Environment event auxiliary prediction method and prediction system based on data analysis

Abstract

The invention provides an environment event auxiliary prediction method and prediction system based on data analysis, and the method comprises the steps: obtaining any environment object and the correlation between the attribute characteristics of the environment object and an environment event, and constructing a primary model; analyzing historical environment objects in the samples and attribute features of the historical environment objects; acquiring a feature set which is spatially associated with the attribute features of the environment object; performing correlation analysis on any element in the feature set and the sample to obtain spatial correlation; adjusting the primary model according to spatial correlation to obtain a prediction model; meanwhile, environment data of a to-be-predicted area and an area having spatial correlation with the to-be-predicted area are collected, and the environment data are screened according to the prediction model; and predicting the environment event of the to-be-predicted area by using the prediction model. According to the method, on the premise of ensuring the prediction accuracy, the data volume required during operation is reduced, and the hardware demand during prediction is reduced.

Description

Technical Field

[0001] The present disclosure relates to the field of environmental event prediction, and in particular to an environmental event auxiliary prediction method and prediction system based on data analysis. Background Art

[0002] Environmental monitoring is a key area in modern society, which involves all aspects of our lives, work and environment. At the same time, environmental monitoring is a complex system involving a large amount of data, which can collect, process and analyze real-time data of environmental parameters. Through these data, we can understand the environmental conditions and formulate effective environmental protection measures; due to its huge amount of data and high data flow, traditional data processing methods can no longer meet the requirements of real-time and accuracy. Existing, artificial intelligence technology is usually applied in the field of environmental monitoring.

[0003] The existing environmental event prediction process is usually as follows: first, the collected data usually needs to be cleaned, filtered, and standardized for subsequent analysis; then, statistical methods, machine learning algorithms, or deep learning models are used to analyze the data in the prediction area to extract meaningful information; then, based on the analysis results, a prediction model is constructed to predict the future environmental conditions or environmental risk events in the prediction area.

[0004] However, when predicting the future environmental conditions of the area to be predicted or the environmental risk events that may occur, due to the huge amount of data, it is usually very dependent on the computing power of the computer and the hardware requirements for data processing are very high.

[0005] Based on this, the purpose of the present invention is to provide an auxiliary prediction method and prediction system for environmental events based on data analysis, so as to use a limited amount of data for prediction while ensuring the accuracy of prediction, thereby reducing the hardware requirements for data processing. Summary of the invention

[0006] In order to solve the problems of the prior art, the present disclosure provides an auxiliary prediction method and prediction system for environmental events based on data analysis, which can solve at least one technical problem in the prior art.

[0007] The technical solution adopted by the present disclosure to solve the above technical problems is:

[0008] An auxiliary prediction method for environmental events based on data analysis, comprising:

[0009] Obtaining an environmental object corresponding to any environmental event and attribute characteristics of the environmental object, and obtaining any environmental object and the correlation between the attribute characteristics of any environmental object and the environmental event, to construct a primary model;

[0010] Collecting a preset amount of historical environmental event data as samples, and analyzing historical environmental objects in the samples and attribute characteristics of the historical environmental objects;

[0011] Analyzing the attribute features of the historical environment object and the attribute features of the environment object to obtain a feature set that is spatially associated with the attribute features of the environment object;

[0012] Performing correlation analysis on any element in the feature set and the sample to obtain spatial correlation;

[0013] Adjusting the primary model according to the spatial correlation to obtain a prediction model for predicting the environmental event;

[0014] At the same time, collect the unused environmental monitoring data of the area to be predicted and the area that has spatial correlation with the area to be predicted, filter the unused environmental monitoring data according to the prediction model, reduce the amount of data, and obtain environmental data containing spatial correlation; based on the environmental data containing spatial correlation, use the prediction model to predict environmental events in the area to be predicted.

[0015] The collecting of environmental data of the area to be predicted and predicting environmental events in the area to be predicted according to the prediction model includes:

[0016] Receive a prediction task, and store the corresponding environmental object to be predicted and the attribute feature data of the environmental object to be predicted in the prediction task in a parent database of a data center;

[0017] Send a request to the front end to obtain the unused environmental monitoring data of the area to be predicted and the area that has spatial correlation with the area to be predicted, and generate a sub-database;

[0018] Sending the environmental monitoring data of any of the sub-databases to a preset terminal, and at the preset terminal, dividing the sub-database into at least two data modules, jointly acquiring the environmental monitoring data related to the prediction model in the sub-database to clean up useless data and obtain environmental data containing spatial associations;

[0019] Record the environmental data containing spatial associations, sort and organize them according to timestamps and / or spatial order, and obtain sorted environmental monitoring data;

[0020] The environmental event is predicted using the prediction model based on the collated environmental monitoring data.

