation in the to-be-predicted area; s2, constructing a satellite internet assisted data transmission network; s3, preprocessing geological disaster influence characteristic data of the transformer substation, and performing contribution degree sorting on the characteristic data; s4, performing clustering analysis on the feature data with high contribution degree, and outputting a membership function; and S5, constructing a fuzzy logic reasoning enhancement module, and outputting the predicted geological disaster risk value of the transformer substation.
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 systems, and in particular to a method for predicting geological disaster risks of substations based on satellite-ground data fusion. Background Art

[0002] As a key hub in the power system, substations undertake the important tasks of receiving, converting and distributing electric energy in the transmission network. Their safety and stability are directly related to the operating efficiency and safety of the entire power grid, as well as the production and life of the majority of users. With the development of smart grids, the construction of substations is accelerating, and more and more problems are coming with it, such as being close to the river, low elevation, and complex and changeable climate, which makes geological disasters frequent. Therefore, geological disaster risk prediction has become one of the main tasks of substation safe operation. Through the geological disaster risk prediction system, potential disaster risks can be known in advance, and preventive measures can be taken, such as strengthening equipment, optimizing facility layout, etc., to reduce the impact of disasters on substations.

[0003] The existing geological disaster early warning methods for electrical systems mainly include scenario simulation methods, while there is little involvement in the prediction of geological disaster risks in substations. The scenario simulation method has certain advantages in revealing system responses and physical mechanisms, but it has significant limitations in terms of system scalability. In other words, the scenario simulation method is often designed for specific geographical areas, disaster types and conditions, and is difficult to directly apply to other different scenarios. At the same time, in order to ensure the safe and stable operation of substations, it is not possible to analyze only one geological disaster, but the impact characteristics of multiple geological disasters should be analyzed simultaneously. However, there are many impact characteristics of geological disasters in substations, and the monitoring range of ground sensor equipment is limited, making it difficult to fully collect the impact characteristics. To this end, some scholars have proposed to monitor the impact characteristics of geological disasters in substations by deploying multiple sensors and drones, and to use machine learning methods to comprehensively analyze the impact characteristics. However, the current substation geological disaster risk prediction methods have the following defects: (1) The lack of collected features makes it difficult to meet the accuracy of risk prediction. Ground sensors have a fixed detection range and sensitivity, which means that they can only monitor environmental conditions near the installation location, and some large-area features such as NDVI cannot be collected. In theory, dense sensors can be deployed to cover the entire area, but this will undoubtedly greatly increase the monitoring cost. In addition, due to network congestion, hardware failure and other limitations, the stability of communication cannot be guaranteed, and data transmission may result in packet loss, which will also cause insufficient feature collection. (2) There are too many influencing features of substation geological disasters, which will greatly increase the complexity of calculation. Substation geological disasters include landslides, collapses, mudslides, etc., each of which has different inducing factors and manifestations. In order to conduct a comprehensive analysis, all the influencing features of each disaster are often considered, which brings considerable trouble to the calculation. Summary of the invention

[0004] In view of this, the present invention provides a method for predicting geological disaster risks of substations based on satellite-ground data fusion, which is used to at least solve the problems of low risk prediction accuracy and complex calculation process in the prior art.

[0005] In order to achieve the above object, the present invention adopts the following technical solution: A method for predicting geological disaster risks of substations based on satellite-ground data fusion includes the following steps: S1. Construct a sky-ground integrated substation geological disaster monitoring system; use space-based platforms, air-based platforms and ground-based platforms to comprehensively perceive the geological environment of substations in the predicted area to obtain risk assessment factors for substation geological disasters; S2. Build a satellite Internet-assisted data transmission network to stably transmit risk assessment factors to the data processing center; S3. Pre-process the risk assessment factors of historical substation geological disasters collected under the sky-ground integrated substation geological disaster monitoring system, and rank the risk assessment factors by contribution; S4. Perform cluster analysis on the risk assessment factors whose contribution is higher than the preset contribution threshold, and output the risk assessment factor membership function; S5. Construct a fuzzy logic reasoning enhancement module to output the predicted substation geological disaster risk value.

