An intelligent monitoring system based on big data analysis of grassroots smart governance

Abstract

An intelligent monitoring system based on big data analysis of grassroots smart governance relates to the field of big data technology, including a monitoring center, wherein the monitoring center is communicatively connected with a data acquisition module, a regional grid division module, a data monitoring module, a data analysis module, a data prediction module, and a data processing module; multi-source heterogeneous data collected by several Internet of Things nodes are obtained and the collection time is marked, and a collection cycle is set; a plane triangulated network model of the grassroots smart governance area is constructed; environmental data and meteorological data of each triangular grid of the plane triangulated network model are monitored, and alarm information is generated according to the monitoring results; grid inspectors are set in each triangular grid; a grassroots risk warning data prediction model is constructed, and a predicted grassroots risk warning data set of each triangular grid is obtained; early warning operations are performed on each triangular grid and the inspection frequency of inspectors is dynamically updated, thereby significantly improving the grassroots governance level and the ability to respond to emergencies.

Description

Technical Field

[0001] The present invention relates to the field of big data technology, and specifically to an intelligent monitoring system based on big data analysis of grassroots smart governance. Background Art

[0002] The prior art CN116932612A "Intelligent monitoring system based on big data analysis of grassroots smart governance" provides grassroots governments or organizations with data acquisition capabilities in different ways, thereby constructing a system data acquisition module, and then classifies the acquired data into format types and business types for algorithmic analysis and processing, thereby constructing a system data analysis module; utilizing multimodal heterogeneous source data, integrating to form domain knowledge, performing automatic and semi-automatic acquisition of file updates, incremental updates, and processing data assets to asset knowledge to form a system, data relationship association mining, forming a broader and deeper query, and building a data reasoning system, which can discover more data association value and multiple application scenarios, form an extraction base, build multi-level applications, and form multi-directional scenarios based on base applications.

[0003] The prior art CN117634799A "CIM-based smart city grassroots governance event command and dispatch method and system" builds a CIM platform, which includes a CIM model and Internet of Things devices, sensor devices and mobile devices that access the CIM platform through the mobile Internet, wherein the Internet of Things devices, sensor devices and mobile devices all interact with the CIM platform through the MQTT communication protocol; based on the CIM platform, the perception data uploaded by the Internet of Things devices and sensor devices are collected and presented, and various grassroots events in the city operation are discovered; based on the CIM platform, the corresponding event handling process is executed for the event, the event handling method is determined, and the event is sent to the mobile device of the corresponding grassroots event handling department or personnel for event processing, so as to complete the real-time response and command and dispatch of the event.

[0004] At present, when the existing grassroots data monitoring system monitors the grassroots data of several areas, these data are scattered and stored in various monitoring systems, making the data isolated. Without effective data recording and analysis methods, it is difficult to obtain meaningful features from the monitoring, especially when it comes to complex data sets. It is impossible to perform comprehensive data analysis on the data. How to perform comprehensive and intelligent analysis based on multi-source test data is one of the problems that need to be solved at present. Summary of the invention

[0005] In order to solve the above technical problems, the object of the present invention is to provide an intelligent monitoring system based on big data analysis of grassroots smart governance, including a monitoring center, which is communicatively connected with a data acquisition module, a regional grid division module, a data monitoring module, a data analysis module, a data prediction module, and a data processing module;

[0006] The data acquisition module is composed of several IoT nodes and is used to obtain multi-source heterogeneous data collected by several IoT nodes, mark the collection time, and set the collection cycle;

[0007] The regional grid division module is used to obtain the grassroots risk warning data set and construct a plane triangulated network model of the grassroots smart governance area;

[0008] The data monitoring module is used to monitor the environmental data and meteorological data of each triangular mesh of the plane triangulated network model, and generate alarm information according to the monitoring results;

[0009] The data analysis module is used to set grid inspectors in each triangular grid according to the grassroots risk warning data set;

[0010] The data prediction module is used to construct a grassroots risk warning data prediction model to obtain a predicted grassroots risk warning data set for each triangular mesh;

[0011] The data processing module is used to perform early warning operations on each triangular grid and dynamically update the inspection frequency of inspectors according to the predicted grassroots risk warning data set.