[0021] The sending of the environmental monitoring data of any of the sub-databases to a preset terminal comprises:

[0022] Counting the nodes between the front end and the preset terminal, and obtaining the security value of any of the nodes, and obtaining an optional node for transmission according to the security value of any of the nodes;

[0023] According to the optional nodes for transmission, an optimal transmission path between the front end and the preset terminal is constructed, and encryption is used for transmission between any sending node on the optimal transmission path and its adjacent receiving node to ensure the security of data transmission.

[0024] The counting of nodes between the front end and the preset terminal, and obtaining a security value of any of the nodes, and obtaining an optional node for transmission according to the security value of any of the nodes; and constructing an optimal transmission path between the front end and the preset terminal according to the optional nodes for transmission, and any sending node on the optimal transmission path and its adjacent receiving node are transmitted through encryption, including:

[0025] Using the front end to send test data, and receiving the test data at the preset terminal, selecting nodes on the transmission path according to a preset delay threshold and / or packet loss rate threshold to obtain available nodes;

[0026] Obtaining an interaction behavior of any of the available nodes, and obtaining a trustworthy metric value of a communication behavior environment according to the interaction behavior of the available nodes;

[0027] After weighting the interaction behaviors of the available nodes respectively, a safety value of the available nodes is obtained;

[0028] Selecting a second node connected to the first node according to the communication time; and selecting a third node having the shortest communication time with the first node as the third communication node of the optimal transmission path without considering the second node, until the optimal transmission path is constructed;

[0029] When transmitting between adjacent nodes, the same / different encryption methods are used for data transmission.

[0030] The step of analyzing the attribute features of the historical environment object and the attribute features of the environment object to obtain a feature set spatially associated with the attribute features of the environment object includes:

[0031] Acquiring attribute characteristic data of an environmental object, including: altitude data of the environmental object;

[0032] Obtaining position data corresponding to the environmental object and a Voronoi diagram of the position data;

[0033] The altitude data is used as the weighted data of any location data in the Voronoi diagram, and the process is as follows:

[0034] Obtain the altitude difference between any position data in the position data and its adjacent position data, and find the undulation data;

[0035] Convert the voltage data into weights; ω i =1 / d i p ;ω i is the weight in position i; d i is the volt data; p is the adjustable data;

[0036] Get the region definition of any point s in the weighted Voronoi diagram:

[0037] Among them, d(x,s) is the distance from point x to point s; ω(s) is the weight of point s; d(x,t) is the distance from point x to point t; ω(t) is the weight of point t; a set of positions spatially associated with point s is obtained to form a set of features spatially associated with the attribute features of the environmental object.

[0038] The obtaining of the position data corresponding to the environmental object and the Voronoi diagram of the position data comprises:

[0039] Acquire the position data of the environmental object, set a reference point, and construct a triangle with the three closest points in the position data to form a triangulated network;

[0040] Recording the discrete points constituting any triangle to be measured and the center of the circumscribed circle of the triangle to be measured;

[0041] Traversing the triangulated network to obtain adjacent triangles adjacent to the triangle to be measured;

[0042] Adjacent triangles are connected to the center of the triangle to be measured until a Voronoi diagram for the position data is obtained.

[0043] The performing correlation analysis on any element in the feature set and the sample to obtain spatial correlation includes:

[0044] Storing any element in the feature set in a distributed file system and performing data mining to obtain a mined data set;

[0045] According to the mined data set, setting the dimension of candidate items;

[0046] A mining rule is obtained, and the spatial correlation between any element in the feature set and the sample is acquired according to the mining rule.

[0047] The mining rules are:

[0048] H=H1(C1)AND / OR H2(C2)……H n (C n ), where H is the environmental event that occurred; H n is the nth environmental factor; C is the confidence level of the occurrence of the environmental event.

[0049] The step of adjusting the primary model according to the spatial correlation to obtain a prediction model for predicting the environmental event includes:

[0050] The environmental factors with higher spatial correlation were added to the primary model to obtain the prediction model by grading them according to spatial correlation;

[0051] The environmental factors corresponding to the different levels of spatial correlation are assigned values, and the assigned values ​​corresponding to any environmental factor in the prediction model are superimposed and compared with the set threshold value to predict the occurrence of environmental events.