[0006] Preferably, the specific content of S1 includes: S11. Determine potential areas of geological hazards through space-based platforms; conduct preliminary scans of the substations and their surrounding areas in the entire predicted area using high-resolution optical imaging and synthetic aperture radar interferometry technology from space-based platforms to identify areas with significant geological deformation in the past or currently undergoing deformation, thereby determining potential areas of geological hazards; S12. Focus on monitoring the potential danger areas through air-based platforms; use the drone aerial photography and airborne LiDAR of the air-based platform to focus on monitoring the surface deformation and damage signs and topography of the potential danger areas. At the same time, with the navigation enhancement assistance provided by the navigation satellite system to the drone, the drone can accurately locate and navigate in the potential danger areas; S13. Use the ground-based platform to conduct accurate and real-time monitoring of the substation geological disaster risk areas that are difficult to monitor via the space-based and air-based platforms.

[0007] Preferably, the specific content of S2 includes: S21. Establish a communication link for the data collected by the air-based platform and the ground-based platform, and transmit the collected data to the ground data processing center; when the ground network is stable, the monitoring data of each sensor of the ground-based platform within the preset range is collected to the data processing center through wireless communication technology; when the ground network is unstable, adopt an adaptive network switching mechanism, automatically switch to the uplink to transmit the data to the satellite Internet, and transmit the data to the ground data processing center through the downlink, establish an uninterrupted information transmission link, and ensure the transmission of geological disaster information of substations; S22. Establish a communication link for the data collected by the space-based platform and transmit the collected data to the ground data processing center; the images collected by the remote sensing satellite and the position information provided by the navigation satellite are transmitted to the ground data processing center through the inter-satellite link and the downlink.

[0008] Preferably, the specific content of S3 includes: S31. Create different decision models based on the risk assessment factors of substation geological disasters collected under the sky-ground integrated monitoring system. Each decision model is trained based on different random samples and feature subsets of the risk assessment factor data set, and calculate the classification impurity of the decision points in each decision model. S32. Analyze the changes in classification impurity before and after the decision to evaluate the classification contribution of each substation geological disaster risk assessment factor in the decision model; S33. Normalize the contribution of all risk assessment factors in the entire decision model set and rank the normalized contribution.

[0009] Preferably, in S31, the classification impurity measurement of the decision point in each decision model, i.e., the Gini coefficient, is calculated Specifically: ;

[0010] in, Indicates A decision model, represents the number of categories, Indicates decision point Medium Category The proportion of Represents possible categories, used to traverse all categories in the category set Elements in S32. Calculate the change in classification impurity before and after the decision, that is, the change in the Gini coefficient before and after the decision point classification, recorded as : ; in, and Represents decision points The Gini index of the left and right sub-decision points after classification, Indicates Risk assessment factors; S33. Normalize the contribution of risk assessment factors in all decision models and rank the contribution of risk assessment factors; the normalization process specifically includes: ; ; ; in, Indicates that it contains Features , represents the total number of decision models, Representation characteristics In the decision model The set of nodes that appear in Risk Assessment Factor In the The contribution of a decision model, Risk Assessment Factor The contribution of all decision models, Risk Assessment Factor Normalized contribution across all decision models.

[0011] Preferably, the specific content of S4 includes: S41. Obtain risk assessment factors with contribution higher than a preset contribution threshold, construct a dynamic clustering optimization module, determine the number of levels of substation geological disaster risk assessment factors, and facilitate the formulation of fuzzy rules; S42. Solve the membership function parameters in the fuzzy processing stage of risk assessment factor data, construct a risk assessment factor grade classification function, and determine the membership of the corresponding value of the risk assessment factor in the substation geological disaster risk assessment factor grade.

