Regional dynamic monitoring and early warning method and system based on multi-source space-time big data fusion

Abstract

The invention discloses a regional dynamic monitoring and early warning method and system based on multi-source space-time big data fusion, and relates to the technical field of regional dynamic monitoring. The system comprises a multi-source space-time big data fusion module used for obtaining feature information data of a to-be-analyzed area; the first disaster type analysis module is used for determining a first disaster type of the to-be-analyzed area based on the feature information data of the to-be-analyzed area; the regional dynamic monitoring analysis module is used for determining index data of a second disaster type based on the first disaster type; and the combined disaster prevention early warning module is used for determining a disaster prevention combined area based on the index data of the second disaster type, outputting a resource reserve plan of the disaster prevention combined area, monitoring resource reserve changes in real time and providing early warning information based on a set threshold value. Technical support and decision basis can be provided for intelligent disaster prevention, disaster prevention mechanism monitoring is dynamically combined, government departments are supported to formulate more efficient disaster prevention policies, and the public safety governance level is improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of regional dynamic monitoring, and particularly relates to a regional dynamic monitoring and early warning method and system based on multi-source spatio-temporal big data fusion. BACKGROUND

[0002] The disaster prevention planning of territorial space is a special planning formulated in the planning system of territorial space to cope with natural disasters, accidents and other emergencies, and to improve the resilience and safety of territorial space. At present, in the disaster prevention planning of territorial space, through the fusion of a plurality of front-end data, the regional layout and resource allocation are adjusted, and the disaster risk is reduced, and at the same time, the post-disaster protection is established.

[0003] However, there are still many deficiencies in the current regional disaster prevention field. For example, in the control of disasters, the current technology often determines the main types of disasters by dividing high-risk and low-risk areas, but ignores the analysis of secondary disasters caused by main disasters, such as traffic congestion pressure caused by fire accidents, which may hinder rescue, etc. In the dynamic configuration monitoring of the region, the problem of how to set up a joint disaster prevention region to reduce resource consumption as much as possible has not been solved. SUMMARY

[0004] The present application aims to provide a regional dynamic monitoring and early warning method and system based on multi-source spatio-temporal big data fusion to solve the problems in the prior art.

[0005] To achieve the above-mentioned purpose, the present application provides the following technical scheme: a regional dynamic monitoring and early warning method based on multi-source spatio-temporal big data fusion, the method comprising:

[0006] obtaining characteristic information data of a region to be analyzed;

[0007] determining a first disaster type of the region to be analyzed based on the characteristic information data of the region to be analyzed;

[0008] determining index data of a second disaster type based on the first disaster type, the index data of the second disaster type comprising determining an influence area of the second disaster type in combination with the first disaster type, establishing regional dynamic monitoring and analysis of the region to be analyzed under multiple disaster types;

[0009] determining a disaster prevention joint region based on the index data of the second disaster type, outputting a resource reserve plan of the disaster prevention joint region, monitoring the change of resource reserve in real time, and providing early warning information based on a set threshold.

[0010] Further, the characteristic information data of the region to be analyzed comprises:

[0011] population density in the region, daily traffic flow in the region, and historical disaster data in the region;

[0012] The daily traffic flow in the region is monthly, and the average is obtained by summing up in a cycle;

[0013] The historical disaster data in the region includes the types of disasters occurring in the region and the number of each disaster;

[0014] The feature information data is collected based on multi-source spatio-temporal big data.

[0015] Further, the feature information data based on the analysis of the region, determine the first disaster type of the region to be analyzed includes:

[0016] Get the regional feature information data when the historical data disaster occurs, recorded as a group of data, get several groups of data to form a first data set;

[0017] Get the feature information data of the region to be analyzed, write it as a first data group, mix the first data group into the first data set, and perform classification operation to determine the output type of the first data group, which includes:

[0018] Get the data amount of the first data set, recorded as N, take the data group less than N in the first data set as the data clustering core, respectively calculate the correlation value Y between other data groups in the first data set and the data clustering core, and take the corresponding data clustering core with the minimum correlation value Y as the assigned data clustering core. Each data clustering core forms a new data combination, and sets a stable threshold for the data combination. If the data amount in a data combination exceeds the stable threshold of the data combination, take the average value in the data combination as a new data clustering core to form a new batch of data clustering cores, and reassign the data in the first data set;