[0012] Furthermore, the process of the regional grid division module obtaining the grassroots risk warning data set and constructing the plane triangulation model of the grassroots smart governance area includes:

[0013] Through GIS means, the GIS geographic data, environmental data and meteorological data of the grassroots smart governance area are obtained, and the multi-source heterogeneous data collected by several IoT nodes are obtained at the same time. The grassroots risk warning data set is constructed based on the GIS geographic data, environmental data, meteorological data and multi-source heterogeneous data of the grassroots smart governance area in several historical collection cycles. The terrain characteristics of the grassroots smart governance area are obtained based on the GIS geographic data, and a plane triangulated network model of the grassroots smart governance area is constructed based on the terrain characteristics of the grassroots smart governance area. The terrain characteristics, environmental data, meteorological data and multi-source heterogeneous data of each triangular grid in the plane triangulated network model are given.

[0014] Furthermore, the data monitoring module monitors the environmental data and meteorological data of each triangular mesh of the plane triangulated network model, and the process of generating alarm information according to the monitoring results includes:

[0015] Preset qualified threshold intervals corresponding to various types of environmental monitoring indicators in environmental data and qualified threshold intervals corresponding to various types of meteorological monitoring indicators in meteorological data, compare the numerical time series corresponding to various types of meteorological monitoring indicators in each triangular grid with the corresponding qualified threshold interval, obtain the cumulative time when the numerical time series corresponding to each type of meteorological monitoring indicator is not located in the corresponding qualified threshold interval, and generate alarm information of the meteorological monitoring indicator of the type when the corresponding cumulative time of the meteorological monitoring indicator of the type exists is greater than the preset time threshold;

[0016] Similarly, the numerical time series corresponding to each type of environmental monitoring indicator in each triangular mesh is compared with the corresponding qualified threshold interval, and the cumulative time that the numerical time series corresponding to each type of environmental monitoring indicator is not located in the corresponding qualified threshold interval is obtained. When the corresponding cumulative time of a type of environmental monitoring indicator is greater than the preset time threshold, the alarm information of the type of environmental monitoring indicator is generated.

[0017] Furthermore, the process of the data analysis module setting grid inspectors in each triangular grid according to the grassroots risk warning data set includes:

[0018] Obtain the number of alarm information of each type of environmental monitoring indicator, the number of alarm information of each type of meteorological monitoring indicator, building and infrastructure data, average human flow, and average vehicle flow of each triangular mesh, use the number of alarm information of each type of environmental monitoring indicator, the number of alarm information of each type of meteorological monitoring indicator, building and infrastructure data, average human flow, and average vehicle flow as evaluation indicators, preset an indicator weight matrix of the evaluation indicators, and obtain the membership matrix of each triangular mesh for the preset importance level through fuzzy comprehensive evaluation;

[0019] The importance level of each triangular mesh is obtained according to the membership matrix and the indicator weight matrix, and the importance level of each triangular mesh is compared with the preset importance level threshold. Grid inspectors are set for triangular meshes with importance levels greater than the importance level threshold, and the number of grid inspectors, inspection frequency and inspection range distribution of triangular meshes are obtained according to the coverage range of triangular meshes with importance levels greater than the importance level threshold and the importance level. The grid inspectors are used to feed back manual inspection abnormal data to the monitoring center, and the manual inspection abnormal data includes abnormal problem types and abnormal problem locations.

