Geological disaster real-time early warning method based on multi-source data dynamic fusion

Abstract

The embodiment of the invention provides a geological disaster real-time early warning method based on multi-source data dynamic fusion. Comprising the steps that physical parameters of various monitoring data capable of being acquired in a monitoring area are determined and preprocessed; dividing early warning levels, and determining interval values of different early warning levels corresponding to each type of monitoring data; calculating a cloud model parameter value of each early warning level of each type of monitoring data; calculating the credibility of each early warning level corresponding to each type of monitoring data based on the parameter values; based on the credibility distribution, calculating the consistent quantity, the conflict quantity, the conflict intensity and the unreliability among the multi-source monitoring data; constructing an early warning level credibility redistribution equation, and obtaining new credibility of each early warning level corresponding to each piece of monitoring data; and fusing the new credibility of the early warning level, obtaining the credibility of the early warning level based on multi-source data fusion, and determining early warning information and dynamically updating an early warning result in combination with a maximum credibility principle. According to the method, the problem of multi-source data conflict fusion is solved, and the geological disaster early warning efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of disaster early warning technology, and in particular to an instant early warning method for geological disasters based on dynamic fusion of multi-source data. Background Art

[0002] Faced with the severe geological disaster situation, carrying out geological disaster monitoring and early warning is an important way to effectively reduce disaster losses. In recent years, the country has attached great importance to geological disaster monitoring and early warning. Driven by high-tech, a multi-dimensional integrated monitoring and early warning system of "sky-air-ground" has been formed. Based on the monitoring and early warning system, various provinces and cities are gradually deploying various monitoring and early warning terminals and corresponding systems in high-risk areas of geological disasters, which provides important guarantees for timely capture of disaster information and rapid and efficient early warning and risk avoidance.

[0003] However, although a large number of monitoring terminals and diversified monitoring methods provide data support for early warning, the physical meanings, data expressions and data structures mapped by monitoring data from different sources are different. What's more, the degree of danger represented by multiple types of monitoring data for the same disaster body is contradictory and conflicting. This leads to a large amount of multi-source monitoring data just lying statically in the database, lacking sufficient mining and utilization, resulting in low accuracy of early warning results. Summary of the invention

[0004] The purpose of the embodiment of the present invention is to provide a geological disaster instant warning method based on dynamic fusion of multi-source data, so as to achieve effective fusion of multi-source conflicting data and improve the accuracy of geological disaster warning results. The specific technical solution is as follows:

[0005] In a first aspect of an embodiment of the present invention, a method for real-time early warning of geological disasters based on dynamic fusion of multi-source data is first provided, the method comprising:

[0006] S1. Determine the monitoring data type and physical parameters of the geological disaster risk point, wherein the risk point monitoring data is multiple types of data from multiple sensors, and the multiple types of monitoring data are real-time monitoring data collected at different times;

[0007] S2. Preprocess the multi-source monitoring data to eliminate missing values ​​and illogical values;

[0008] S3. Based on the actual warning needs, the warning levels are divided and the interval values ​​of different warning levels corresponding to each type of monitoring data are determined;

[0009] S4. Based on the cloud model theory, calculate the cloud model parameter values ​​of different warning levels corresponding to each type of monitoring data, wherein the cloud model parameter values ​​include: expected value, entropy value and super entropy;

[0010] S5. Based on the cloud model parameter value and the interval values ​​of the different warning levels, calculate the credibility of the different warning levels corresponding to various types of monitoring data, and normalize the calculated credibility;

[0011] S6. Based on the normalized credibility distribution, calculate the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data;

[0012] S7, based on the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data, construct an early warning level credibility redistribution equation, correct the conflict of the multi-source monitoring data, and obtain a new credibility of the early warning level;

[0013] S8. Using evidence theory, the new credibility of different warning levels corresponding to multi-source monitoring data is integrated to obtain the credibility distribution of warning levels based on multi-source data fusion;

[0014] S9. Determine the warning level based on the maximum credibility value in the credibility distribution, and release warning information through the warning system in combination with warning rules.

[0015] In a possible implementation manner, the potential hazard point monitoring data information includes: the type of monitoring equipment, the monitored physical parameters, the format of the monitoring data, and the physical meaning represented by the monitoring data.

[0016] In a possible implementation, the division of warning levels based on actual warning requirements and determination of interval values ​​of different warning levels corresponding to each type of monitoring data include:

[0017] Select the corresponding warning level according to the actual situation of the monitoring area;

[0018] Determine the relevant parameters for early warning classification levels;

[0019] According to the relevant parameters of the warning classification level, the interval values ​​of different warning levels corresponding to each type of monitoring data are determined.

[0020] In a possible implementation, the cloud model parameter values ​​corresponding to different warning levels for each type of monitoring data are calculated based on the cloud model theory, including:

[0021] Based on the interval values ​​of each warning level and in accordance with cloud model theory, on the assumption that the monitoring data corresponding to different warning levels belong to a normal distribution, the cloud model parameter values ​​corresponding to different warning levels of various types of monitoring data are calculated.

[0022] In a possible implementation, the calculation of cloud model parameter values ​​corresponding to different warning levels of various types of monitoring data includes:

[0023] By the first preset formula:

[0024] ;

[0025] ;

[0026] ;

[0027] Calculate the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, among which, For the The monitoring data of this category correspond to Expected value of each warning level; For the The monitoring data of this category correspond to Entropy value of warning level; For the The monitoring data of this category correspond to The super entropy value of the warning level; The value range is 0 to Constants of; and Respectively The monitoring data of this category correspond to The lower and upper limits of the warning level range.

[0028] In a possible implementation, the calculation of the credibility of different warning levels corresponding to various types of monitoring data based on cloud model parameters includes:

[0029] Based on the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, the credibility of each warning level corresponding to each monitoring data is calculated.

[0030] In a possible implementation, the calculation of the credibility of each warning level corresponding to each type of monitoring data based on the cloud model parameter values ​​of different warning levels corresponding to each type of monitoring data includes:

[0031] Based on the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, the second preset formula is used:

[0032] ;

[0033] Calculate the credibility of each warning level corresponding to each type of monitoring data, where i=1,2…n; j=1,2…5; t=1,2…m; Indicates The monitoring data of this category is Monitoring data values ​​within a time period The corresponding warning level is The credibility value of is a random variable.

[0034] In a possible implementation, the normalizing the calculated credibility includes:

[0035] Through the third preset formula:

[0036] ;

[0037] Normalize the calculated credibility to obtain the normalized credibility vector , They respectively represent the credibility values ​​of multiple types of monitoring data corresponding to different warning levels.