[0019] Constantly reassign, until the total number of data clustering cores formed by two consecutive assignments is the same or there is no data amount in a data combination exceeding the stable threshold of the data combination;

[0020] The calculation method of the correlation value Y includes:

[0021] Y=a1x1+a2x2+a3x3

[0022] Wherein, a1, a2, a3 respectively refer to the weight coefficient set by the system; x1, x2, x3 respectively refer to the normalized data of population density in the region, daily traffic flow in the region and the number of occurrence of this type of disaster data in the region;

[0023] The average value in the data combination is formed by summing up and averaging all x1, x2, x3 in the data combination.

[0024] Further, further comprising:

[0025] According to the finally formed classification group, find the group where the feature information data of the region to be analyzed is located, analyze the disaster types of other data in the group, and select the disaster type with the most disaster types as the first disaster type of the region to be analyzed;

[0026] If there is no other data in the group, data warning is performed to the administrator end;

[0027] If there are multiple disaster types in the group with the same number, the disaster type with fewer disaster types in the first data set is preferentially selected for output, and if they are still the same, one of them is randomly selected for output.

[0028] Further, the index data of the second disaster type based on the first disaster type comprises:

[0029] Randomly divide the region to be analyzed into a plurality of sub-regions, and mark the ratio relationship between the feature information data of each sub-region and the feature information data of the region to be analyzed;

[0030] In each sub-region, the corresponding feature information data with the largest ratio is the main information data of the sub-region, and each sub-region has and only one main information data;

[0031] Based on the occurred disaster data, the second disaster type that can occur under the first disaster type is obtained;

[0032] In the historical data, the region data where any second disaster type occurs is obtained, a threshold value is set for each feature information data, if a certain feature information data in a certain region data exceeds the set threshold value, the region data is sent to the second training group, and the region data is marked as having a certain feature information data;

[0033] The system sets the division number of the region to be analyzed, randomly divides a plurality of sub-regions based on the division number, determines the main information data of any sub-region, and determines the influence range of any sub-region based on the main information data determined by any sub-region in the second training group:

[0034] In the second training group, select the region data where the main information data determined by any sub-region exists as a training set, calculate the ratio of the influence area of the region to the area of the region itself, output a plurality of data ratios based on the training set, determine the prediction output value of the plurality of data ratios as the prediction ratio of the current any sub-region based on the gray prediction algorithm, and output the influence area of the current any sub-region excluding the region itself based on the prediction ratio;

[0035] The system sets the number of divisions based on computer computing power, and in each division, the sum of the affected area of each sub-region excluding the self-region is calculated, and the sum of the affected area of each sub-region excluding the self-region is marked in each division;

[0036] In each division, the sum of the affected area of each sub-region excluding the self-region is calculated for all possible second disaster types, and a data set after each division is formed, denoted as disaster data division set, expressed as {b1, b2, …, b n} where b1, b2, …, b n represent all possible second disaster types corresponding to the first disaster type.

[0037] Further, it also includes:

[0038] For any division, determine the division score:

[0039] K = c1*g-c2*h+c3*f

[0040] Where K refers to the division score; c1, c2, c3 refer to the weight of the system division respectively; g refers to the normalized value of the sum of the area sum marked in each division; h refers to the normalized value of the regional joint capacity score; f refers to the number of divisions of the region to be analyzed;

[0041] Take the minimum value of K as the division of the region to be analyzed, and perform dynamic monitoring of the region to be analyzed;

[0042] The regional joint capacity score includes:

[0043] Get the sub-region under any division, if there are any two sub-regions intersecting and the absolute value difference of the affected area excluding the self-region under the same second disaster type is less than the system preset absolute value difference threshold, then count 1, and combine all sub-region data to form the regional joint capacity score under any division.

[0044] Further, the index data based on the second disaster type determines the disaster prevention joint region, outputs the resource reserve plan of the disaster prevention joint region, monitors the resource reserve change in real time, and provides early warning information based on the set threshold, including:

[0045] Under the dynamic monitoring of the region to be analyzed, take the intersection point between each sub-region to set up a disaster prevention joint region, set the resource reserve index of the disaster prevention joint region based on the possible second disaster type on the disaster prevention joint region, monitor the resource reserve change in real time, and provide early warning information when the resource reserve is lower than the set resource reserve threshold.