[0052] An environmental event auxiliary prediction system based on data analysis, using the environmental event auxiliary prediction method as described above, includes:

[0053] A collection module, used to collect environmental events, environmental objects corresponding to the environmental events, attribute feature data of the environmental objects, and environmental data of areas having spatial correlation with the environmental objects;

[0054] A storage module for storing historical environmental event data and prediction models;

[0055] A first processing module performs data interaction with the acquisition module to obtain the environmental objects corresponding to any environmental event and the attribute characteristics of the environmental objects, and obtain the correlation between any environmental object, the attribute characteristics of any environmental object and the environmental event, and construct a primary model;

[0056] A second processing module performs data interaction with the storage module, and is used to collect a preset number of historical environmental event data as samples, and parse the historical environmental objects in the samples and the attribute characteristics of the historical environmental objects; and analyze the attribute characteristics of the historical environmental objects with the attribute characteristics of the environmental objects to obtain a feature set that is spatially associated with the attribute characteristics of the environmental objects;

[0057] A third processing module, which performs data interaction with the second processing module, and is used to perform correlation analysis between any element in the feature set and the sample to obtain spatial correlation;

[0058] a fourth processing module, which performs data interaction with the third processing module and the first processing module, and is used to adjust the primary model according to the spatial correlation, obtain a prediction model for predicting the environmental event, and store the model in the storage module;

[0059] A prediction module, which interacts with the fourth processing module and the acquisition module to obtain a prediction model for predicting the environmental event; and obtains environmental objects in the acquisition module and environmental data of areas where the environmental objects have spatial correlation, and uses the prediction model to predict environmental events;

[0060] The output module performs data interaction with the prediction module and is used to output the prediction result.

[0061] The beneficial effects of the present disclosure are:

[0062] The auxiliary prediction method for environmental events disclosed in the present invention first obtains the environmental objects corresponding to any environmental event and the attribute characteristics of the environmental objects, and obtains the correlation between any of the environmental objects, the attribute characteristics of any of the environmental objects and the environmental events, and constructs a primary model; then collects a preset number of historical environmental event data as samples, and parses the attribute characteristics of the historical environmental objects and the historical environmental objects in the samples; analyzes the attribute characteristics of the historical environmental objects and the attribute characteristics of the environmental objects, and obtains a feature set that is spatially associated with the attribute characteristics of the environmental objects; and, performs a correlation analysis on any element in the feature set and the sample to obtain spatial correlation; adjusts the primary model according to the spatial correlation to obtain a prediction model for predicting the environmental event; finally, simultaneously collects environmental data of the area to be predicted and the area that is spatially correlated with the area to be predicted, screens the environmental data according to the prediction model to reduce the amount of data; and uses the prediction model to predict environmental events in the area to be predicted. The auxiliary prediction method described in the present application utilizes a prediction model to screen the environmental data, thereby reducing the amount of data required for calculations and lowering the hardware requirements for predictions. At the same time, when making predictions, not only the environmental monitoring data of the environmental object itself is taken into account, but also the environmental monitoring data of the area spatially associated with the environmental object is taken into account, thereby ensuring the accuracy of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a flow chart of the environmental event auxiliary prediction method disclosed in the present invention;

[0064] Figure 2 for Figure 1 A flowchart of obtaining a spatial correlation feature set in the flowchart;

[0065] Figure 3To obtain the flowchart of the Voronoi diagram;

[0066] Figure 4 A flowchart for making predictions using a prediction model;

[0067] Figure 5 The flowchart of data processing when making predictions;

[0068] Figure 6 It is a data transmission flow chart during data collection;

[0069] Figure 7 Flowchart for constructing the optimal transmission path;

[0070] Figure 8 This is a system block diagram of the environmental event auxiliary prediction system described in the present invention.

[0071] exist Figure 8 middle:

[0072] 100. Acquisition module; 200. Storage module; 300. First processing module; 400. Second processing module; 500. Third processing module; 600. Fourth processing module; 700. Prediction module; 800. Output module. DETAILED DESCRIPTION

[0073] Those skilled in the art will appreciate that the modules in the device in the implementation scenario can be distributed in the device in the implementation scenario according to the implementation scenario description, or can be changed accordingly and located in one or more devices different from the implementation scenario. The modules in the above implementation scenario can be combined into one module, or can be further split into multiple sub-modules.

[0074] In this embodiment, a debris flow event in mountain area A is used as the predicted environmental event. Specific embodiment 1:

[0076] The present disclosure provides an embodiment:

[0077] like Figure 1, an auxiliary prediction method for environmental events based on data analysis, including: obtaining the environmental objects corresponding to the debris flow event in mountain area A, the attribute characteristics of the environmental objects, and obtaining the correlation between any of the environmental objects, the attribute characteristics of any of the environmental objects and the environmental events, and constructing a primary model; collecting a preset number of historical environmental event data as samples, and parsing the historical environmental objects and the attribute characteristics of the historical environmental objects in the samples; analyzing the attribute characteristics of the historical environmental objects and the attribute characteristics of the environmental objects, and obtaining a feature set that is spatially associated with the attribute characteristics of the environmental objects; performing correlation analysis on any element in the feature set and the sample to obtain spatial correlation; adjusting the primary model according to the spatial correlation to obtain a prediction model for predicting the environmental event; collecting standby environmental monitoring data of the area to be predicted and the area that is spatially correlated with the area to be predicted at the same time, screening the standby environmental monitoring data according to the prediction model to reduce the amount of data and obtain environmental data containing spatial correlation; using the prediction model to predict environmental events in the area to be predicted according to the environmental data containing spatial correlation. In this embodiment, in addition to the data of mountain area A, the data used for prediction also includes environmental data of the area that is spatially correlated with mountain area A.