[0038] In a possible implementation, the calculation of the consistency, conflict, conflict intensity and unreliability among multi-source monitoring data based on the normalized credibility distribution includes:

[0039] According to the normalized credibility, through the fourth preset formula:

[0040] ;

[0041] ;

[0042] ;

[0043] ;

[0044] Calculate the consistency and conflict between multi-source monitoring data, where: Indicates monitoring data and The amount of agreement between Indicates monitoring data and The amount of conflict between Indicates monitoring data The consistency between the monitoring data of other monitoring devices in the tth time period; Indicates monitoring data The amount of conflict between the monitoring data of other monitoring devices in the tth time period; is the default value;

[0045] According to the normalized credibility, through the fifth preset formula:

[0046] ;

[0047] Calculate the conflict intensity between multi-source monitoring data, where: Indicates monitoring data The conflict intensity between the monitoring data of other monitoring devices in the tth time period;

[0048] According to the normalized credibility, through the sixth preset formula:

[0049] ;

[0050] ;

[0051] Calculate the unreliability between multi-source monitoring data, where: Indicates monitoring data and The correlation coefficient between Indicates monitoring data Unreliability;

[0052] Based on the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data, an early warning level credibility redistribution equation is constructed to correct the conflict of the multi-source monitoring data, including:

[0053] Through the seventh preset formula:

[0054] ;

[0055] The conflict of multi-source monitoring data was corrected for the first time, including: Indicates monitoring data Corresponding warning level after correction The credibility value of Θ represents the set of different warning levels. , Indicates monitoring data The original warning levels The partial value is transferred to the value of the whole set Θ; is the adjustment coefficient, when the monitoring data can be determined When the credibility values ​​of the corresponding warning levels are reasonable, The value is 0, when the monitoring data cannot be determined Whether the credibility values ​​of the corresponding warning levels are reasonable The value of is 0.5;

[0056] Through the eighth preset formula:

[0057] ;

[0058] ;

[0059] ;

[0060] The first correction result is corrected for the second time to obtain the corrected monitoring data information of various types of hidden danger points, wherein the and Respectively represent monitoring data Information entropy and precision; Indicates monitoring data The final revised credibility value corresponding to each warning level;

[0061] The above-mentioned use of evidence theory integrates the new credibility of different warning levels corresponding to multi-source monitoring data to obtain the warning level credibility distribution based on multi-source data fusion, including:

[0062] Through the ninth preset formula:

[0063] ;

[0064] ;

[0065] ;

[0066] By integrating multi-source monitoring data, the credibility distribution of warning levels based on multi-source data fusion is obtained, where: express The monitoring data is in Fusion results within the time period; Indicates that the warning level in the fusion result of multi-source monitoring data is The credibility value of express The conflict coefficient between the monitoring data.

[0067] In a possible implementation, in the step of determining the warning level based on the maximum credibility value in the credibility distribution, and issuing warning information through the warning system in combination with the warning rule, the method further includes:

[0068] Acquire and dynamically integrate multi-source monitoring data from each time period based on S5 to S9 based on monitoring data collected at multiple different times, and dynamically update geological disaster warning information at the monitoring points in real time.

[0069] Another aspect of the embodiment of the present invention provides a geological disaster instant warning device based on dynamic fusion of multi-source data, the device comprising:

[0070] A data acquisition module is used to determine the type and physical parameters of the monitoring data of the geological hazard point, wherein the monitoring data of the hazard point is multiple types of data from multiple sensors, and the multiple types of monitoring data are real-time monitoring data collected at different times;

[0071] A preprocessing module, used to preprocess the multi-source monitoring data to eliminate missing values ​​and illogical values;

[0072] The warning classification module is used to classify the warning levels based on the actual warning needs and determine the interval values ​​of different warning levels corresponding to each type of monitoring data;

[0073] A credibility calculation module is used to calculate the cloud model parameter values ​​of different warning levels corresponding to each type of monitoring data based on the cloud model theory, wherein the cloud model parameter values ​​include: expected value, entropy value and super entropy; based on the cloud model parameter values ​​and the interval values ​​of the different warning levels, calculate the credibility of different warning levels corresponding to each type of monitoring data, and normalize the calculated credibility;

[0074] The conflict calculation module is used to calculate the consistency, conflict, conflict intensity and unreliability among multi-source monitoring data based on the normalized credibility distribution;

[0075] A data correction module is used to construct a warning level credibility redistribution equation based on the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data, correct the conflict of the multi-source monitoring data, and obtain a new credibility of the warning level;

[0076] The data fusion module is used to use the evidence theory to fuse the new credibility of different warning levels of multi-source monitoring data to obtain the credibility distribution of warning levels based on multi-source data fusion;

[0077] The warning release module is used to determine the warning level based on the maximum credibility value in the credibility distribution, release warning information through the warning system in combination with the warning rules, and dynamically update the warning results based on the dynamic fusion of multi-source data collected at each moment.

[0078] In a possible implementation manner, the potential hazard point monitoring data information includes: the type of monitoring equipment, the monitored physical parameters, the format of the monitoring data, and the physical meaning represented by the monitoring data.

[0079] In a possible implementation, the warning classification module is specifically used to select a corresponding warning classification level according to the actual status of the monitoring area; determine relevant parameters of the warning classification level; and determine the interval values ​​of different warning levels corresponding to each type of monitoring data based on the relevant parameters of the warning classification level.

[0080] In a possible implementation, the credibility calculation module is specifically used to calculate the cloud model parameter values ​​corresponding to different warning levels of various types of monitoring data based on the interval values ​​of each warning level and in accordance with cloud model theory, on the assumption that the different warning levels corresponding to the monitoring data belong to a normal distribution.

[0081] In a possible implementation manner, the credibility calculation module is specifically configured to calculate the credibility through a first preset formula:

[0082] ;

[0083] ;

[0084] ;

[0085] Calculate the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, among which, For the The monitoring data of this category correspond to Expected value of each warning level; For the The monitoring data of this category correspond to Entropy value of warning level; For the The monitoring data of this category correspond to The super entropy value of the warning level; The value range is 0 to Constants of; and Respectively The monitoring data of this category correspond to The lower and upper limits of the warning level range.

[0086] In a possible implementation, the calculation of the credibility of different warning levels corresponding to various types of monitoring data based on cloud model parameters includes:

[0087] Based on the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, the credibility of each warning level corresponding to each monitoring data is calculated.

[0088] In a possible implementation, the calculation of the credibility of each warning level corresponding to each type of monitoring data based on the cloud model parameter values ​​of different warning levels corresponding to each type of monitoring data includes:

[0089] Based on the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, the second preset formula is used:

[0090] ;

[0091] Calculate the credibility of each warning level corresponding to each type of monitoring data, where i=1,2…n; j=1,2…5; t=1,2…m; Indicates The monitoring data of this category is Monitoring data within a time period The corresponding warning level is The credibility value of is a random variable.

[0092] In a possible implementation, the normalizing the calculated credibility includes:

[0093] Through the third preset formula:

[0094] ;

[0095] Normalize the calculated credibility to obtain the normalized credibility vector , They respectively represent the credibility values ​​of multiple types of monitoring data corresponding to different warning levels.