[0046] The regional dynamic monitoring and early warning system based on multi-source spatio-temporal big data fusion comprises:

[0047] A multi-source spatio-temporal big data fusion module is configured to acquire characteristic information data of a region to be analyzed.

[0048] A first disaster type analysis module is configured to determine a first disaster type of the region to be analyzed based on the characteristic information data of the region to be analyzed.

[0049] A regional dynamic monitoring and analysis module is configured to determine index data of a second disaster type based on the first disaster type, wherein the index data of the second disaster type comprises an influence region of the second disaster type determined in combination with the first disaster type, and the regional dynamic monitoring and analysis of the region to be analyzed under multiple disaster types is established.

[0050] A joint disaster prevention and early warning module is configured to determine a joint disaster prevention region based on the index data of the second disaster type, output a resource reserve plan of the joint disaster prevention region, monitor resource reserves in real time, and provide early warning information based on a set threshold.

[0051] A terminal comprises a processor and a storage medium, wherein the storage medium is configured to store instructions, and the processor is configured to operate according to the instructions to perform the steps of the method.

[0052] A computer readable storage medium stores a computer program, which is executed by a processor to implement the steps of the method.

[0053] Compared with the prior art, the present application has the following advantages: the present application comprehensively considers major disasters and secondary disasters (such as traffic congestion and environmental pollution caused by fire) caused by the major disasters, establishes a multi-disaster coupling analysis, can more comprehensively evaluate disaster risks, and optimizes emergency resource allocation. Avoiding rescue delay or resource waste caused by neglecting secondary disasters, significantly improving the accuracy and effectiveness of disaster prevention and mitigation. Optimizing the dynamic monitoring and resource allocation of the joint disaster prevention region, realizing real-time data sharing and collaborative response of the joint region and multiple disasters. Optimizing the deployment of rescue forces, reducing repeated investment, reducing resource consumption, and improving overall disaster prevention efficiency. The present application can provide technical support and decision basis for intelligent disaster prevention, dynamic joint disaster prevention mechanism monitoring, support government departments to develop more efficient disaster prevention policies, and improve the level of public safety governance. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 The figure is a step schematic diagram of the regional dynamic monitoring and early warning method based on multi-source spatio-temporal big data fusion of the present application.

[0055] Figure 2 The figure is a structure schematic diagram of the regional dynamic monitoring and early warning system based on multi-source spatio-temporal big data fusion of the present application. DETAILED DESCRIPTION

[0056] All other embodiments obtained by a person of ordinary skill in the art based on the embodiments in the present application without creative labor fall within the scope of protection of the present application.

[0057] Embodiment: As shown in Figure 1 and Figure 2 The present application provides a regional dynamic monitoring and early warning method based on multi-source spatio-temporal big data fusion, which comprises the following steps:

[0058] obtaining characteristic information data of a region to be analyzed;

[0059] The characteristic information data of the region to be analyzed comprises:

[0060] population density in the region, daily traffic flow in the region, and historical disaster data in the region;

[0061] The daily traffic flow in the region is averaged by summing up in a cycle of one month;

[0062] The historical disaster data in the region comprises the types of disasters occurring in the region and the number of each type of disaster;

[0063] The characteristic information data is collected and obtained based on multi-source spatio-temporal big data. For example, satellite navigation positioning, remote sensing image, mobile phone signaling, and vehicle networking trajectory data. Specifically, data collected by GPS, Beidou, GLONASS, and other satellite positioning systems comprises:

[0064] Integrating a GNSS receiving module in a target device (such as a smart phone, a vehicle-mounted terminal, and a wearable device). Receiving satellite signals, recording longitude, latitude, elevation, timestamp, speed, direction, and other information. The sampling frequency is set according to the requirements (such as 1Hz for normal positioning and 10Hz for high dynamic scene).

[0065] Eliminating outliers (such as drift points caused by signal shielding), improving accuracy (centimeter level) through differential positioning (RTK / PPK), and then uploading data through 4G / 5G or Internet of Things protocol (MQTT).