[0020] Furthermore, the process of obtaining the number of grid inspectors, inspection frequency, and inspection range distribution of each triangular grid includes:

[0021] A patrol requirement comparison table is constructed, wherein the patrol requirement comparison table includes the number of grid patrollers and patrol frequencies corresponding to different importance levels; the number of grid patrollers and patrol frequencies of the triangular mesh are set according to the importance level of the triangular mesh and the patrol requirement comparison table; the coverage of the triangular mesh and the maximum patrol coverage of the grid patroller are obtained at the same time; the comprehensive coverage rate of the grid patrollers is obtained according to the maximum patrol coverage of the grid patrollers, the number of grid patrollers and the coverage of the triangular mesh; the comprehensive coverage rate is used as the fitness function of the particle swarm algorithm; the optimal patrol position of each grid patroller in the triangular mesh is obtained through the particle swarm algorithm; the patrol area of ​​each grid patroller in the triangular mesh is set according to the optimal patrol position and the maximum patrol coverage of each grid patroller.

[0022] Furthermore, the data prediction module constructs a grassroots risk warning data prediction model, and the process of obtaining the predicted grassroots risk warning data set of each triangular mesh includes:

[0023] Construct a grassroots risk warning data prediction model based on deep learning, obtain environmental data, meteorological data and multi-source heterogeneous data in several historical collection cycles of each triangular grid as a training set and a test set, input the training set into the grassroots risk warning data prediction model for training until the loss function training is stable, and save the model parameters, test the grassroots risk warning data prediction model through the test set until it meets the preset requirements, and output the grassroots risk warning data prediction model;

[0024] According to the grassroots risk warning data prediction model, the predicted grassroots risk warning data set of each triangular grid in the current collection period is output, and the predicted grassroots risk warning data set includes the predicted numerical time series sequence corresponding to each type of environmental monitoring indicators in the environmental data, the predicted numerical time series sequence corresponding to each type of meteorological monitoring indicators in the meteorological data, the predicted pedestrian flow time series sequence and the predicted vehicle flow time series sequence.

[0025] Furthermore, the process of the data processing module performing early warning operations on each triangular mesh and dynamically updating the inspection frequency of the inspectors according to the predicted grassroots risk warning data set includes:

[0026] Compare the predicted numerical time series corresponding to each type of environmental monitoring indicator in the environmental data of the triangular grid and the predicted numerical time series corresponding to each type of meteorological monitoring indicator in the meteorological data with the corresponding qualified threshold intervals, and determine whether to generate warning information of a type of meteorological monitoring indicator or a type of environmental monitoring indicator according to the comparison results; if warning information of a type of meteorological monitoring indicator or a type of environmental monitoring indicator is generated, arrange relevant departments to implement grassroots warning measures according to the warning information;

[0027] A feature sharing database is constructed, and the predicted grassroots risk warning data set of the triangular mesh is retrieved and analyzed in the feature sharing database to obtain the probability of occurrence of abnormal data from manual inspections of the triangular mesh, and determine whether the probability of occurrence of the abnormal data from manual inspections is greater than a preset probability threshold. If so, the abnormal problem type and abnormal problem location in the abnormal data from manual inspections are extracted, and the grid inspectors whose inspection areas in the triangular mesh overlap with the abnormal problem locations are obtained, and the grid inspectors are marked as key grid inspectors. The inspection frequency of key grid inspectors is updated according to the abnormal problem type and the probability of occurrence of abnormal data from manual inspections.

[0028] Furthermore, a feature sharing database is constructed, and the predicted grassroots risk warning data set of the triangular mesh is retrieved and analyzed in the feature sharing database. The process of obtaining the probability of occurrence of manual inspection abnormal data of the triangular mesh includes:

[0029] The numerical time series corresponding to each type of environmental monitoring indicator in the environmental data within the acquisition period of the triangular mesh, the numerical time series corresponding to each type of meteorological monitoring indicator in the meteorological data, the time series of human flow, the time series of vehicle flow, and the building and infrastructure data are marked as feature data sets. At the same time, if there is abnormal manual inspection data fed back to the monitoring center by the grid inspector within the acquisition period of the triangular mesh, the abnormal manual inspection data is added to the feature data set, a feature sharing database is constructed, and the feature data sets of each triangular mesh within several acquisition periods are stored in the feature sharing database;