[0078] In this embodiment, the environmental object corresponding to a debris flow event in a certain place is mountain area A; the attribute characteristics of the environmental object include altitude, rainfall, groundwater level, etc. According to the experience of the engineering personnel, the environmental factors related to the occurrence of debris flow are analyzed and combined with the relevant rough set theory and fuzzy logic methods, and the primary model is constructed and presented in the form of a table, as shown in Table 1.

[0079] Table 1: Primary model of debris flow occurring in a certain place;

[0080]

[0081] In the prior art, the environmental events in the mountain area A are generally predicted by using the correlation factors shown in Table 1 and the spatial position of the mountain area A itself. The present disclosure also fully considers the spatial correlation of environmental events.

[0082] Specifically, Figure 2, the attribute characteristics of the historical environment object are analyzed with the attribute characteristics of the environment object to obtain a feature set that is spatially associated with the attribute characteristics of the environment object, including: obtaining attribute characteristic data of the environment object, including: location data of the environment object, such as altitude data of mountain area A; obtaining longitude and latitude data corresponding to mountain area A, using a Voronoi diagram; using the altitude data as weighted data of any location data in the Voronoi diagram, the process is as follows: obtaining the altitude difference between any location data in the location data and its adjacent location data, and finding the undulation data; terrain undulation refers to the difference between the highest point altitude and the lowest point altitude in a specific area; converting the undulation data into a weight, the larger the undulation data, the smaller the weight; ω i =1 / d i p ;ω i is the weight in position i; d i is the volt data; p is the adjustable data; get the area definition of any point s in the weighted Voronoi diagram:

[0083] Among them, d(x,s) is the distance from point x to point s; ω(s) is the weight of point s; d(x,t) is the distance from point x to point t; ω(t) is the weight of point t; in this Voronoi diagram, the smaller the altitude difference, the greater the weight; the larger the altitude difference, the smaller the weight, and the position set that is spatially associated with point s is obtained to form a spatially associated feature set.

[0084] The Voronoi diagram of the position data corresponding to the environment object is obtained, such as Figure 3 , including: taking the highest altitude point of the mountain area A as a reference point; constructing a triangle with the three nearest points in the position data to form a triangulated network; recording the discrete points constituting the triangle to be measured and the center of the circumscribed circle of the triangle to be measured; traversing the triangulated network to obtain adjacent triangles adjacent to the triangle to be measured; connecting the adjacent triangles with the center of the triangle to be measured; if the triangle to be measured has no adjacent triangles, the edge is the outermost edge of the triangulated network until the Voronoi diagram for the position data is obtained.

[0085] Specifically, the method of performing a correlation analysis on each element in the feature set and the sample to obtain a spatial correlation includes: storing each element in the feature set in a distributed file system, and performing data mining to obtain a mined data set; setting the dimension of candidate items according to the mined data set: when the mined frequent item set is an n-item set, and no candidate item set is generated, then it can be determined that the environmental event is caused by n elements; in this embodiment, n is 3; obtaining a mining rule, and obtaining the spatial correlation between the elements in the feature set and the sample according to the mining rule; n=1, 2, 3, ...

[0086] Preferably, the mining rule is:

[0087] H=H1(C1)AND / OR H2(C2)……H n (C n ), where H is the environmental event that occurred; and C is the confidence level of the environmental event that occurred, as shown in Table 2.

[0088] Table 2: Confidence of historical environmental factors and historical environmental events;

[0089] Association Rules Confidence Difference in altitude between adjacent areas of Mountain A, difference in rainfall between adjacent areas of Mountain A → debris flow in Mountain A 0.85 Rainfall in Mountain A, altitude difference between adjacent areas → debris flow in Mountain A 0.79 Difference in altitude between adjacent areas of Mountain A, height of groundwater level → debris flow occurs in Mountain A 0.3 Difference in altitude between adjacent areas of Mountain A, rainfall in Mountain A → debris flow in Mountain A 0.95

[0090] According to Table 2, the preset confidence level exceeding 0.7 is considered to be an ideal rule; therefore, for mountain area A, when there is “difference in altitude between adjacent areas of mountain area A, difference in rainfall between adjacent areas of mountain area A / rainfall in mountain area A”, debris flow will occur.