[0066] Based on the characteristic information data of the region to be analyzed, determining a first disaster type of the region to be analyzed;

[0067] The determination of the first disaster type of the region to be analyzed based on the characteristic information data of the region to be analyzed comprises:

[0068] Obtaining regional characteristic information data at the time of disaster occurrence of historical data, denoted as a group of data, and obtaining a plurality of groups of data to form a first data set;

[0069] Obtaining the characteristic information data of the region to be analyzed, writing as a first data group, mixing the first data group into a first data set, performing a classification operation, and judging the output type of the first data group, specifically including:

[0070] Obtaining the data amount of the first data set, denoted as N, taking a data group smaller than N in the first data set as a data clustering core, respectively calculating the correlation value Y between other data groups in the first data set and the data clustering core, and taking the corresponding data clustering core with the smallest correlation value Y as the assigned data clustering core, each data clustering core forming a new data combination, setting a stable threshold of the data combination, if the data amount in a data combination exceeds the stable threshold of the data combination, taking the average value in the data combination as a new data clustering core, forming a new batch of data clustering cores, and re-distributing the data in the first data set;

[0071] Continuously re-distributing until the total number of data clustering cores formed by two consecutive distributions is the same or there is no data amount in a data combination exceeding the stable threshold of the data combination;

[0072] The calculation method of the correlation value Y includes:

[0073] Y=a1x1+a2x2+a3x3

[0074] Wherein, a1, a2, a3 respectively represent the weight coefficients set by the system; x1, x2, x3 respectively represent the normalized data of the population density in the region, the daily traffic flow in the region, and the occurrence frequency of the disaster data of this type in the region;

[0075] The average value in the data combination is formed by respectively summing and averaging all x1, x2, x3 in the data combination.

[0076] Further comprising:

[0077] According to the finally formed classification group, finding the group where the characteristic information data of the region to be analyzed is located, analyzing the disaster types of other data in the group, and selecting the disaster type with the most disaster types as the first disaster type of the region to be analyzed;

[0078] If there is no other data in the group, a data warning is sent to the administrator end;

[0079] If there are multiple disaster types in the group with the same number, the disaster type with fewer disaster types in the first data set is preferentially selected as the output, and if they are still the same, one of them is randomly selected as the output.

[0080] Based on the first disaster type, determine the index data of the second disaster type, including determining the impact area of the second disaster type in combination with the first disaster type, establishing the regional dynamic monitoring analysis of the to-be-analyzed area under multiple disaster types;

[0081] The determination of the index data of the second disaster type based on the first disaster type includes:

[0082] Randomly divide the to-be-analyzed area into a plurality of sub-regions, and mark the ratio relationship between the characteristic information data of each sub-region and the characteristic information data of the to-be-analyzed area;

[0083] Among them, the disaster type includes: earthquake, fire, disease transmission, chemical leakage and other geological disasters;

[0084] In each sub-region, the corresponding characteristic information data with the largest ratio is the main information data of the sub-region, and each sub-region has and only one main information data;

[0085] Based on the disaster data that has occurred, the second disaster type that can occur under the first disaster type is obtained;

[0086] In the historical data, the region data of any second disaster type is obtained, and a threshold value is set for each characteristic information data. If a certain characteristic information data in a certain region data exceeds the set threshold value, the region data is sent to the second training group, and it is marked that the region data has a certain characteristic information data;

[0087] The system sets the division number of the to-be-analyzed area, randomly divides a plurality of sub-regions based on the division number, determines the main information data of any sub-region, and determines the influence range of any sub-region based on the main information data determined by any sub-region in the second training group:

[0088] In the second training group, the region data with the main information data determined by any sub-region is selected as the training set, and the ratio of the influence area to the area of the region itself is calculated. Based on the training set, output a plurality of data ratios, and based on the gray prediction algorithm, determine the prediction output value of the plurality of data ratios as the prediction ratio of the current any sub-region, and output the influence area of the current any sub-region excluding the region itself based on the prediction ratio.

[0089] Specifically, it includes setting a plurality of data ratios as a first data set, performing whitening differentiation after gray accumulation generation processing on the first data set, and outputting development coefficient and gray action amount;

[0090] Based on the development coefficient and the gray action amount, the least square method is constructed to solve:

[0091]

[0092] wherein, d n+1 represents the predicted ratio; L represents the development coefficient; p represents the gray amount; m refers to the data amount of the first data set; d1 represents the first data in the first data set;

[0093] The system sets the division times based on computer computing power. In each division, the sum of the affected area of all sub-regions except the self-region after each division is calculated and marked in each division time.