[0096] In a possible implementation manner, the conflict calculation module is specifically configured to calculate the conflict according to the normalized credibility through a fourth preset formula:

[0097] ;

[0098] ;

[0099] ;

[0100] ;

[0101] Calculate the consistency and conflict between multi-source monitoring data, where: Indicates monitoring data and The amount of agreement between Indicates monitoring data and The amount of conflict between Indicates monitoring data With other monitoring equipment in The amount of consistency between monitoring data within a time period; Indicates monitoring data With other monitoring equipment in The amount of conflict between monitoring data within a time period; is the default value;

[0102] According to the normalized credibility, through the fifth preset formula:

[0103] ;

[0104] Calculate the conflict intensity between multi-source monitoring data, where: Indicates monitoring data With other monitoring equipment in The conflict intensity between monitoring data within the time period;

[0105] According to the normalized credibility, through the sixth preset formula:

[0106] ;

[0107] ;

[0108] Calculate the unreliability between multi-source monitoring data, where: Indicates monitoring data and The correlation coefficient between Indicates monitoring data Unreliability;

[0109] The data correction module is specifically used to correct the consistency, conflict, conflict intensity and unreliability of the multi-source monitoring data through the seventh preset formula:

[0110] ;

[0111] The conflict of multi-source monitoring data was corrected for the first time, including: Indicates monitoring data Corresponding warning level after correction The credibility value of Θ represents the set of different warning levels. , Indicates monitoring data The original warning levels The partial value is transferred to the value of the whole set Θ; is the adjustment coefficient, when the monitoring data can be determined When the credibility values ​​of the corresponding warning levels are reasonable, The value is 0, when the monitoring data cannot be determined Whether the credibility values ​​of the corresponding warning levels are reasonable The value of is 0.5;

[0112] Through the eighth preset formula:

[0113] ;

[0114] ;

[0115] ;

[0116] The first correction result is corrected for the second time to obtain the corrected monitoring data information of various types of hidden danger points, wherein the and Respectively represent monitoring data Information entropy and precision; Indicates monitoring data The final revised credibility value corresponding to each warning level;

[0117] The data fusion module is specifically used to pass the ninth preset formula:

[0118] ;

[0119] ;

[0120] ;

[0121] By integrating multi-source monitoring data, the credibility distribution of warning levels based on multi-source data fusion is obtained, where: express The monitoring data is in Fusion results within the time period; Indicates that the warning level in the fusion result of multi-source monitoring data is The credibility value of express The conflict coefficient between the monitoring data.

[0122] In a possible implementation, the device further includes:

[0123] The real-time update module is used to obtain and dynamically integrate multi-source monitoring data at each moment based on monitoring data collected at multiple different moments, and update the warning results in real time.

[0124] An embodiment of the present invention further provides an electronic device, comprising a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus;

[0125] Memory, used to store computer programs;

[0126] The processor is used to implement any of the above-mentioned methods for real-time early warning of geological disasters based on dynamic fusion of multi-source data when executing the program stored in the memory.

[0127] An embodiment of the present invention also provides a computer-readable storage medium, in which a computer program is stored. When the computer program is executed by a processor, any of the above-mentioned methods for real-time warning of geological disasters based on dynamic fusion of multi-source data is implemented.

[0128] An embodiment of the present invention also provides a computer program product containing instructions, which, when executed on a computer, enables the computer to execute any of the above-mentioned methods for real-time early warning of geological disasters based on dynamic fusion of multi-source data.

[0129] Beneficial effects of the embodiments of the present invention:

[0130] The embodiment of the present invention provides a geological disaster instant warning method based on dynamic fusion of multi-source data. The main steps of the method include: clarifying the physical parameters of various types of monitoring data that can be obtained in the monitoring area, pre-processing the multi-source monitoring data, and dividing the warning level based on the warning needs of the monitoring area, and determining the interval values ​​of different warning levels corresponding to each type of monitoring data; calculating the cloud model parameter values ​​of different warning levels for each type of monitoring data; calculating the credibility of each warning level corresponding to each type of monitoring data according to the cloud model parameter values; calculating the consistency, conflict, conflict intensity and unreliability between the multi-source monitoring data based on the credibility distribution; constructing a warning level credibility redistribution equation to solve the conflict problem of multi-source monitoring data fusion, and obtaining the new credibility of each warning level corresponding to each monitoring data; using the fusion rule to fuse the new credibility of the warning level to obtain the warning level credibility distribution based on the fusion of multi-source data; determining the warning information in the current time period based on the maximum credibility principle; dynamically and real-time fusing multi-source monitoring data to instantly warn of geological disaster risks. The embodiment of the present invention reduces the conflict between data and increases the compatibility between data by mapping the credibility distribution of monitoring data and warning levels and correcting the conflict of multi-source monitoring data. It utilizes multi-source data fusion rules and combines real-time collected monitoring data to achieve full fusion of multi-source monitoring data and immediate warning of geological disasters, solves the problem of efficient fusion of multi-source conflicting data, and improves the accuracy of geological disaster warnings.

[0131] Of course, it is not necessary to achieve all of the advantages described above at the same time to implement any product or method of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0132] 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 in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other embodiments can also be obtained based on these drawings.

[0133] Figure 1 A schematic flow chart of a method for real-time early warning of geological disasters based on dynamic fusion of multi-source data provided by an embodiment of the present invention;

[0134] Figure 2 A schematic diagram of a conflict detection process provided by an embodiment of the present invention;

[0135] Figure 3 A schematic diagram of a process of data fusion provided by an embodiment of the present invention;

[0136] Figure 4 An example diagram of a geological disaster instant warning method based on dynamic fusion of multi-source data provided by an embodiment of the present invention;

[0137] Figure 5 A schematic diagram of the structure of a geological disaster instant warning device based on dynamic fusion of multi-source data provided by an embodiment of the present invention;

[0138] Figure 6 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0139] 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 based on the present invention belong to the scope of protection of the present invention.

[0140] In view of the uncertainty of geological disasters, in order to reduce the harm caused by the suddenness of disasters, it is necessary to conduct dynamic and immediate warning of the whole process of disaster-causing and harm-causing bodies. Although dynamic warning has been achieved in some areas, the update of warning results is lagging and it is difficult to meet the timeliness requirements. How to improve the accuracy and timeliness of geological disaster warning results has become an important research topic in the field of disaster prevention and mitigation. At present, there are two main methods for geological disaster warning: (1) empirical model method, based on historical disaster data, an evaluation index system is established, and an empirical warning model is constructed using statistical analysis methods. The collected monitoring data is input into the trained model, and then the warning information is output. This method has poor universality. Some models cannot express the uncertainty of the warning results. The accuracy of the warning results also varies from place to place. The overall warning accuracy is not high, and the dynamic update of the warning results is lagging; (2) multi-factor comprehensive method, based on historical disaster data, the warning thresholds of different indicators are determined. According to the field data collected in real time, the multi-source data is comprehensively analyzed and compared with the warning thresholds to give the geological disaster warning level. This method has good timeliness, but cannot scientifically express the uncertainty of the warning results, and lacks the full integration of multi-source monitoring data and the effective processing of multi-source conflicting data, resulting in a waste of data resources and restricting the improvement of the accuracy of the warning results.

[0141] In order to improve the accuracy of geological disaster warning results, an embodiment of the present invention provides a geological disaster instant warning method based on dynamic fusion of multi-source data.