[0091] Further, after obtaining the confidence of the occurrence of the environmental event, the primary model is adjusted according to the spatial correlation to obtain a prediction model for predicting the environmental event, such as Figure 4 , including: grading by spatial correlation, adding environmental factors with higher spatial correlation to the primary model to adjust the primary model; sorting the environmental factors in the updated primary model to obtain a prediction model for predicting the environmental event, which is presented in a table; assigning values ​​to environmental factors corresponding to different levels of spatial correlation, and superimposing the assigned values ​​corresponding to any environmental factor in the prediction model, comparing them with the set threshold, and predicting the occurrence of environmental events.

[0092] Specifically, after obtaining the content of Table 2, the primary model is adjusted, as shown in Table 3.

[0093] Table 3: Adjusted primary model;

[0094]

[0095] According to the contents of Table 2, “the altitude difference between adjacent areas of mountain area A, the rainfall difference between adjacent areas of mountain area A, and the rainfall in mountain area A” are all environmental factors that are strongly correlated with debris flow environmental events, while “altitude” and “groundwater level” are weakly correlated environmental factors. Therefore, when assigning values, the strongly correlated environmental factors are assigned a value of 0.5, while the weakly correlated environmental factors are assigned a value of 0.1. The resulting prediction model is: H=b1+b2+b3+b4+b5. When H is greater than 1, it is believed that debris flow will occur in mountain area A.

[0096] In order to ensure the accuracy of prediction, when receiving the environmental data of mountain area A, such as "the altitude of the adjacent areas of mountain area A" and "the rainfall of the adjacent areas of mountain area A", it is also necessary to make a threshold judgment. For environmental factors with high spatial correlation, the difference comparison is performed, such as the value of "the altitude difference of the adjacent areas of mountain area A" and "the rainfall difference of the adjacent areas of mountain area A", and then the threshold is set respectively. If it exceeds the threshold, it is assigned a value, and if it does not exceed the threshold, it is assigned a value of 0; similarly, the threshold judgment is also performed on weakly correlated environmental factors.

[0097] If "the difference in altitude between adjacent areas of mountain area A" meets the threshold requirement, "the difference in rainfall between adjacent areas of mountain area A" does not meet the threshold requirement, "the rainfall in mountain area A" meets the threshold requirement; "the altitude" meets the threshold requirement, "the groundwater level" does not meet the threshold requirement; then H = 0.5 + 0.1 + 0 + 0.5 + 0 = 1.3, the difference in rainfall between mountain area A and adjacent areas is very small, and the rainfall is very large, then mudslides will occur in mountain area A.

[0098] Due to the complexity of environmental data, the unified processing method will cause a lot of energy consumption and high hardware requirements. Figure 5, the environmental data of the area to be predicted is collected, and the environmental events of the area to be predicted are predicted according to the prediction model, including: receiving a prediction task, storing the environmental objects to be predicted and the attribute feature data of the environmental objects to be predicted corresponding to the prediction task in the parent database of the data center; sending a request to the front end to obtain the standby environmental monitoring data of the area to be predicted and the area with spatial correlation with the area to be predicted, and generating a sub-database; sending the environmental monitoring data of any of the sub-databases to a preset terminal, and at the preset terminal, dividing the sub-database into two data modules, respectively obtaining the environmental monitoring data related to the prediction model in the sub-database to clean up useless data, and then combining them to obtain environmental data containing spatial associations; recording the environmental data containing spatial associations, sorting and arranging them according to timestamps and / or spatial order, and obtaining the sorted environmental monitoring data; according to the sorted environmental monitoring data, using the prediction model to predict the environmental events. When in use, since the environmental data is stored in a sub-database, the prediction task can be broken down into specific demand terminals. The demand terminals only read data related to the prediction task from the database center, reducing the data requirements for the database center, thereby reducing resource usage and helping to improve data security.

[0099] When cleaning or screening environmental data, the data is screened according to the spatial correlation factors in Table 3, and the other data are treated as useless data; this can make the data have spatial correlation relationships, and can also reduce a lot of complex data, reduce the total amount of data, reduce the data used in calculations, and reduce hardware requirements; moreover, the sub-database is divided into multiple data modules, and when cleaning data, data cleaning can be performed simultaneously for each data module to improve work efficiency.

[0100] At the same time, in this embodiment, the confidentiality of environmental data is also taken into consideration. In this embodiment, the environmental monitoring data of any of the sub-databases is sent to a preset terminal, such as Figure 6 , including: counting the nodes between the front end and the preset terminal, and obtaining the security value of any of the nodes, and obtaining an optional node for transmission according to the security value of any of the nodes; constructing an optimal transmission path between the front end and the preset terminal according to the optional nodes for transmission, and any sending node on the optimal transmission path and its adjacent receiving node are transmitted through encryption to ensure the security of data transmission.