[0094] In each division, the sum of the affected area of all sub-regions except the self-region of all possible second type disasters is marked to form a data set after each division, denoted as disaster data division set, expressed as {b1, b2, …, b n}, wherein b1, b2, …, b n represent all possible second disaster types corresponding to the first disaster type, respectively.

[0095] Further comprising:

[0096] The division score is determined for each division:

[0097] K = c1*g-c2*h+c3*f

[0098] wherein, K refers to the division score; c1, c2, c3 refer to the weight of the system division, respectively; g refers to the normalized value of the sum of the area sum marked in each division time; h refers to the normalized value of the regional joint capacity score; f refers to the division number of the region to be analyzed.

[0099] The minimum value of K is taken as the division of the region to be analyzed, and the dynamic monitoring of the region to be analyzed is carried out.

[0100] The regional joint capacity score includes:

[0101] The sub-regions under any division time are obtained. If the absolute value difference of the affected area of any two sub-regions intersecting and under the same second disaster type except the self-region is less than the absolute value difference threshold preset by the system, then the count is 1. The data of all sub-regions are integrated to form the regional joint capacity score under any division time.

[0102] Specifically, any two sub-regions intersecting means that the two sub-regions are adjacent and there is an area in contact;

[0103] The second disaster type index data is used to determine a disaster prevention joint area, output a resource reserve plan of the disaster prevention joint area, monitor resource reserve changes in real time, and provide early warning information based on a set threshold.

[0104] Under dynamic monitoring of the to-be-analyzed area, an intersection between each sub-area is taken as a disaster prevention joint area, a resource reserve index of the disaster prevention joint area is set based on a possible second disaster type, resource reserve changes are monitored in real time, and early warning information is provided when the resource reserve is lower than a set resource reserve threshold.

[0105] The regional dynamic monitoring and early warning system based on multi-source spatio-temporal big data fusion includes:

[0106] The multi-source spatio-temporal big data fusion module 101 is configured to acquire feature information data of a to-be-analyzed area.

[0107] The first disaster type analysis module 102 is configured to determine a first disaster type of the to-be-analyzed area based on the feature information data of the to-be-analyzed area.

[0108] The regional dynamic monitoring and analysis module 103 is configured to determine index data of a second disaster type based on the first disaster type, wherein the index data of the second disaster type includes determining an influence area of the second disaster type in combination with the first disaster type, and establishing regional dynamic monitoring and analysis of the to-be-analyzed area under multiple disaster types.

[0109] The joint disaster prevention and early warning module 104 is configured to determine a disaster prevention joint area based on the index data of the second disaster type, output a resource reserve plan of the disaster prevention joint area, monitor resource reserve changes in real time, and provide early warning based on a set threshold.

[0110] The embodiments of the present application also provide a computer program product, which includes computer program codes, and when the computer program codes are run on a computer, the computer is caused to implement the method in the above-mentioned embodiments of the present application.

[0111] The embodiments of the present application also provide a computer readable storage medium, which stores computer instructions, and when the computer instructions are run on a computer, the computer is caused to implement the method in the above-mentioned embodiments of the present application.

[0112] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described platform, device and unit can refer to the corresponding process in the foregoing method embodiments, which will not be described herein.

[0113] The prefix of "first", "second" and the like in the embodiments of the present application are merely used to distinguish different description objects, and do not have the limitation on the position, order, priority, number or content of the described objects. The use of the prefix of ordinal numbers and the like for distinguishing the description objects in the embodiments of the present application does not constitute the limitation on the described objects, and the description of the described objects should refer to the description in the context of the claims or embodiments, and should not constitute the redundant limitation because of the use of the prefix.

[0114] In several embodiments provided in the present application, it should be understood that the disclosed system, device and method can be implemented in other manners. For example, the described device embodiments are merely schematic, and the division of the units is merely a logical function division, and there can be another division manner in actual implementation. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In addition, the displayed or discussed mutual couplings or direct couplings or communication connections between different units, can be indirect couplings or communication connections through some interfaces, devices or units, and can be electrical, mechanical or in other forms.