[0012] Preferably, the specific content of S42 includes: ; ; ; in, , and Respectively represent the membership values ​​of the fuzzy sets Low, Medium, and High. Each function is between 0 and 1, indicating the degree of belonging to the corresponding fuzzy set. represents the variable that determines the degree of membership in the fuzzy set, represents the set of all possible values, and It is used as the cluster center to define the membership function, which is used to divide the most suitable areas into different categories. is the minimum value of the variable data sample, is the maximum value of the variable data sample.

[0013] Preferably, the specific content of S5 includes: S51. Fuzzify the risk assessment factors and perform fuzzy reasoning on the risk level of geological disasters of the substation according to the fuzzy reasoning rules; S52. Defuzzification processing, output substation geological disaster prediction risk value.

[0014] Preferably, the process of S51 is as follows: fuzzifying the risk assessment factors, constructing the membership functions and fuzzy rules of each fuzzy inference system FIS, FIS constitutes FIST and adopts an incremental structure: the first FIS takes the first two variables with the highest VCD as input, and obtains the output of the first FIS according to the fuzzy inference rules. The second FIS takes the output of the first FIS and the variable with the third highest VCD as input, and so on. The output of a FIS is the output of FIST; the fuzzy reasoning includes: fuzzy rule set and fuzzy operator; ; The fuzzy operator used is the intersection, and its mathematical expression is: ; in, and Respectively represent data For collections A and collection B The membership degree of , Intersection (AND) represents the intersection of fuzzy operators; The defuzzification process in S52 is specifically as follows: ; in, It represents the centroid position of the fuzzy set, which is the predicted risk value of geological disasters in substations.

[0015] It can be seen from the above technical solutions that, compared with the prior art, the present invention discloses a method for predicting substation geological disaster risks based on satellite-ground data fusion, which has the following beneficial effects: The present invention constructs a "sky-ground integrated" substation geological disaster monitoring system, integrates geological data collected by space-based platforms, air-based platforms and ground-based platforms, and uses a satellite Internet-assisted network transmission mechanism to transmit data to a data center. At the same time, a multi-level adaptive intelligent integrated prediction method is adopted, which effectively overcomes the problem of large-scale data collection. The prediction method proposed in the present invention can collect various influencing factors of geological disasters and analyze them in a unified manner, accurately predict the risk of geological disasters in substations, not only improves the comprehensiveness and applicability of disaster prediction, but also enhances the substation's response capabilities to various potential disasters. The present invention proposes a method for actively preventing substation geological disasters under sky-ground coordination, enhances data coverage and accuracy, and provides an innovative solution for accurately predicting and effectively managing substation geological disasters. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 The present invention is a general flow chart of the method for predicting the risk of geological disasters in substations based on satellite-ground data fusion; Figure 2 It is a "sky-ground integrated" substation geological disaster monitoring system; Figure 3 Input membership functions of risk assessment factors for dynamic clustering optimization processing; (a) SG input membership function; (b) DH input membership function; (c) MAR input membership function; (d) RE input membership function; Figure 4 is the output membership function of each level of FIS; Figure 5 Comparison of substation geological disaster risk prediction performance using different data sources and different processing methods; (a) Comparison of prediction performance of three prediction methods; (b) Comparison of prediction performance of two data sources; Figure 6 It is the overall flow chart of the present invention. DETAILED DESCRIPTION

[0018] 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.

[0019] like Figure 1 The process of data collection, transmission, and fusion processing in the actual application of the substation geological disaster risk prediction system is shown in the following figure: Figure 6 The topology of the present invention includes a "sky-ground integrated" substation geological disaster data acquisition module, a satellite Internet-assisted data transmission module, a feature importance ranking module, a dynamic clustering optimization module, and a fuzzy logic reasoning enhancement module.