[0101] Specifically, Figure 7, the front end can be used to send test data, and the test data can be received at the preset terminal, and the nodes on the transmission path can be selected according to the preset delay threshold and / or packet loss rate threshold to obtain available nodes; the interaction behavior of any of the available nodes is obtained, and the trust measurement value of the communication behavior environment is obtained according to the interaction behavior of the available nodes; the interaction behavior of the available nodes is weighted respectively to obtain the security value of the available nodes; according to the communication time, the second node connected to the first node is selected; without considering the second node, the third node with the shortest communication time with the first node is selected as the third communication node of the optimal transmission path, until the construction of the optimal transmission path is completed; when transmitting between adjacent nodes, the same / different encryption methods are used for data transmission.

[0102] Specifically, the interactive behavior of any of the available nodes is obtained; based on the interactive behavior of the available nodes, a trust metric value of the behavioral environment is obtained; and the variance explanation rate is used to perform weight calculation, and the interactive behaviors of the available nodes are weighted separately to obtain the safety value of the available nodes.

[0103] The interaction behavior of any of the available nodes includes: transmission efficiency L a , response efficiency L b , request efficiency L c and forwarding efficiency L d ;

[0104] L a =C s / C a ;

[0105] L b =C RR / C RT ;

[0106] L c =C TA / C TR ;

[0107] L d =C ts / C tt ;

[0108] Among them, C s The total amount of interaction data successfully received by available nodes; C a The total amount of interactive data sent to available nodes; C RR Answer the request for available nodes; C RT The total number of times available node requests are answered; C TA The total number of requests received by available nodes; C T The total number of requests for available nodes; Cts The total number of interactive data forwarded by available nodes; C tt The number of interaction data forwarded by available nodes;

[0109] The security value of the available node includes:

[0110] P = β1P1 + β2P2 + β3P3;

[0111]

[0112] P3=η1L a +η2L b +η3L c +η4L d ;

[0113] Among them, P represents the security value of the available node when data interaction occurs; P1 represents the trust value of the behavior subject; P2 is the trust value of the behavior object; P3 is the trust value of the behavior environment; β1, β2 and β3 are the weight coefficients of P1, P2 and P3 respectively; ∑H(Pi) is the trustworthiness corresponding to the behavior object i; λ is the trust factor; η1, η2, η3 and η4 are the corresponding coefficients of the transmission efficiency, response efficiency, request efficiency and forwarding efficiency of the available nodes respectively.

[0114] The security value of each node is obtained through the security value equation of the available nodes; then, the optimal transmission path between the data transmission end and the data receiving end is constructed according to the optional nodes for transmission.

[0115] After obtaining the security value of the node, this embodiment selects nodes with high security, and then uses the nodes with high security values ​​to build a transmission path, thereby improving transmission security and selecting the optimal solution according to speed, capacity, cost, etc., saving energy and reducing consumption. Specific embodiment 2:

[0117] The present disclosure also provides an embodiment:

[0118] like Figure 8, an environmental event auxiliary prediction system based on data analysis, using the environmental event auxiliary prediction method as described in specific embodiment 1, including: a collection module 100, a storage module 200, a first processing module 300, a second processing module 400, a third processing module 500, a fourth processing module 600, a prediction module 700 and an output module 800; wherein the collection module 100 is used to collect environmental events, environmental objects corresponding to the environmental events, attribute feature data of the environmental objects, and environmental data of areas with spatial correlation with the environmental objects; the storage module 200 is used to store historical environmental event data and prediction models; the first processing module 300 interacts with the collection module 100 for data acquisition, and is used to obtain environmental objects corresponding to any environmental event, attribute features of the environmental objects, and obtain any of the environmental objects, the correlation between the attribute features of any of the environmental objects and the environmental events, and construct a primary model; the second processing module 400 interacts with the storage module 200 for data acquisition, and is used to collect a preset number of historical environmental event data as samples, and parse the samples The historical environment object and the attribute characteristics of the historical environment object are analyzed with the attribute characteristics of the environment object to obtain a feature set that is spatially associated with the attribute characteristics of the environment object; the third processing module 500 interacts with the second processing module 400 for performing correlation analysis on any element in the feature set and the sample to obtain spatial correlation; the fourth processing module 600 interacts with the third processing module 500 and the first processing module 300 for adjusting the primary model according to the spatial correlation to obtain a prediction model for predicting the environmental event and storing it in the storage module; the prediction module 700 interacts with the fourth processing module 600 and the acquisition module 100 for obtaining a prediction model for predicting the environmental event; and obtains the environmental object in the acquisition module and the environmental data of the area where the environmental object has spatial correlation, and uses the prediction model to predict the environmental event; the output module 800 interacts with the prediction module 700 for outputting the prediction result. Specific embodiment 3:

[0120] The present disclosure also provides an embodiment:

[0121] An electronic device comprises: a storage medium for storing a computer program and a processing unit; wherein the processing unit exchanges data with the storage medium, and is used to execute the computer program through the processing unit when predicting environmental events, and perform the steps of the environmental event auxiliary prediction method as described in specific example 1.