[0030] Perform similarity matching on the predicted grassroots risk warning dataset and the feature dataset, obtain the similarity between each feature dataset in the feature sharing database and the predicted grassroots risk warning dataset, screen out feature datasets with a similarity greater than a preset similarity threshold, mark the feature dataset as a key feature dataset, obtain the number of key feature sets including manual inspection abnormal data in the key feature dataset and the total number of key feature datasets, and obtain the probability of occurrence of manual inspection abnormalities based on the number of key feature sets including manual inspection abnormal data in the key feature dataset and the total number of key feature datasets.

[0031] Compared with the prior art, the present invention has the following beneficial effects:

[0032] 1. Real-time and accuracy: Data can be obtained in real time through IoT nodes, which improves the timeliness and accuracy of data. The data monitoring module can immediately generate alarm information, enabling relevant departments to respond quickly.

[0033] 2. Efficient data processing capabilities: The system is able to process heterogeneous data from different sources and types, which helps to more comprehensively assess risk conditions, build predictive models using historical data and deep learning algorithms, and help decision makers predict possible risks and development trends in advance.

[0034] 3. Dynamic resource allocation: Adjust the work plan of patrol personnel based on forecast data to make resource allocation more reasonable and effective, and achieve refined management through regional grid division modules to ensure that each area has corresponding monitoring and management measures.

[0035] 4. Enhance decision support: Through in-depth analysis of data, it provides a more scientific basis for decision-making and reduces the uncertainty caused by subjective judgment. At the same time, through the data prediction module, the system can identify potential risks in advance and buy time for taking preventive measures.

[0036] To sum up, this system, through its unique architectural design and functional modules, has obvious advantages in improving the real-time nature of data collection, the accuracy of data analysis, the rationality of resource allocation, etc., and can significantly enhance the level of grassroots governance and the ability to respond to emergencies. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a schematic diagram of an intelligent monitoring system based on big data analysis of grassroots smart governance in an embodiment of the present application. DETAILED DESCRIPTION

[0038] The following is a clear and complete description of the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0039] like Figure 1 As shown, an intelligent monitoring system based on big data analysis of grassroots smart governance includes a monitoring center, and the monitoring center is communicatively connected with a data acquisition module, a regional grid division module, a data monitoring module, a data analysis module, a data prediction module, and a data processing module;

[0040] The data acquisition module is composed of several IoT nodes and is used to obtain multi-source heterogeneous data collected by several IoT nodes, mark the collection time, and set the collection cycle;

[0041] The regional grid division module is used to obtain the grassroots risk warning data set and construct a plane triangulated network model of the grassroots smart governance area;

[0042] The data monitoring module is used to monitor the environmental data and meteorological data of each triangular mesh of the plane triangulated network model, and generate alarm information according to the monitoring results;

[0043] The data analysis module is used to set grid inspectors in each triangular grid according to the grassroots risk warning data set;

[0044] The data prediction module is used to construct a grassroots risk warning data prediction model to obtain a predicted grassroots risk warning data set for each triangular mesh;

[0045] The data processing module is used to perform early warning operations on each triangular grid and dynamically update the inspection frequency of inspectors according to the predicted grassroots risk warning data set.

[0046] It should be further explained that, in the specific implementation process, the process of the regional grid division module obtaining the grassroots risk warning data set and constructing the plane triangulation model of the grassroots smart governance area includes:

[0047] The GIS geographic data, environmental data and meteorological data of the grassroots smart governance area are obtained by GIS means, and the multi-source heterogeneous data collected by several IoT nodes are obtained at the same time. The grassroots risk warning data set is constructed according to the GIS geographic data, environmental data, meteorological data and multi-source heterogeneous data of the grassroots smart governance area in several historical collection periods. The terrain characteristics of the grassroots smart governance area are obtained according to the GIS geographic data. The plane triangulated network model of the grassroots smart governance area is constructed according to the terrain characteristics of the grassroots smart governance area, and the terrain characteristics, environmental data, meteorological data and multi-source heterogeneous data of each triangular mesh in the plane triangulated network model are given; wherein the terrain characteristics include plains, mountains, rivers, hills, deserts, coastlines, etc., the meteorological data include meteorological monitoring indicators such as temperature, humidity, wind speed, precipitation, etc., the environmental data include environmental monitoring indicators such as the chemical composition, turbidity, pH value of water bodies such as rivers, lakes and oceans, and the concentration of pollutants in the air, such as PM2.5, sulfur dioxide, nitrogen oxides and other environmental monitoring indicators, and the multi-source heterogeneous data include building and infrastructure data, pedestrian flow data and vehicle flow data.

[0048] It should be further explained that, in the specific implementation process, the data monitoring module monitors the environmental data and meteorological data of each triangular mesh of the plane triangulated network model, and the process of generating alarm information according to the monitoring results includes:

[0049] Preset qualified threshold intervals corresponding to various types of environmental monitoring indicators in environmental data and qualified threshold intervals corresponding to various types of meteorological monitoring indicators in meteorological data, compare the numerical time series corresponding to various types of meteorological monitoring indicators in each triangular grid with the corresponding qualified threshold interval, obtain the cumulative time when the numerical time series corresponding to each type of meteorological monitoring indicator is not located in the corresponding qualified threshold interval, and generate alarm information of the meteorological monitoring indicator of the type when the corresponding cumulative time of the meteorological monitoring indicator of the type exists is greater than the preset time threshold;

[0050] Similarly, the numerical time series corresponding to each type of environmental monitoring indicator in each triangular mesh is compared with the corresponding qualified threshold interval, and the cumulative time that the numerical time series corresponding to each type of environmental monitoring indicator is not located in the corresponding qualified threshold interval is obtained. When the corresponding cumulative time of a type of environmental monitoring indicator is greater than the preset time threshold, the alarm information of the type of environmental monitoring indicator is generated.

[0051] It should be further explained that, in the specific implementation process, the process of the data analysis module setting grid inspectors in each triangular grid according to the grassroots risk warning data set includes:

[0052] According to the grassroots risk warning data set, the number of alarm information of each type of environmental monitoring indicator, the number of alarm information of each type of meteorological monitoring indicator, building and infrastructure data, average pedestrian flow, and average vehicle flow of each triangular grid are obtained, and the number of alarm information of each type of environmental monitoring indicator, the number of alarm information of each type of meteorological monitoring indicator, building and infrastructure data, average pedestrian flow, and average vehicle flow are used as evaluation indicators, and the indicator weight matrix of the evaluation indicator is preset, and the membership matrix of each triangular grid for the preset importance level is obtained through fuzzy comprehensive evaluation;

[0053] Among them, the building and infrastructure data includes the building and infrastructure data type and the number of each type of building and infrastructure. Since the building and infrastructure data type is text data, the building and infrastructure data type is converted as follows, so that the data size represents the importance of the building and infrastructure data type. The larger the value, the more important the building and infrastructure data type is. For example, residential = 0.7, commercial building = 0.8, industrial building = 0.6, public facilities = 0.9, transportation infrastructure = 0.9, water conservancy facilities = 0.8;

[0054] The importance level of each triangular mesh is obtained according to the membership matrix and the indicator weight matrix, and the importance level of each triangular mesh is compared with the preset importance level threshold. Grid inspectors are set for triangular meshes with importance levels greater than the importance level threshold, and the number of grid inspectors, inspection frequency and inspection range distribution of triangular meshes are obtained according to the coverage range of triangular meshes with importance levels greater than the importance level threshold and the importance level. The grid inspectors are used to feed back manual inspection abnormal data to the monitoring center, and the manual inspection abnormal data includes abnormal problem types and abnormal problem locations.