[0020] The present invention provides a method for predicting the risk of geological disasters in substations based on satellite-ground data fusion, which specifically includes: S1. “Sky-ground integrated” substation geological disaster data collection; Building a "sky-ground integrated" substation geological disaster monitoring system is the data basis for substation geological disaster risk prediction. It uses space-based monitoring platform, air-based monitoring platform and ground-based monitoring platform to comprehensively perceive the geological environment of the substation and obtain comprehensive substation geological disaster risk assessment factors.

[0021] S1.1. Conduct a preliminary large-scale scan of the substations and their surrounding areas in the entire region using high-resolution optical imaging and synthetic aperture radar interferometry (InSAR) technology from a space-based platform to identify areas with significant historical geological deformation or current deformation, thereby determining areas with potential geological disaster risks; S1.2. Use aerial photography and airborne LiDAR from airborne drones to focus on monitoring the signs of surface deformation and damage and topography in the potential danger area. At the same time, with the navigation enhancement assistance provided by navigation satellite systems (such as GPS, GLONASS, Galileo and BeiDou), drones can accurately locate and navigate in the potential danger area. S1.3, through the ground-based platform to monitor the substation geological disaster risk areas that are difficult to monitor by space-based and air-based platforms accurately and in real time: using the ground-based monitoring system, including geological radar, displacement sensors and inclinometers, etc., to observe the geological deformation that is difficult to monitor by space-based and air-based platforms in real time and accurately. In addition, by deploying groundwater level monitoring devices and soil moisture sensors, further assess the development trend of geological disasters, realize the continuous monitoring of the geological stability of the substation and its surrounding key areas, and ensure the effective response of the early warning system and the timely implementation of disaster prevention measures.

[0022] S1.4. Collect geological data around power substations by utilizing the complementary advantages of space-based monitoring platforms, air-based monitoring platforms, and ground-based monitoring platforms. Remote sensing satellites capture large-scale geological features, such as NDVI, and adjust the frequency of data collection according to the level of geological hazard risk. In high-risk areas, the frequency of data collection may be as high as once every 1-2 days, or even once every few hours, while in low-risk areas, it may be once every half a month. On the other hand, ground sensors collect local geological features as needed, and the frequency may range from once every few minutes to once an hour.

[0023] The proposed "sky-ground integrated" substation geological disaster collaborative monitoring system, on the one hand, can be based on the integration of the "communication, navigation and sensing" functions of satellite Internet, and use "sky-ground" monitoring devices to comprehensively collect the disaster-causing factors of geological disasters in the substation and its surrounding areas; on the other hand, it can effectively coordinate resources and propose a resource allocation plan of "comprehensive monitoring of business areas, focused monitoring of hidden danger areas, and in-depth monitoring of key areas", so as to ensure the efficient use of limited resources.

[0024] S2. Satellite Internet-assisted data transmission S2.1. Data collected from air-based and ground-based platforms. First, wireless communication technology (LoRa or NB-IoT) is used to collect monitoring data from multiple sensors within a radius of 5km to the data processing center. Then, an adaptive network switching mechanism is used. When the ground network is unstable, the data is automatically switched to the uplink for transmission to the satellite Internet. Then, the data is transmitted to the ground data processing center via the downlink to establish an uninterrupted information transmission link to ensure the transmission of geological disaster information at the substation. By monitoring the signal strength, it can be determined in real time whether the network needs to be switched: ; in, is the ground network signal strength, is the preset signal strength threshold, signal strength The calculation of is usually based on the received signal power and transmission loss and can be expressed as: ; in is the transmit power, is the transmission loss, which is usually calculated using the free space transmission loss model: ; in, is the transmission distance, is the signal frequency, is the speed of light, about 3×10 8 m / s.

[0025] S2.2. Data collected by space-based platforms: Images collected by remote sensing satellites and location information provided by navigation satellites are transmitted to ground data processing centers through intersatellite links and downlinks.

[0026] Smart substations empowered by satellite Internet will have a stable and reliable data transmission network, capable of transmitting high-quality data in real time, forming a massive and rich substation geological environment database, and supporting planning decisions under the diversified environmental data of smart substations.