[0142] In a first aspect of the embodiment of the present invention, a geological disaster instant warning method based on dynamic fusion of multi-source data is first provided. Figure 1 , Figure 1 A schematic flow chart of a method for real-time early warning of geological disasters based on dynamic fusion of multi-source data provided by an embodiment of the present invention, the method comprising:

[0143] Step S1, determining the monitoring data type and physical parameters of the geological disaster risk point;

[0144] Step S2, preprocessing the multi-source monitoring data to eliminate missing values ​​and illogical values;

[0145] Step S3: based on actual warning requirements, divide the warning levels and determine the interval values ​​of different warning levels corresponding to each type of monitoring data;

[0146] Step S4: based on the cloud model theory, calculating the cloud model parameter values ​​of different warning levels corresponding to each type of monitoring data, wherein the cloud model parameter values ​​include: expected value, entropy value and super entropy;

[0147] Step S5: based on the cloud model parameter value and the interval values ​​of the different warning levels, calculate the credibility of the different warning levels corresponding to various types of monitoring data, and normalize the calculated credibility;

[0148] Step S6: Based on the normalized credibility distribution, calculate the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data;

[0149] Step S7: constructing a warning level credibility redistribution equation based on the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data, correcting the conflict of the multi-source monitoring data, and obtaining a new credibility of the warning level;

[0150] Step S8: Using evidence theory, fuse the new credibility of different warning levels corresponding to multi-source monitoring data to obtain the warning level credibility distribution based on multi-source data fusion;

[0151] Step S9: determine the warning level based on the maximum credibility value in the credibility distribution, and release warning information through the warning system in combination with the warning rules.

[0152] Corresponding to the above-mentioned step S1, wherein the hidden danger point monitoring data is multiple types of data from multiple sensors, and the multiple types of monitoring data are real-time monitoring data collected at different times. The geological disaster hidden danger point can be multiple points in a pre-set area, specifically, it can be a monitoring point in a hillside, an administrative area, etc. When determining the monitoring data type and physical parameters of the geological disaster hidden danger point, the type of monitoring equipment, the monitored physical parameters, the format of the monitoring data, and the physical meaning represented by the monitoring data can be clarified. Specifically, the monitoring data type may include displacement, rainfall, soil moisture content, etc. In a possible implementation, the hidden danger point monitoring data information includes: the type of monitoring equipment, the monitored physical parameters, the format of the monitoring data, and the physical meaning represented by the monitoring data.

[0153] Corresponding to the above step S2, the multi-source monitoring data is preprocessed to eliminate missing values ​​and illogical values, and prevent missing values ​​and illogical values ​​from appearing, so as to facilitate subsequent processing. Specifically, the preprocessing method can refer to the prior art.

[0154] Corresponding to the above step S3, based on the actual warning needs, the warning level is divided, and the interval value of the different warning levels corresponding to each type of monitoring data is determined. The warning level can be set in advance according to the actual situation. my country's warning classification level often adopts three levels, four levels and five levels. In the embodiment of the present invention, the corresponding warning classification level can be freely selected according to the actual status of the geological disaster monitoring area, such as three or five levels. In an example, see Table 1. The geological disaster in the embodiment of the present invention can be a debris flow, a landslide and other geological disasters.

[0155] Table 1 is a table of parameter information of warning classification levels provided by an embodiment of the present invention

[0156] When determining the interval values ​​of different warning levels corresponding to each type of monitoring data, please refer to Table 2.

[0157] Table 2 is a numerical range table of warning levels provided by an embodiment of the present invention

[0158] in, Indicates The monitoring data of this category correspond to The lower limit of the level warning interval; Indicates The corresponding monitoring data The upper limit of the warning interval value, .

[0159] In a possible implementation, the warning levels are divided based on actual warning needs, and the interval values ​​of different warning levels corresponding to each type of monitoring data are determined, including: selecting a corresponding warning classification level according to the actual status of the monitoring area; determining relevant parameters of the warning classification level, wherein the warning classification level is divided into 5 levels; and determining the interval values ​​of different warning levels corresponding to each type of monitoring data according to the relevant parameters of the warning classification level.

[0160] Corresponding to the above-mentioned step S4, wherein the cloud model parameter values ​​include: expected value, entropy value and super entropy. Based on cloud model theory, the cloud model parameter values ​​corresponding to different warning levels of each type of monitoring data are calculated. In a possible implementation, the cloud model parameter values ​​corresponding to different warning levels of each type of monitoring data are calculated based on cloud model theory, including: based on the interval values ​​of each warning level, according to cloud model theory, on the assumption that the different warning levels corresponding to the monitoring data belong to a normal distribution, the cloud model parameter values ​​corresponding to different warning levels of each type of monitoring data are calculated.

[0161] Corresponding to the above step S5, based on the cloud model parameter value and the interval value of the different warning levels, the credibility of each type of monitoring data corresponding to the different warning levels is calculated, and the calculated credibility is normalized. In a possible implementation, the credibility of each type of monitoring data corresponding to the different warning levels is calculated based on the cloud model parameters, including: based on the cloud model parameter values ​​of the different warning levels corresponding to the various types of monitoring data, the credibility of each warning level corresponding to the various types of monitoring data is calculated.

[0162] Corresponding to the above step S6, based on the normalized credibility distribution, the consistency, conflict, conflict intensity and unreliability between the multi-source monitoring data are calculated. Since in the same time and space, the credibility values ​​of the various warning levels mapped by different types of monitoring data to the geological disaster risk points are different, and even there are serious contradictions and conflicts, which is the direct reason why the conflicting data cannot be effectively integrated. In the embodiment of the present application, the method in Liu Youfang's paper can be used to calculate the consistency, conflict, conflict intensity and unreliability between the monitoring data, providing a reference for the correction of conflicting data.

[0163] Corresponding to the above step S7, based on the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data, an early warning level credibility redistribution equation is constructed to correct the conflict of the multi-source monitoring data and obtain a new credibility of the early warning level. In order to effectively handle the contradictory data among the multi-source monitoring data, an early warning level credibility redistribution equation is constructed to correct the conflict of the multi-source monitoring data. For details, see Figure 2 , construct the warning level credibility redistribution equation, realize the conflict correction of multi-source monitoring data, thereby reducing the conflict between data and increasing the compatibility between data. Through the warning level credibility distribution mapped by various monitoring data, determine the consistency measurement between data, the conflict measurement between data, the data unreliability measurement, the data conflict intensity measurement, the information entropy and the accuracy calculation, and then adjust the system, make the first correction, construct the warning level credibility value redistribution equation, make the second correction, and correct the warning level credibility redistribution correction. Finally, the conflict is reduced and the compatibility is increased, realizing the conflict correction of multi-source monitoring data.

[0164] Corresponding to the above step S8, the new credibility of different warning levels corresponding to multi-source monitoring data is integrated by using evidence theory to obtain the warning level credibility distribution based on multi-source data fusion. Specifically, DS evidence theory (d-sevidence theory) can be used, which is an uncertainty reasoning method. This method can be used to integrate the new credibility of different warning levels of multi-source monitoring data to obtain the warning level credibility distribution based on multi-source data fusion. In an example, see Figure 3 , the corrected monitoring data information of various types of hidden danger points can be integrated, and the multi-source monitoring data can be integrated based on the DS evidence theory fusion formula to obtain the credibility distribution of the warning level based on multi-source data fusion. By monitoring data source 1, monitoring data source 2... monitoring data source i... monitoring data source n, the credibility distribution of data source 1 in time period t is determined , Credibility distribution of data source 2 in time period t , credibility distribution of data source i in time period t , credibility distribution of data source n in time period t , after correction, we get the corrected credibility distribution , Corrected Credibility Distribution , Corrected Credibility Distribution , Corrected Credibility Distribution ,Through DS evidence theory fusion rules, the credibility distribution of multi-source monitoring data fusion is determined, and the warning level is determined by selecting the maximum credibility.