[0055] It needs to be further explained that, in the specific implementation process, the process of grid inspectors feeding back manual patrol abnormal data to the monitoring center is: use the camera function to take photos of the locations where abnormal problems exist and upload them, describe the abnormal problems in text, clearly explain the specific problems and classify the types of abnormal problems. The abnormal problem types include environmental sanitation abnormal data (for example, garbage cleaning: check whether the garbage bin is overflowing and whether the garbage is correctly classified; public facility cleaning: whether park seats, public toilets, etc. are clean and tidy; road cleaning: observe whether there is littering), public safety abnormal data (for example, security patrols: pay attention to whether there are suspicious persons or activities; fire hazards: check whether the fire passages are unobstructed and whether the fire facilities are intact; traffic safety: observe whether the traffic signs and markings are clear, whether there is illegal parking, etc.) and building and infrastructure abnormal data (for example, housing safety: check whether there are safety hazards in old houses; power facilities: observe whether facilities such as telephone poles and transformers are normal; water supply network: check whether there are leaks in water pipes).

[0056] It should be further explained that, in the specific implementation process, the process of obtaining the importance level of each triangular mesh according to the membership matrix and the index weight matrix includes:

[0057] The indicator weight matrix and the membership matrix of the evaluation indicator are integrated through a formula to obtain a fuzzy comprehensive evaluation matrix of the evaluation indicator, the membership of each triangular mesh for different importance levels is obtained according to the fuzzy comprehensive evaluation matrix, the importance level with the highest membership corresponding to each triangular mesh is screened out, and the importance level with the highest membership corresponding to each triangular mesh is used as the importance level of each triangular mesh;

[0058] Wherein, the formula is: ;

[0059] in, is the fuzzy comprehensive evaluation matrix of the evaluation index, is the indicator weight matrix of the evaluation indicator, is the membership matrix, " represents the multiplication of the weight matrix of the evaluation index and the elements at the corresponding positions of the membership matrix, It is a weighted parameter for controlling the balance between the weight matrix and the membership matrix in the fuzzy comprehensive evaluation matrix of the evaluation index.

[0060] It should be further explained that, in the specific implementation process, the process of obtaining the number of grid inspectors, inspection frequency, and inspection range distribution of each triangular grid includes:

[0061] Construct an inspection requirement comparison table, the inspection requirement comparison table includes the number of grid inspectors and inspection frequencies corresponding to different importance levels, set the number of grid inspectors and inspection frequencies of the triangular mesh according to the importance level of the triangular mesh and the inspection requirement comparison table, and simultaneously obtain the coverage of the triangular mesh and the maximum inspection coverage of the grid inspectors, obtain the comprehensive coverage rate of the grid inspectors according to the maximum inspection coverage of the grid inspectors, the number of grid inspectors, and the coverage of the triangular mesh, use the comprehensive coverage rate as the fitness function of the particle swarm algorithm, obtain the best inspection position of each grid inspector in the triangular mesh through the particle swarm algorithm, and set the inspection area of ​​each grid inspector in the triangular mesh according to the best inspection position and the maximum inspection coverage of each grid inspector;

[0062] Among them, the calculation formula for obtaining the comprehensive coverage rate of grid inspectors based on the maximum patrol coverage of grid inspectors, the number of grid inspectors, and the coverage of triangular grids is:

[0063] ;

[0064] in, represents the comprehensive coverage rate of the grid inspector z, S represents the coverage of the triangular grid, Indicates the maximum patrol coverage of the grid inspector. represents the distance from the grid inspector to the geometric center b of the triangle grid, Indicates the conversion factor;

[0065] Then define the objective function. In this embodiment, the objective function is the maximum comprehensive coverage rate. Determine the size of the particle swarm, randomly generate the initial position and speed for each particle (i.e., the position of each grid inspector in the triangular grid), set the parameters, and then perform the optimization process. After multiple iterations, determine the optimal solution, i.e., the optimal position of the group gbest, which represents the optimal layout position of each grid inspector. The calculation formula of the optimization process is:

[0066]

[0067]

[0068] in, represents the inertia weight; , Represents the acceleration coefficient, which is used to weight the optimal position of the particle itself and the optimal position of the group; Indicates the current iteration of particle i speed; Indicates the current iteration of particle i location; , is a random number between 0 and 1; represents the optimal position of particle i.