[0027] S3. Construct feature importance ranking module The variable contribution degree of each variable is expressed as VCD. VCD is usually measured by the reduction of the Gini index when the risk assessment factor is used as a decision point in all decision models. Assuming that Risk Assessment Factor , The decision-making model normalizes the Gini index score of each risk assessment factor and can be expressed as: ; ; ; in, Indicates that it contains Features , represents the total number of decision models, Representation characteristics In the decision model The set of nodes that appear in Risk Assessment Factor In the The contribution of a decision model, Risk Assessment Factor The contribution of all decision models, Risk Assessment Factor Normalized contribution of all decision models. Calculate the change in Gini coefficient before and after the decision point classification , which can be expressed as: ; in, and Represents decision points Gini index of the left and right sub-decision points after classification. Indicates Tree Node The Gini index, can be expressed as: ; in, Indicates A tree, represents the number of categories, Representation Node Medium Category The proportion of Represents possible categories, used to traverse all categories in the category set Intuitively speaking, it is the elements from the node The probability that two samples are randomly selected from the same category and their category labels are inconsistent.

[0028] S4. Build dynamic clustering optimization module In order to determine whether the value of the substation geological hazard risk assessment factor belongs to the corresponding assessment factor level, a rigorous input membership function needs to be constructed. The mean clustering algorithm is used to determine the parameters of the membership function. Specifically, The mean clustering algorithm is a commonly used unsupervised learning method that discovers the potential structure in the data by dividing the data into multiple levels. In the present invention, the algorithm is used to analyze and process the input data to determine which level of the substation geological hazard risk assessment factor the data point belongs to. In this way, we can calculate the membership for each data point, that is, the possibility that the data point belongs to a certain level of the risk assessment factor. In the present invention, the number of levels of the substation geological hazard risk assessment factor is selected as 3, which are low, medium and high as the representative values ​​of the data points. Among them, the "low" and "high" labels are represented by trapezoidal membership functions; the "medium" label is represented by a triangular membership function: ; ; ; in, , and Respectively represent the membership values ​​of the fuzzy sets Low, Medium, and High. Each function is between 0 and 1, indicating the degree of belonging to the corresponding fuzzy set. represents the variable that determines the degree of membership in the fuzzy set, represents the set of all possible values, It is used as the cluster center to define the membership function, which is used to divide the most suitable areas into different categories. is the minimum value of the variable data sample, is the maximum value of the variable data sample.

[0029] S5. Constructing fuzzy logic reasoning enhancement module S5.1. In this embodiment, 6 Mamdani FISs are deployed in the form of a tree to form FIST. In order to design the FIST risk prediction model, the six FISs are initialized one by one and their membership functions and fuzzy rules are declared. In order to improve the accuracy of the prediction model, the contribution of the input variables is sorted and fuzzified, and then input into FIST layer by layer from high to low according to the contribution of the input variables. In other words, the first FIS takes the first two variables with the highest VCD as input, the second FIS takes the output of the first FIS and the variable with the third highest VCD as input, and so on. The output of the sixth FIS is the output of FIST. Taking a system with seven input variables and one output variable as an example, three membership functions are assigned to each input variable. A simple FIS needs to consider 3 7 = 2187 rules, while the FIST in the present invention only needs to consider 54 rules, which greatly reduces the computational complexity of the evaluation system.

[0030] S5.2. Construct a fuzzy reasoning method that considers the risk assessment factors of substation geological hazards, which includes two parts: a fuzzy rule set and a fuzzy operator. Based on the correlation between different characteristic parameters and the experience reference provided by geological experts, a fuzzy rule set is constructed to convert qualitative expert knowledge into quantitative fuzzy rules. The fuzzy operator used in the present invention is an intersection, and its mathematical expression is: ; in, and Respectively represent data For collections A and collection B The membership degree of , Intersection (AND) represents the intersection of fuzzy operators; S5.3. In the final stage of FIST, the defuzzification method is used to obtain the risk assessment results. The defuzzification process is as follows: ; in, It represents the centroid position of the fuzzy set, which is the predicted risk value of geological disasters in substations.