[0165] Corresponding to the above step S9, based on the maximum credibility value in the credibility distribution, the warning level is determined, and in combination with the warning rules, the warning information is issued through the warning system. In a possible implementation, in the determination of the warning level based on the maximum credibility value in the credibility distribution, and in combination with the warning rules, the warning information is issued through the warning system, the method further includes: repeating the above steps according to the multi-source monitoring data collected at each moment, dynamically integrating the multi-source monitoring data at each moment, dynamically updating the warning information, and realizing the real-time warning of geological disaster risks.

[0166] It can be seen that through the method of the present invention, the conflict between data can be reduced and the compatibility between data can be increased by mapping the credibility distribution of monitoring data and warning levels and correcting the conflict of multi-source monitoring data. By utilizing the multi-source data fusion rules and combining the real-time collected monitoring data, the full fusion of multi-source monitoring data and the immediate warning of geological disasters are achieved, the problem of efficient fusion of multi-source conflicting data is solved, and the accuracy of geological disaster warnings is improved.

[0167] In the embodiment of the present invention, based on the cloud model theory, on the assumption that the different warning levels corresponding to the monitoring data belong to the normal distribution, the cloud model parameters corresponding to the different warning levels of various types of monitoring data are calculated to provide a basic preparation for the uncertainty expression of the warning level. In a possible implementation, the calculation of the cloud model parameter values ​​corresponding to the different warning levels of various types of monitoring data includes:

[0168] By the first preset formula:

[0169] ;

[0170] ;

[0171] ;

[0172] Calculate the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, among which, For the The monitoring data of this category correspond to Expected value of each warning level; For the The monitoring data of this category correspond to Entropy value of warning level; For the The monitoring data of this category correspond to The super entropy value of the warning level; The value range is 0 to Constants of; and Respectively The monitoring data of this category correspond to The lower and upper limits of the warning level range.

[0173] In the embodiment of the present invention, based on the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, The monitoring data from different sources at all times are converted into the credibility under each warning level, so as to realize the mapping between the monitoring data and each warning level. In a possible implementation, the cloud model parameter values ​​corresponding to different warning levels of various types of monitoring data are used to calculate the credibility of each warning level corresponding to each type of monitoring data, including:

[0174] Based on the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, the second preset formula is used:

[0175] ;

[0176] Calculate the credibility of each warning level corresponding to each type of monitoring data, where i=1,2…n; j=1,2…5; t=1,2…m; Indicates The monitoring data of this category is Monitoring data within a time period The corresponding warning level is The credibility value of is a random variable.

[0177] To facilitate calculation, the credibility of each warning level is normalized. In a possible implementation, the normalization of the calculated credibility includes:

[0178] Through the third preset formula:

[0179] ;

[0180] Normalize the calculated credibility to obtain the normalized credibility vector , They represent the credibility values ​​of different warning levels corresponding to multiple types of monitoring data. The credibility of each warning level corresponding to each type of monitoring data forms a feature vector, which is called the credibility distribution of this type of monitoring data, where the warning level corresponding to the maximum credibility is the main level.

[0181] In the same time and space, the credibility values ​​of different types of monitoring data for each warning level mapped to geological disaster risk points are different, and there are even serious contradictions and conflicts. This is the direct reason why conflicting data cannot be effectively integrated. In the embodiment of the present invention, by calculating the consistency, conflict, conflict intensity and unreliability between multi-source monitoring data, parameters are provided for conflict data correction. In a possible implementation, the calculation of the consistency, conflict, conflict intensity and unreliability between multi-source monitoring data based on the normalized credibility distribution includes:

[0182] According to the normalized credibility, through the fourth preset formula:

[0183] ;

[0184] ;

[0185] ;

[0186] ;

[0187] Calculate the consistency and conflict between multi-source monitoring data, where: Indicates monitoring data and The amount of agreement between Indicates monitoring data and The amount of conflict between Indicates monitoring data With other monitoring equipment in The amount of consistency between monitoring data within a time period; Indicates monitoring data With other monitoring equipment in The amount of conflict between monitoring data within a time period; is the default value; Indicates the hth warning level.

[0188] In the embodiment of the present invention, based on the monitoring data With other monitoring equipment in The consistency and conflict between the monitoring data in a time period define the conflict intensity between the monitoring data. Specifically, according to the normalized credibility, the fifth preset formula is used:

[0189] ;

[0190] Calculate the conflict intensity between multi-source monitoring data, where: Indicates monitoring data With other monitoring equipment in The intensity of conflict between monitoring data within a time period; it combines the conflict amount and consistency amount of monitoring data to jointly characterize the degree of conflict between monitoring data, and is a comprehensive measure of the degree of conflict between monitoring data.

[0191] Due to the interference of various factors, monitoring data from different sources have a certain degree of unreliability. In the absence of accurate information, the unreliability of monitoring data is mainly estimated based on the "majority principle". Generally, the higher the degree of consistency between a piece of monitoring data and other monitoring data, the more the information expressed by the monitoring data is supported by other monitoring data, which means that the monitoring data is more reliable. Conversely, the monitoring data is more unreliable. Therefore, the consistency between monitoring data can be used to calculate the unreliability between monitoring data.

[0192] According to the normalized credibility, through the sixth preset formula:

[0193] ;

[0194] ;

[0195] Calculate the unreliability between multi-source monitoring data, where: Indicates monitoring data and The correlation coefficient between Indicates monitoring data unreliability.

[0196] Monitoring data from different sources sometimes conflict with each other, resulting in substantial differences in the credibility of different warning levels. The warning level obtained is =0.2, =0.8, while the monitoring data The corresponding warning information is =0.9, =0.1, the warning levels mapped by the two data are completely conflicting and have no compatible space, thus failing to provide effective information for warning. In order to reduce the conflict between data and increase the compatibility between data, in the embodiment of the present invention, the credibility distribution of the warning levels corresponding to various monitoring data is corrected.

[0197] Specifically, based on the conflict intensity and unreliability of the monitoring data, combined with the adjustment coefficient, an early warning level credibility redistribution equation can be constructed to achieve the initial correction of the early warning level credibility distribution. In a possible implementation, the early warning level credibility redistribution equation is constructed based on the consistency, conflict, conflict intensity and unreliability among multi-source monitoring data to correct the conflict of multi-source monitoring data, including:

[0198] Based on the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data, the seventh preset formula is used:

[0199] ;

[0200] The conflict of multi-source monitoring data was corrected for the first time, including: Indicates monitoring data Corresponding warning level after correction The credibility value of Θ represents the set of different warning levels. , Indicates monitoring data The original warning levels The partial value is transferred to the value of the whole set Θ; is the adjustment coefficient, when the monitoring data can be determined When the credibility values ​​of the corresponding warning levels are reasonable, The value is 0, when the monitoring data cannot be determined Whether the credibility values ​​of the corresponding warning levels are reasonable The value of is 0.5;