[0115] In each embodiment of the present application, the terms and / or descriptions between different embodiments are consistent and can be mutually referred to, and the technical features in different embodiments can be combined to form a new embodiment according to the inherent logical relationship, if there is no special description and logical conflict.

[0116] The units described as separate components can or can not be physically separate, and the components displayed as units can or can not be physical units, i.e. can be located in one place, or can be distributed on a plurality of network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the embodiment scheme.

[0117] In addition, each functional unit in the embodiments of the present application can be integrated in one processing unit, or each unit can be physically present separately, or two or more units can be integrated in one unit.

[0118] The above is merely specific implementation of the present application, but the protection scope of the present application is not limited to this, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed in the present application, which should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A regional dynamic monitoring and early warning method based on multi-source spatio-temporal big data fusion, characterized in that: The method comprises: obtaining characteristic information data of a region to be analyzed; determining a first disaster type of the region to be analyzed based on the characteristic information data of the region to be analyzed; determining index data of a second disaster type based on the first disaster type, the index data of the second disaster type comprising determining an influence region of the second disaster type in combination with the first disaster type, establishing regional dynamic monitoring and analysis of the region to be analyzed under multiple disaster types; determining a disaster prevention joint region based on the index data of the second disaster type, outputting a resource reserve plan of the disaster prevention joint region, monitoring resource reserve changes in real time, and providing early warning information based on a set threshold. 2.The regional dynamic monitoring and early warning method based on multi-source spatio-temporal big data fusion according to claim 1, characterized in that: The characteristic information data of the region to be analyzed comprises: population density in the region, daily traffic flow in the region, and historical disaster data in the region; the daily traffic flow in the region is averaged by summing up in a monthly cycle; the historical disaster data in the region comprises disaster types occurring in the region and the number of times of each disaster; the characteristic information data is collected based on multi-source spatio-temporal big data. 3.The regional dynamic monitoring and early warning method based on multi-source spatio-temporal big data fusion according to claim 1, characterized in that: The determination of the first disaster type of the region to be analyzed based on the characteristic information data of the region to be analyzed comprises: obtaining regional characteristic information data at the time of disaster occurrence, denoted as a group of data, obtaining a plurality of groups of data to form a first data set; obtaining the characteristic information data of the region to be analyzed, writing it as a first data group, mixing the first data group into the first data set, performing classification operation, and determining the output type of the first data group, specifically comprising: obtaining the data amount of the first data set, denoted as N, taking data groups smaller than N in the first data set as data clustering cores, respectively calculating the correlation value Y between other data groups in the first data set and the data clustering cores, taking the data clustering core corresponding to the minimum correlation value Y of each other data group as the allocated data clustering core, forming a completely new data combination for each data clustering core, setting a stable threshold for the data combination, and if the data amount in a data combination exceeds the stable threshold of the data combination, taking the average value in the data combination as a new data clustering core to form a new batch of data clustering cores, and re-distributing the data in the first data set; continuously re-distributing until the total number of data clustering cores formed by two consecutive distributions is the same or there is no data amount in a data combination exceeding the stable threshold of the data combination; The calculation method of the correlation value Y comprises: Y = a1x1 + a2x2 + a3x3 wherein a1, a2, and a3 are weight coefficients set by the system; x1, x2, and x3 are normalized data of population density in the region, daily traffic flow in the region, and the number of times of the type of disaster data in the region, respectively; The average value in the data combination is formed by summing up and averaging all x1, x2, and x3 in the data combination.

4. The regional dynamic monitoring and early warning method based on multi-source spatio-temporal big data fusion according to claim 3, characterized in that: Further comprising: according to the finally formed classification group, finding the group where the characteristic information data of the region to be analyzed is located, analyzing the disaster types of other data in the group, and selecting the disaster type with the most as the first disaster type of the region to be analyzed; If there is no other data in the group, a data warning is sent to the administrator end; If there are multiple disaster types in the group with the same number, the disaster type with fewer disaster types in the first data set is selected as the output, and if they are still the same, one of them is randomly selected as the output.