[0031] The present invention compares the predicted fault risk with the actual fault processing result of the corresponding record in the test set (0 or 1: the fault processing result occurs or does not occur) when the data sources are the sky-ground integrated monitoring data and the ground sensor data, and when the prediction methods are the fuzzy logic reasoning enhancement (FLRE) prediction method, the dynamic clustering optimization with fuzzy logic reasoning enhancement (DCO-FLRE) prediction method and the multi-level adaptive intelligent integration (MLAII) prediction method, so as to evaluate the performance of the proposed prediction model. Figure 5 As shown in the figure, under the "sky-ground integrated" substation geological disaster monitoring system, the MLAII prediction method can effectively improve the prediction effect of substation geological disaster risks.

[0032] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A method for predicting geological disaster risks in substations based on satellite-ground data fusion, characterized in that: The following steps are involved: S1. Construct a sky-ground integrated substation geological disaster monitoring system; use space-based platforms, air-based platforms and ground-based platforms to comprehensively perceive the geological environment of substations in the predicted area to obtain risk assessment factors for substation geological disasters; S2. Build a satellite Internet-assisted data transmission network to stably transmit risk assessment factors to the data processing center; S3. Pre-process the risk assessment factors of historical substation geological disasters collected under the sky-ground integrated substation geological disaster monitoring system, and rank the risk assessment factors by contribution; S4. Perform cluster analysis on the risk assessment factors whose contribution is higher than the preset contribution threshold, and output the risk assessment factor membership function; S5. Construct a fuzzy logic reasoning enhancement module to output the predicted substation geological disaster risk value.

2. The method for predicting substation geological disaster risk based on satellite-ground data fusion according to claim 1 is characterized in that: The specific contents of S1 include: S11. Determine potential areas of geological hazards through space-based platforms; conduct preliminary scans of the substations and their surrounding areas in the entire predicted area using high-resolution optical imaging and synthetic aperture radar interferometry technology from space-based platforms to identify areas with significant geological deformation in the past or currently undergoing deformation, thereby determining potential areas of geological hazards; S12. Focus on monitoring the potential danger areas through air-based platforms; use the drone aerial photography and airborne LiDAR of the air-based platform to focus on monitoring the surface deformation and damage signs and topography of the potential danger areas. At the same time, with the navigation enhancement assistance provided by the navigation satellite system to the drone, the drone can accurately locate and navigate in the potential danger areas; S13. Use the ground-based platform to conduct accurate and real-time monitoring of the substation geological disaster risk areas that are difficult to monitor via the space-based and air-based platforms.

3. The method for predicting substation geological disaster risk based on satellite-ground data fusion according to claim 1 is characterized in that: The specific contents of S2 include: S21. Establish a communication link for the data collected by the air-based platform and the ground-based platform, and transmit the collected data to the ground data processing center; when the ground network is stable, the monitoring data of each sensor of the ground-based platform within the preset range is collected to the data processing center through wireless communication technology; when the ground network is unstable, adopt an adaptive network switching mechanism, automatically switch to the uplink to transmit the data to the satellite Internet, and transmit the data to the ground data processing center through the downlink, establish an uninterrupted information transmission link, and ensure the transmission of geological disaster information of substations; S22. Establish a communication link for the data collected by the space-based platform and transmit the collected data to the ground data processing center; the images collected by the remote sensing satellite and the position information provided by the navigation satellite are transmitted to the ground data processing center through the inter-satellite link and the downlink.