[0122] The CPU can perform various appropriate actions and processes according to the program stored in the storage medium. The electronic device also includes the following peripherals, including input parts such as a keyboard and a mouse, and may also include output parts such as a cathode ray tube (CRT), a liquid crystal display (LCD), and a speaker; in particular, according to the embodiments of the present disclosure, Figure 1-Figure 7 Any of the processes described in can be implemented as a computer software program.

[0123] The present disclosure also provides an embodiment:

[0124] A computer-readable storage medium: the computer-readable storage medium stores a computer program; when the computer program is run, the steps of the environmental event auxiliary prediction method as described in specific example 1 are executed.

[0125] In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, device or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. The computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in combination with an instruction execution system, device or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0126] The above disclosures are only several specific implementation scenarios of the present disclosure, but the present disclosure is not limited thereto, and any changes that can be thought of by a person skilled in the art should fall within the protection scope of the present disclosure. The above disclosure serial numbers are only for description and do not represent the advantages and disadvantages of the implementation scenarios.

Claims

1. An auxiliary prediction method for environmental events based on data analysis, characterized in that: include: Obtaining an environmental object corresponding to any environmental event and attribute characteristics of the environmental object, and obtaining any environmental object and the correlation between the attribute characteristics of any environmental object and the environmental event, to construct a primary model; Collecting a preset amount of historical environmental event data as samples, and analyzing historical environmental objects in the samples and attribute characteristics of the historical environmental objects; Analyzing the attribute features of the historical environment object and the attribute features of the environment object to obtain a feature set that is spatially associated with the attribute features of the environment object; Performing correlation analysis on any element in the feature set and the sample to obtain spatial correlation; adjusting the primary model according to the spatial correlation to obtain a prediction model for predicting the environmental event; At the same time, collect the unused environmental monitoring data of the area to be predicted and the area that has spatial correlation with the area to be predicted, filter the unused environmental monitoring data according to the prediction model, reduce the amount of data, and obtain environmental data containing spatial correlation; based on the environmental data containing spatial correlation, use the prediction model to predict environmental events in the area to be predicted.

2. The environmental event auxiliary prediction method based on data analysis according to claim 1 is characterized in that: The step of predicting the environmental events in the area to be predicted by using the prediction model according to the environmental data containing spatial associations includes: Receive a prediction task, and store the corresponding environmental object to be predicted and the attribute feature data of the environmental object to be predicted in the prediction task in a parent database of a data center; Send a request to the front end to obtain the unused environmental monitoring data of the area to be predicted and the area that has spatial correlation with the area to be predicted, and generate a sub-database; Sending the environmental monitoring data of any of the sub-databases to a preset terminal, and dividing the sub-database into at least two data modules at the preset terminal; respectively obtaining the environmental monitoring data related to the prediction model in any of the data modules to clean up useless data, and combining them to obtain environmental data containing spatial associations; Record the environmental data containing spatial associations, sort and organize them according to timestamps and / or spatial order, and obtain sorted environmental monitoring data; The environmental event is predicted using the prediction model based on the collated environmental monitoring data.

3. The environmental event auxiliary prediction method based on data analysis according to claim 2 is characterized in that: The sending of the environmental monitoring data of any of the sub-databases to a preset terminal comprises: Counting the nodes between the front end and the preset terminal, and obtaining the security value of any of the nodes, and obtaining an optional node for transmission according to the security value of any of the nodes; According to the optional nodes for transmission, an optimal transmission path between the front end and the preset terminal is constructed, and encryption is used for transmission between any sending node on the optimal transmission path and its adjacent receiving node to ensure the security of data transmission.

4. The environmental event auxiliary prediction method based on data analysis according to claim 3 is characterized in that: The counting of nodes between the front end and the preset terminal, and obtaining a security value of any of the nodes, and obtaining an optional node for transmission according to the security value of any of the nodes; According to the transmission optional node, an optimal transmission path between the front end and the preset terminal is constructed, and any sending node on the optimal transmission path is transmitted with an adjacent receiving node through encryption, including: Using the front end to send test data, and receiving the test data at the preset terminal, selecting nodes on the transmission path according to a preset delay threshold and / or packet loss rate threshold to obtain available nodes; Obtaining an interaction behavior of any of the available nodes, and obtaining a trustworthy metric value of a communication behavior environment according to the interaction behavior of the available nodes; After weighting the interaction behaviors of the available nodes respectively, a safety value of the available nodes is obtained; Selecting a second node connected to the first node according to the communication time; and selecting a third node having the shortest communication time with the first node as the third communication node of the optimal transmission path without considering the second node, until the optimal transmission path is constructed; When transmitting between adjacent nodes, the same / different encryption methods are used for data transmission.