[0069] It should be further explained that, in the specific implementation process, the data prediction module constructs a grassroots risk warning data prediction model, and the process of obtaining the predicted grassroots risk warning data set of each triangular mesh includes:

[0070] Construct a grassroots risk warning data prediction model based on deep learning, obtain environmental data, meteorological data and multi-source heterogeneous data in several historical collection cycles of each triangular grid as a training set and a test set, input the training set into the grassroots risk warning data prediction model for training until the loss function training is stable, and save the model parameters, test the grassroots risk warning data prediction model through the test set until it meets the preset requirements, and output the grassroots risk warning data prediction model;

[0071] According to the grassroots risk warning data prediction model, the predicted grassroots risk warning data set of each triangular grid in the current collection period is output, and the predicted grassroots risk warning data set includes the predicted numerical time series sequence corresponding to each type of environmental monitoring indicators in the environmental data, the predicted numerical time series sequence corresponding to each type of meteorological monitoring indicators in the meteorological data, the predicted pedestrian flow time series sequence and the predicted vehicle flow time series sequence.

[0072] It should be further explained that, in the specific implementation process, the process in which the data processing module performs early warning operations on each triangular grid and dynamically updates the inspection frequency of the inspectors according to the predicted grassroots risk warning data set includes:

[0073] Compare the predicted numerical time series corresponding to each type of environmental monitoring indicator in the environmental data of the triangular grid and the predicted numerical time series corresponding to each type of meteorological monitoring indicator in the meteorological data with the corresponding qualified threshold intervals, and determine whether to generate warning information of a type of meteorological monitoring indicator or a type of environmental monitoring indicator according to the comparison results; if warning information of a type of meteorological monitoring indicator or a type of environmental monitoring indicator is generated, arrange relevant departments to implement grassroots warning measures according to the warning information;

[0074] A feature sharing database is constructed, and the predicted grassroots risk warning data set of the triangular mesh is retrieved and analyzed in the feature sharing database to obtain the probability of occurrence of abnormal data from manual inspections of the triangular mesh, and determine whether the probability of occurrence of the abnormal data from manual inspections is greater than a preset probability threshold. If so, the abnormal problem type and abnormal problem location in the abnormal data from manual inspections are extracted, and the grid inspectors whose inspection areas in the triangular mesh overlap with the abnormal problem locations are obtained, and the grid inspectors are marked as key grid inspectors. The inspection frequency of key grid inspectors is updated according to the abnormal problem type and the probability of occurrence of abnormal data from manual inspections.

[0075] It should be further explained that, in the specific implementation process, the calculation formula for updating the inspection frequency of key grid inspectors according to the abnormal problem type and the probability of occurrence of abnormal data in manual inspection is:

[0076] ;

[0077] in, represents the new inspection frequency of key grid inspector z, represents the original inspection frequency of key grid inspector z, It indicates the probability of abnormal data in manual inspection. Represents the baseline adjustment factor for the abnormal problem type r.

[0078] It should be further explained that, in the specific implementation process, a feature sharing database is constructed, and the predicted grassroots risk warning data set of the triangular mesh is retrieved and analyzed in the feature sharing database, and the process of obtaining the probability of occurrence of manual inspection abnormal data of the triangular mesh includes:

[0079] The numerical time series corresponding to each type of environmental monitoring indicator in the environmental data within the acquisition period of the triangular mesh, the numerical time series corresponding to each type of meteorological monitoring indicator in the meteorological data, the time series of human flow, the time series of vehicle flow, and the building and infrastructure data are marked as feature data sets. At the same time, if there is abnormal manual inspection data fed back to the monitoring center by the grid inspector within the acquisition period of the triangular mesh, the abnormal manual inspection data is added to the feature data set, a feature sharing database is constructed, and the feature data sets of each triangular mesh within several acquisition periods are stored in the feature sharing database;