4. The method for predicting substation geological disaster risk based on satellite-ground data fusion according to claim 1 is characterized in that: The specific contents of S3 include: S31. Create different decision models based on the risk assessment factors of substation geological disasters collected under the integrated sky-ground monitoring system. Each decision model is trained based on different random samples and feature subsets of the risk assessment factor data set, and calculate the classification impurity of the decision points in each decision model. S32. Analyze the changes in classification impurity before and after the decision to evaluate the classification contribution of each substation geological disaster risk assessment factor in the decision model; S33. Normalize the contribution of all risk assessment factors in the entire decision model set and rank the normalized contribution.

5. The method for predicting substation geological disaster risk based on satellite-ground data fusion according to claim 4 is characterized in that: In S31, the classification impurity of the decision point in each decision model is calculated, that is, the Gini coefficient, which is recorded as : ; in, Indicates A decision model, represents the number of categories, Indicates decision point Medium Category The proportion of Represents possible categories, used to traverse all categories in the category set Elements in S32. Calculate the change in classification impurity before and after the decision, that is, the change in the Gini coefficient before and after the decision point classification, recorded as : ; in, and Represents decision points The Gini index of the left and right sub-decision points after classification, Indicates Risk assessment factors; S33. Normalize the contribution of risk assessment factors in all decision models and rank the contribution of risk assessment factors; the normalization process specifically includes: ; ; ; in, Indicates that it contains Features , represents the total number of decision models, Representation characteristics In the decision model The set of nodes that appear in Risk Assessment Factor In the The contribution of a decision model, Risk Assessment Factor The contribution of all decision models, Risk Assessment Factor Normalized contribution across all decision models.

6. The method for predicting substation geological disaster risk based on satellite-ground data fusion according to claim 1 is characterized in that: The specific contents of S4 include: S41. Obtain risk assessment factors with contribution higher than a preset contribution threshold, construct a dynamic clustering optimization module, determine the number of levels of substation geological disaster risk assessment factors, and facilitate the formulation of fuzzy rules; S42. Solve the membership function parameters in the fuzzy processing stage of risk assessment factor data, construct a risk assessment factor grade classification function, and determine the membership of the corresponding value of the risk assessment factor in the substation geological disaster risk assessment factor grade.

7. The method for predicting substation geological disaster risk based on satellite-ground data fusion according to claim 6 is characterized in that: The specific contents of S42 include: ; ; ; in, , and Respectively represent the membership values ​​of the fuzzy sets Low, Medium, and High. Each function is between 0 and 1, indicating the degree of belonging to the corresponding fuzzy set. represents the variable that determines the degree of membership in the fuzzy set, represents the set of all possible values, and It is used as the cluster center to define the membership function, which is used to divide the most suitable areas into different categories. is the minimum value of the variable data sample, is the maximum value of the variable data sample.

8. The method for predicting substation geological disaster risk based on satellite-ground data fusion according to claim 1 is characterized in that: The specific contents of S5 include: S51. Fuzzify the risk assessment factors and perform fuzzy reasoning on the risk level of geological disasters of substations according to fuzzy reasoning rules; S52. Defuzzification processing, output substation geological disaster prediction risk value.

9. The method for predicting substation geological disaster risk based on satellite-ground data fusion according to claim 8, characterized in that: The specific process of S51 is: fuzzify the risk assessment factors, construct the membership function and fuzzy rules of each fuzzy inference system FIS, FIS constitutes FIST and adopts an incremental structure: the first FIS takes the first two variables with the highest VCD as input, and obtains the output of the first FIS according to the fuzzy inference rules. The second FIS takes the output of the first FIS and the variable with the third highest VCD as input, and so on. The output of each FIS is the output of FIST; Fuzzy reasoning includes: fuzzy rule set and fuzzy operator; ; The fuzzy operator used is the intersection, and its mathematical expression is: ; in, and Respectively represent data For collections A and collection B The membership degree of , Intersection (AND) represents the intersection of fuzzy operators; The defuzzification process in S52 is specifically as follows: ; in, It represents the centroid position of the fuzzy set, which is the predicted risk value of geological disasters in substations.

Citation Information

Patent Citations

  • Geological disaster early warning system based on radar remote sensing technology

    CN116580532A