5. The regional dynamic monitoring and early warning method based on multi-source spatio-temporal big data fusion according to claim 4, characterized in that: The index data of the second disaster type is determined based on the first disaster type, including: Randomly divide the area to be analyzed into several sub-regions, and mark the ratio relationship between the characteristic information data of each sub-region and the characteristic information data of the area to be analyzed; In each sub-region, the corresponding characteristic information data with the largest ratio is the main information data of the sub-region, and each sub-region has only one main information data; Based on the occurred disaster data, the second disaster type that can occur under the first disaster type is obtained; In the historical data, the region data of any second disaster type is obtained, and a threshold value is set for each characteristic information data. If a certain characteristic information data in a certain region data exceeds the set threshold value, the region data is sent to the second training group, and the region data is marked as having a certain characteristic information data; The system sets the division number of the area to be analyzed, randomly divides several sub-regions based on the division number, determines the main information data of any sub-region, and determines the influence range of any sub-region in the second training group based on the main information data determined by any sub-region: In the second training group, the region data with the main information data determined by any sub-region is selected as the training set, the ratio of the influence area to the area of the region itself is calculated, the prediction output value of the several data ratios is output based on the training set, the prediction output value of the several data ratios is determined based on the gray prediction algorithm as the prediction ratio of the current any sub-region, and the influence area of the current any sub-region excluding the area itself is output based on the prediction ratio. The system sets the division number based on the computer computing power. In each division, the sum of the influence area of all sub-regions excluding the area itself after each division is calculated, and the sum is marked in each division number. In each division, all the second type disasters that can occur are marked to sum up the impact area of all sub-regions except the self-region, forming the data set after each division, recorded as disaster data division set, expressed as {b1, b2, …, b n}, wherein b1, b2, …, b n represent all the second type disasters that can occur corresponding to the first disaster type respectively. 6.The regional dynamic monitoring and early warning method based on multi-source spatio-temporal big data fusion according to claim 5, characterized in that: Also includes: Determine the division score for each division: K=c1*g-c2*h+c3*f Where K is the division score; c1, c2, and c3 are the weights of the system division; g is the normalized value of the sum of the area sum marked in each division number; h is the normalized value of the region joint ability score; f is the division number of the area to be analyzed; Take the minimum value of K as the division of the area to be analyzed, and perform dynamic monitoring of the area to be analyzed; The region joint ability score includes: Get the sub-region under any division number. If there are two sub-regions intersecting and the absolute value difference of the influence area excluding the area itself under the same second disaster type is less than the system preset absolute value difference threshold, count 1, and combine all sub-region data to form the region joint ability score under any division number.

7. The regional dynamic monitoring and early warning method based on multi-source spatio-temporal big data fusion according to claim 6, characterized in that: The index data based on the second disaster type determines a joint disaster prevention area, outputs a resource reserve plan of the joint disaster prevention area, monitors resource reserve changes in real time, and provides early warning information based on a set threshold. Under dynamic monitoring of the area to be analyzed, an intersection between each sub-area is taken to set up a joint disaster prevention area, resource reserve indexes of the joint disaster prevention area are set based on a possible second disaster type on the joint disaster prevention area, resource reserve changes are monitored in real time, and early warning information is provided when the resource reserve is lower than a set resource reserve threshold.

8. A regional dynamic monitoring and early warning system based on multi-source spatio-temporal big data fusion, characterized in that: The system comprises: a multi-source spatio-temporal big data fusion module configured to acquire feature information data of an area to be analyzed; a first disaster type analysis module configured to determine a first disaster type of the area to be analyzed based on the feature information data of the area to be analyzed; a regional dynamic monitoring analysis module configured to determine index data of a second disaster type based on the first disaster type, wherein the index data of the second disaster type comprises determining an influence area of the second disaster type in combination with the first disaster type, and establishing regional dynamic monitoring analysis of the area to be analyzed under multiple disaster types; a joint disaster prevention early warning module configured to determine a joint disaster prevention area based on the index data of the second disaster type, output a resource reserve plan of the joint disaster prevention area, monitor resource reserve changes in real time, and provide early warning information based on a set threshold. 9.A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is configured to store instructions; and the processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-7.

10. A computer readable storage medium having stored thereon a computer program, characterized in that, The program, when executed by the processor, implements the steps of the method according to any one of claims 1-7.

Images