[0201] In the embodiment of the present invention, due to the first correction by the seventh preset formula It may cause an alert level If the value transferred to Θ is too high, the ambiguity of the information expressed by the monitoring data will increase, which will lead to an increase in the uncertainty of the fusion result. The adverse effects after redistribution need to be corrected by the eighth preset formula A second correction is performed. Among them, the main level contains the maximum amount of information mapped by the monitoring data. If the main level of the monitoring data does not change before and after the redistribution of the warning level credibility value, and the credibility value of the main level and the credibility value of other warning levels do not change significantly, then there will be no obvious difference in the information expressed by the monitoring data before and after the redistribution of the warning level credibility value. The purpose of the second correction of the warning level credibility value is to keep the credibility values ​​of the main level and other warning levels from changing dramatically before and after the correction, so as to avoid the generation of "overcorrection" in the correction of monitoring data. The greater the difference between the credibility value of the main level corresponding to a certain monitoring data and the credibility value of other warning levels corresponding to the data, the more accurate the information expressed by the data, but the more accurate the monitoring data, the more likely the corresponding credibility value is to be overcorrected. Therefore, the problem of overcorrection can be corrected from the perspective of the accuracy of the monitoring data. Information entropy is one of the effective ways to measure data accuracy. The embodiment of the present invention uses information entropy to calculate the accuracy of each data, and then uses the accuracy as a transfer factor to transfer part of the credibility value transferred to Θ during the redistribution process to the main level. The second correction is made through the eighth preset formula:

[0202] ;

[0203] ;

[0204] ;

[0205] The first correction result is corrected for the second time to obtain the corrected monitoring data information of various types of hidden danger points, wherein the and Respectively represent monitoring data Information entropy and precision; Indicates monitoring data The final revised credibility value corresponding to each warning level.

[0206] In a possible implementation, the method of using evidence theory to fuse new credibility of different warning levels corresponding to multi-source monitoring data to obtain a warning level credibility distribution based on multi-source data fusion includes:

[0207] Through the ninth preset formula:

[0208] ;

[0209] ;

[0210] ;

[0211] By integrating multi-source monitoring data, the credibility distribution of warning levels based on multi-source data fusion is obtained, where: express The monitoring data is in Fusion results within the time period; Indicates that the warning level in the fusion result of multi-source monitoring data is The credibility value of express The conflict coefficient between the monitoring data.

[0212] The warning level can be determined based on the fusion results and can be determined by the maximum credibility principle. Specifically, based on the multi-source data fusion results and the maximum credibility value principle, the warning level can be determined using the following formula ( ), and combined with the warning release rules, release the warning information to the outside through the system and related equipment terminals.

[0213] ;

[0214] in accordance with The warning level calculation process of multi-source monitoring data fusion within the time period is combined with the monitoring data collected by the monitoring equipment in real time to calculate The early warning results within the time period are updated in real time to achieve dynamic and instant early warning of geological disasters.

[0215] See also Figure 4 , Figure 4 An example diagram of a geological disaster instant warning method based on dynamic fusion of multi-source data provided by an embodiment of the present invention includes:

[0216] S41. Clarify the type of monitoring equipment deployed for the monitoring objects in the monitoring area, the monitoring physical parameters, and the monitoring data format;

[0217] S42. Preprocess multi-source monitoring data to prevent missing values ​​and illogical values;

[0218] S43. Based on the actual warning needs, clarify the warning level, and determine the different warning level intervals corresponding to each type of monitoring data in combination with historical disaster data;

[0219] S44. Based on the cloud model theory, calculate the cloud model parameter values ​​corresponding to different warning levels for each type of monitoring data: Expected ( ),entropy( )、Super Entropy( );

[0220] S45. Based on the cloud model parameters, the credibility of various monitoring data corresponding to different warning levels is calculated and normalized;

[0221] S46. Based on the credibility distribution of warning levels, measure the consistency, conflict, conflict intensity and unreliability among multi-source monitoring data;

[0222] S47. Construct a warning level credibility value redistribution equation to correct the conflict of multi-source monitoring data, reduce the conflict between data, and increase the compatibility between data;

[0223] S48. Based on the DS evidence theory fusion formula, multi-source monitoring data is integrated to obtain the credibility distribution of warning levels based on multi-source data fusion;

[0224] S49. Based on the principle of maximum credibility, determine the warning level, combine the warning rules, and release the warning information through the warning system;

[0225] S410. Based on the monitoring data collected at different times, repeat S45 to S49, dynamically integrate the multi-source monitoring data at each time, update the warning results in real time, and realize dynamic and immediate warning of geological disasters.

[0226] The present invention first maps various monitoring data into the credibility distribution of the warning level through the cloud model theory, and uses credibility to characterize the uncertainty of the warning level, thus overcoming the limitation that conventional methods are difficult to characterize the uncertainty of geological disasters; then, by calculating the consistency, conflict, conflict intensity, unreliability, accuracy, etc. between multi-source monitoring data, the warning level credibility redistribution equation is constructed to reduce the conflict of conflicting data, increase the compatibility between data, and use the DS evidence theory fusion rule to fuse multi-source monitoring data, solve the fusion problem of conflicting data, and provide a method for effectively utilizing multiple types of monitoring data; finally, based on the proposed method, combined with real-time dynamically collected data, the instant warning of geological disasters is realized, and the timeliness of the warning is improved. The present invention fully taps the value of various monitoring data through the dynamic real-time fusion of multi-source monitoring data, overcomes the limitation that conventional methods are difficult to utilize multiple types of data, and provides a scientific and feasible method for improving the accuracy and timeliness of warning results.

[0227] Another aspect of the present invention provides a geological disaster instant warning device based on dynamic fusion of multi-source data, see Figure 5 , Figure 5 A schematic diagram of the structure of a geological disaster instant warning device based on dynamic fusion of multi-source data provided by an embodiment of the present invention, the device comprises:

[0228] The data acquisition module 501 is used to determine the monitoring data type and physical parameters of the geological disaster potential point, wherein the potential point monitoring data is multiple types of data from multiple sensors, and the multiple types of monitoring data are real-time monitoring data collected at different times;

[0229] A preprocessing module 502 is used to preprocess the multi-source monitoring data to eliminate missing values ​​and illogical values;

[0230] The warning classification module 503 is used to classify the warning levels based on the actual warning requirements and determine the interval values ​​of different warning levels corresponding to each type of monitoring data;

[0231] The credibility calculation module 504 is used to calculate the cloud model parameter values ​​of different warning levels corresponding to each type of monitoring data based on the cloud model theory, wherein the cloud model parameter values ​​include: expected value, entropy value and super entropy; based on the cloud model parameter values ​​and the interval values ​​of the different warning levels, calculate the credibility of different warning levels corresponding to each type of monitoring data, and normalize the calculated credibility;

[0232] The conflict calculation module 505 is used to calculate the consistency, conflict, conflict intensity and unreliability between the multi-source monitoring data based on the normalized credibility distribution;

[0233] The data correction module 506 is used to construct a warning level credibility redistribution equation based on the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data, correct the conflict of the multi-source monitoring data, and obtain a new credibility of the warning level;

[0234] The data fusion module 507 is used to fuse the new credibility of different warning levels of multi-source monitoring data by using evidence theory to obtain the credibility distribution of warning levels based on multi-source data fusion;

[0235] The warning release module 508 is used to determine the warning level based on the maximum credibility value in the credibility distribution, release warning information through the warning system in combination with warning rules, and dynamically update the warning results based on the dynamic fusion of multi-source data collected at each moment.

[0236] In a possible implementation manner, the potential hazard point monitoring data information includes: the type of monitoring equipment, the monitored physical parameters, the format of the monitoring data, and the physical meaning represented by the monitoring data.

[0237] In a possible implementation, the warning classification module is specifically used to select a corresponding warning classification level according to the actual status of the monitoring area; determine relevant parameters of the warning classification level, wherein the warning classification level described in this embodiment takes level 5 as an example; and determine the interval values ​​of different warning levels corresponding to each type of monitoring data based on the relevant parameters of the warning classification level.