5. The environmental event auxiliary prediction method based on data analysis according to claim 1 is characterized in that: The step of analyzing the attribute features of the historical environment object and the attribute features of the environment object to obtain a feature set spatially associated with the attribute features of the environment object includes: Acquiring attribute characteristic data of an environmental object, including: altitude data of the environmental object; Obtaining position data corresponding to the environmental object and a Voronoi diagram of the position data; The altitude data is used as the weighted data of any location data in the Voronoi diagram, and the process is as follows: Obtain the altitude difference between any position data in the position data and its adjacent position data, and find the undulation data; converting the volt data into weights; ω i is the weight in position i; d i is the volt data; p is the adjustable data; Get the region definition of any point s in the weighted Voronoi diagram: Among them, d(x,s) is the distance from point x to point s; ω(s) is the weight of point s; d(x,t) is the distance from point x to point t; ω(t) is the weight of point t; a set of positions spatially associated with point s is obtained to form a set of features spatially associated with the attribute features of the environmental object.

6. The environmental event auxiliary prediction method based on data analysis according to claim 5 is characterized in that: The obtaining of the position data corresponding to the environmental object and the Voronoi diagram of the position data comprises: Acquire the position data of the environmental object, set a reference point, and construct a triangle with the three closest points in the position data to form a triangulated network; Recording the discrete points constituting any triangle to be measured and the center of the circumscribed circle of the triangle to be measured; Traversing the triangulated network to obtain adjacent triangles adjacent to the triangle to be measured; Adjacent triangles are connected to the center of the triangle to be measured until a Voronoi diagram for the position data is obtained.

7. The environmental event auxiliary prediction method based on data analysis according to claim 1 is characterized in that: The performing correlation analysis on any element in the feature set and the sample to obtain spatial correlation includes: Storing any element in the feature set in a distributed file system and performing data mining to obtain a mined data set; According to the mined data set, setting the dimension of candidate items; A mining rule is obtained, and the spatial correlation between any element in the feature set and the sample is acquired according to the mining rule.

8. The environmental event auxiliary prediction method based on data analysis according to claim 7 is characterized in that: The mining rules are: H=H1(C1)AND / OR H2(C2)……H n (C n ), where H is the environmental event that occurred; H n is the nth environmental factor; C is the confidence level of the occurrence of the environmental event.

9. The environmental event auxiliary prediction method based on data analysis according to claim 1 is characterized in that: The step of adjusting the primary model according to the spatial correlation to obtain a prediction model for predicting the environmental event includes: The environmental factors with higher spatial correlation were added to the primary model to obtain the prediction model by grading them according to spatial correlation; The environmental factors corresponding to the different levels of spatial correlation are assigned values ​​respectively, and the assigned values ​​corresponding to any environmental factor in the prediction model are superimposed and compared with the set threshold to predict the occurrence of environmental events.

10. An environmental event auxiliary prediction system based on data analysis, using the environmental event auxiliary prediction method according to any one of claims 1 to 9, characterized in that: include: A collection module, used to collect environmental events, environmental objects corresponding to the environmental events, attribute feature data of the environmental objects, and environmental data of areas having spatial correlation with the environmental objects; A storage module for storing historical environmental event data and prediction models; A first processing module performs data interaction with the acquisition module to obtain the environmental objects corresponding to any environmental event and the attribute characteristics of the environmental objects, and obtain the correlation between any environmental object, the attribute characteristics of any environmental object and the environmental event, and construct a primary model; A second processing module performs data interaction with the storage module, and is used to collect a preset number of historical environmental event data as samples, and parse the historical environmental objects in the samples and the attribute characteristics of the historical environmental objects; and analyze the attribute characteristics of the historical environmental objects with the attribute characteristics of the environmental objects to obtain a feature set that is spatially associated with the attribute characteristics of the environmental objects; A third processing module, which performs data interaction with the second processing module, and is used to perform correlation analysis between any element in the feature set and the sample to obtain spatial correlation; a fourth processing module, which performs data interaction with the third processing module and the first processing module, and is used to adjust the primary model according to the spatial correlation, obtain a prediction model for predicting the environmental event, and store the model in the storage module; A prediction module, which interacts with the fourth processing module and the acquisition module to obtain a prediction model for predicting the environmental event; and obtains environmental objects in the acquisition module and environmental data of areas where the environmental objects have spatial correlation, and uses the prediction model to predict environmental events; The output module performs data interaction with the prediction module and is used to output the prediction result.

Images