[0238] In a possible implementation, the credibility calculation module is specifically used to calculate the cloud model parameter values ​​corresponding to different warning levels of various types of monitoring data based on the interval values ​​of each warning level and in accordance with cloud model theory, on the assumption that the different warning levels corresponding to the monitoring data belong to a normal distribution.

[0239] In a possible implementation manner, the credibility calculation module is specifically configured to calculate the credibility through a first preset formula:

[0240] ;

[0241] ;

[0242] ;

[0243] Calculate the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, among which, For the The monitoring data of this category correspond to Expected value of each warning level; For the The monitoring data of this category correspond to Entropy value of warning level; For the The monitoring data of this category correspond to The super entropy value of the warning level; λ is the value range of 0 to Constants of; and Respectively The monitoring data of this category correspond to The lower and upper limits of the warning level range.

[0244] In a possible implementation, the calculation of the credibility of different warning levels corresponding to various types of monitoring data based on cloud model parameters includes:

[0245] Based on the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, the credibility of each warning level corresponding to each monitoring data is calculated.

[0246] In a possible implementation, the calculation of the credibility of each warning level corresponding to each type of monitoring data based on the cloud model parameter values ​​of different warning levels corresponding to each type of monitoring data includes:

[0247] Based on the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, the second preset formula is used:

[0248] ;

[0249] Calculate the credibility of each warning level corresponding to each type of monitoring data, where i=1,2…n; j=1,2…5; t=1,2…m; Indicates The monitoring data of this category is Monitoring data within a time period The corresponding warning level is The credibility value of is a random variable.

[0250] In a possible implementation, the normalizing the calculated credibility includes:

[0251] Through the third preset formula:

[0252] ;

[0253] Normalize the calculated credibility to obtain the normalized credibility vector , They respectively represent the credibility values ​​of multiple types of monitoring data corresponding to different warning levels.

[0254] In a possible implementation manner, the conflict calculation module is specifically configured to calculate the conflict according to the normalized credibility through a fourth preset formula:

[0255] ;

[0256] ;

[0257] ;

[0258] ;

[0259] Calculate the consistency and conflict between multi-source monitoring data, where: Indicates monitoring data and The amount of agreement between Indicates monitoring data and The amount of conflict between Indicates monitoring data With other monitoring equipment in The amount of consistency between monitoring data within a time period; Indicates monitoring data With other monitoring equipment in The amount of conflict between monitoring data within a time period; is the default value; Indicates the hth warning level;

[0260] According to the normalized credibility, through the fifth preset formula:

[0261] ;

[0262] Calculate the conflict intensity between multi-source monitoring data, where: Indicates monitoring data With other monitoring equipment in The conflict intensity between monitoring data within the time period;

[0263] According to the normalized credibility, through the sixth preset formula:

[0264] ;

[0265] ;

[0266] Calculate the unreliability between multi-source monitoring data, where: Indicates monitoring data and The correlation coefficient between Indicates monitoring data Unreliability;

[0267] The data correction module is specifically used to correct the consistency, conflict, conflict intensity and unreliability of the multi-source monitoring data through the seventh preset formula:

[0268] ;

[0269] The conflict of multi-source monitoring data was corrected for the first time, including: Indicates monitoring data Corresponding warning level after correction The credibility value of Θ represents the set of different warning levels. , Indicates monitoring data The original warning levels The partial value is transferred to the value of the whole set Θ; is the adjustment coefficient, when the monitoring data can be determined When the credibility values ​​of the corresponding warning levels are reasonable, The value is 0, when the monitoring data cannot be determined Whether the credibility values ​​of the corresponding warning levels are reasonable The value of is 0.5;

[0270] Through the eighth preset formula:

[0271] ;

[0272] ;

[0273] ;

[0274] The first correction result is corrected for the second time to obtain the corrected monitoring data information of various types of hidden danger points, wherein the and Respectively represent monitoring data Information entropy and precision; Indicates monitoring data The final revised credibility value corresponding to each warning level;

[0275] The data fusion module is specifically used to pass the ninth preset formula:

[0276] ;

[0277] ;

[0278] ;

[0279] By integrating multi-source monitoring data, the credibility distribution of warning levels based on multi-source data fusion is obtained, where: express The monitoring data is in Fusion results within the time period; Indicates that the warning level in the fusion result of multi-source monitoring data is The credibility value of express The conflict coefficient between the monitoring data.

[0280] In a possible implementation, the device further includes:

[0281] The real-time update module is used to obtain and dynamically integrate multi-source monitoring data at each moment based on monitoring data collected at multiple different moments, and update the warning results in real time.

[0282] It can be seen that through the device of the present invention, the conflict between data can be reduced and the compatibility between data can be increased by mapping the credibility distribution of monitoring data and warning levels and correcting the conflict of multi-source monitoring data. By utilizing the multi-source data fusion rules and combining the real-time collected monitoring data, the full fusion of multi-source monitoring data and the immediate warning of geological disasters are achieved, the problem of efficient fusion of multi-source conflicting data is solved, and the accuracy of geological disaster warnings is improved.

[0283] The embodiment of the present invention further provides an electronic device, such as Figure 6 As shown, it includes a processor 601 , a communication interface 602 , a memory 603 and a communication bus 604 , wherein the processor 601 , the communication interface 602 , and the memory 603 communicate with each other via the communication bus 604 .

[0284] Memory 603, used for storing computer programs;

[0285] The processor 601 is used to execute the program stored in the memory 603 to implement the following steps:

[0286] S1. Determine the monitoring data type and physical parameters of the geological disaster risk point, wherein the risk point monitoring data is multiple types of data from multiple sensors, and the multiple types of monitoring data are real-time monitoring data collected at different times;

[0287] S2. Preprocess the multi-source monitoring data to eliminate missing values ​​and illogical values;

[0288] S3. Based on the actual warning needs, the warning levels are divided and the interval values ​​of different warning levels corresponding to each type of monitoring data are determined;

[0289] S4. Based on the cloud model theory, calculate the cloud model parameter values ​​of different warning levels corresponding to each type of monitoring data, wherein the cloud model parameter values ​​include: expected value, entropy value and super entropy;

[0290] S5. Based on the cloud model parameter value and the interval values ​​of the different warning levels, calculate the credibility of the different warning levels corresponding to various types of monitoring data, and normalize the calculated credibility;

[0291] S6. Based on the normalized credibility distribution, calculate the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data;

[0292] S7, based on the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data, construct an early warning level credibility redistribution equation, correct the conflict of the multi-source monitoring data, and obtain a new credibility of the early warning level;

[0293] S8. Using evidence theory, the new credibility of different warning levels corresponding to multi-source monitoring data is integrated to obtain the credibility distribution of warning levels based on multi-source data fusion;

[0294] S9. Determine the warning level based on the maximum credibility value in the credibility distribution, and release warning information through the warning system in combination with warning rules.

[0295] The communication bus mentioned in the above electronic device can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. The communication bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus.

[0296] The communication interface is used for communication between the above electronic device and other devices.

[0297] The memory may include a random access memory (RAM) or a non-volatile memory (NVM), such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0298] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0299] In one embodiment provided by the present invention, a computer-readable storage medium is also provided, which stores a computer program. When the computer program is executed by a processor, the steps of any of the above-mentioned methods for instant warning of geological disasters based on dynamic fusion of multi-source data are implemented.

[0300] In one embodiment provided by the present invention, a computer program product containing instructions is also provided. When the computer is run on the computer, the computer executes any of the geological disaster instant warning methods based on dynamic fusion of multi-source data in the above embodiments.

[0301] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions may be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium may be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive Solid State Disk (SSD)), etc.

[0302] 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. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0303] Each embodiment in this specification is described in a related manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, electronic device, storage medium and computer program product embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0304] The above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention are included in the protection scope of the present invention.

Claims

1. A geological disaster instant warning method based on dynamic fusion of multi-source data, characterized in that: The method comprises: S1. Determine the monitoring data type and physical parameters of the geological disaster risk point, wherein the risk point monitoring data is multiple types of data from multiple sensors, and the multiple types of monitoring data are real-time monitoring data collected at different times; S2. Preprocess the multi-source monitoring data to eliminate missing values ​​and illogical values; S3. Based on the actual warning needs, the warning levels are divided and the interval values ​​of different warning levels corresponding to each type of monitoring data are determined; S4. Based on the cloud model theory, calculate the cloud model parameter values ​​of different warning levels corresponding to each type of monitoring data, wherein the cloud model parameter values ​​include: expected value, entropy value and super entropy; S5. Based on the cloud model parameter value and the interval values ​​of the different warning levels, calculate the credibility of the different warning levels corresponding to various types of monitoring data, and normalize the calculated credibility; S6. Based on the normalized credibility distribution, calculate the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data; S7, based on the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data, construct an early warning level credibility redistribution equation, correct the conflict of the multi-source monitoring data, and obtain a new credibility of the early warning level; S8. Using evidence theory, the new credibility of different warning levels corresponding to multi-source monitoring data is integrated to obtain the credibility distribution of warning levels based on multi-source data fusion; S9. Determine the warning level based on the maximum credibility value in the credibility distribution, and release warning information through the warning system in combination with warning rules.

2. The method according to claim 1, characterized in that The hidden danger point monitoring data information includes: the type of monitoring equipment, the monitored physical parameters, the format of the monitoring data and the physical meaning represented by the monitoring data.

3. The method according to claim 1, characterized in that Based on the actual warning needs, the warning levels are divided and the interval values ​​of different warning levels corresponding to each type of monitoring data are determined, including: Select the corresponding warning level according to the actual situation of the monitoring area; Determine the relevant parameters for warning classification levels; According to the relevant parameters of the warning classification level, the interval values ​​of different warning levels corresponding to each type of monitoring data are determined.

4. The method according to claim 1, characterized in that: The cloud model parameter values ​​corresponding to different warning levels for each type of monitoring data are calculated based on the cloud model theory, including: Based on the interval values ​​of each warning level and in accordance with cloud model theory, on the assumption that the monitoring data corresponding to different warning levels belong to a normal distribution, the cloud model parameter values ​​corresponding to different warning levels of various types of monitoring data are calculated.

5. The method according to claim 4, characterized in that The calculation of cloud model parameter values ​​corresponding to different warning levels of various monitoring data includes: By the first preset formula: ； ； ； Calculate the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, among which, For the The monitoring data of this category correspond to Expected value of each warning level; For the The monitoring data of this category correspond to Entropy value of warning level; For the The monitoring data of this category correspond to The super entropy value of the warning level; λ is the value range of 0 to Constants of; and Respectively The monitoring data of this category corresponds to The lower and upper limits of the warning level range.

6. The method according to claim 5, characterized in that The cloud model parameters are used to calculate the credibility of various monitoring data corresponding to different warning levels, including: Based on the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, the credibility of each warning level corresponding to each monitoring data is calculated.

7. The method according to claim 6, characterized in that The cloud model parameter values ​​corresponding to different warning levels of various monitoring data are used to calculate the credibility of each warning level corresponding to each monitoring data, including: Based on the cloud model parameter values ​​corresponding to different warning levels of various monitoring data, the second preset formula is used: ； Calculate the credibility of each warning level corresponding to each type of monitoring data, where i=1,2…n; j=1,2…5; t=1,2…m; Indicates The monitoring data of this category is Monitoring data within a time period The corresponding warning level is The credibility value of is a random variable.

8. The method according to claim 7, characterized in that The normalizing of the calculated credibility includes: Through the third preset formula: ; Normalize the calculated credibility to obtain the normalized credibility vector , They respectively represent the credibility values ​​of multiple types of monitoring data corresponding to different warning levels.

9. The method according to claim 8, characterized in that The calculation of the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data based on the normalized credibility distribution includes: According to the normalized credibility, through the fourth preset formula: ； ； ； ； Calculate the consistency and conflict between multi-source monitoring data, where: Indicates monitoring data and The amount of agreement between Indicates monitoring data and The amount of conflict between Indicates monitoring data The consistency between the monitoring data of other monitoring devices in the tth time period; Indicates monitoring data The amount of conflict between the monitoring data of other monitoring devices in the tth time period; is the default value; According to the normalized credibility, through the fifth preset formula: ； Calculate the conflict intensity between multi-source monitoring data, where: Indicates monitoring data The conflict intensity between the monitoring data of other monitoring devices in the tth time period; According to the normalized credibility, through the sixth preset formula: ； ； Calculate the unreliability between multi-source monitoring data, where: Indicates monitoring data and The correlation coefficient between Indicates monitoring data Unreliability; Based on the consistency, conflict, conflict intensity and unreliability among the multi-source monitoring data, an early warning level credibility redistribution equation is constructed to correct the conflict of the multi-source monitoring data, including: Through the seventh preset formula: ； The conflict of multi-source monitoring data was corrected for the first time, including: Indicates monitoring data Corresponding warning level after correction The credibility value of Θ represents the set of different warning levels. , Indicates monitoring data The original warning levels The partial value is transferred to the value of the whole set Θ; is the adjustment coefficient, when the monitoring data can be determined When the credibility values ​​of the corresponding warning levels are reasonable, The value is 0, when the monitoring data cannot be determined Whether the credibility values ​​of the corresponding warning levels are reasonable The value of is 0.5; Through the eighth preset formula: ； ； ； The first correction result is corrected for the second time to obtain the corrected monitoring data information of various types of hidden danger points, wherein the and Respectively represent monitoring data Information entropy and precision; Indicates monitoring data The final revised credibility value corresponding to each warning level; The above-mentioned use of evidence theory integrates the new credibility of different warning levels corresponding to multi-source monitoring data to obtain the warning level credibility distribution based on multi-source data fusion, including: Through the ninth preset formula: ； ； ; By integrating multi-source monitoring data, the credibility distribution of warning levels based on multi-source data fusion is obtained, where: express The monitoring data is in Fusion results within the time period; Indicates that the warning level in the fusion result of multi-source monitoring data is The credibility value of express The conflict coefficient between the monitoring data.

10. The method according to claim 1, characterized in that In the step of determining the warning level based on the maximum credibility value in the credibility distribution, and issuing warning information through the warning system in combination with the warning rule, the method further includes: Acquire and dynamically integrate multi-source monitoring data from each time period based on S5 to S9 based on monitoring data collected at multiple different times, and dynamically update geological disaster warning information at the monitoring points in real time.

Images