[0080] Perform similarity matching on the predicted grassroots risk warning data set and the feature data set, obtain the similarity between each feature data set in the feature sharing database and the predicted grassroots risk warning data set, screen out feature data sets with similarity greater than a preset similarity threshold, mark the feature data sets as key feature data sets, obtain the number of key feature sets including manual patrol abnormal data in the key feature data set and the total number of key feature data sets, and obtain the probability of manual patrol abnormality according to the number of key feature sets including manual patrol abnormal data in the key feature data set and the total number of key feature data sets;

[0081] Among them, the formula for obtaining the similarity between each feature data set in the feature sharing database and the predicted grassroots risk warning data set is:

[0082] ;

[0083] Among them, q represents the similarity between the feature dataset and the predicted grassroots risk warning dataset; Represents the value of the p-th type of data in the feature data set at the t-th time; Represents the average value of the p-th type of data in the feature data set; It indicates the value of the p-th type of data in the predicted grassroots risk warning data set at the tth moment; It represents the average value of the pth type of data in the predicted grassroots risk warning data set; n represents the total number of moments in the time series; M represents a set of data types, including various types of environmental monitoring indicators in environmental data, various types of meteorological monitoring indicators in meteorological data, pedestrian flow, vehicle flow, and abnormal problem types in manual inspection abnormal data.

[0084] Among them, the formula for obtaining the probability of manual inspection anomaly occurrence based on the number of key feature sets including manual inspection anomaly data in the key feature data set and the total number of key feature data sets is:

[0085] ;

[0086] in, Indicates the number of key feature sets in the key feature data set that include manual inspection abnormal data, The total number of key feature datasets, Indicates the conversion factor.

[0087] The above embodiments are only used to illustrate the technical method of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical method of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical method of the present invention.

Claims

1. An intelligent monitoring system based on big data analysis of grassroots smart governance, characterized in that: It includes a monitoring center, which is communicatively connected with a data acquisition module, a regional grid division module, a data monitoring module, a data analysis module, a data prediction module, and a data processing module; The data acquisition module is composed of several IoT nodes and is used to obtain multi-source heterogeneous data collected by several IoT nodes, mark the collection time, and set the collection cycle; The regional grid division module is used to obtain the grassroots risk warning data set and construct a plane triangulated network model of the grassroots smart governance area, including: Obtain GIS geographic data, environmental data, and meteorological data of grassroots smart governance areas through GIS means, and simultaneously obtain multi-source heterogeneous data collected by several IoT nodes. Construct a grassroots risk warning dataset based on the GIS geographic data, environmental data, meteorological data, and multi-source heterogeneous data of grassroots smart governance areas in several historical collection cycles. Obtain the terrain characteristics of grassroots smart governance areas based on GIS geographic data. Construct a plane triangulated network model of the grassroots smart governance areas based on the terrain characteristics of the grassroots smart governance areas, and assign terrain characteristics, environmental data, meteorological data, and multi-source heterogeneous data to each triangular mesh in the plane triangulated network model. The data monitoring module is used to monitor the environmental data and meteorological data of each triangular mesh of the plane triangulated network model, and generate alarm information according to the monitoring results, including: Preset qualified threshold intervals corresponding to various types of environmental monitoring indicators in environmental data and qualified threshold intervals corresponding to various types of meteorological monitoring indicators in meteorological data, compare the numerical time series corresponding to various types of meteorological monitoring indicators in each triangular grid with the corresponding qualified threshold interval, obtain the cumulative time when the numerical time series corresponding to each type of meteorological monitoring indicator is not located in the corresponding qualified threshold interval, and generate alarm information of the meteorological monitoring indicator of the type when the corresponding cumulative time of the meteorological monitoring indicator of the type exists is greater than the preset time threshold; Similarly, the numerical time series corresponding to each type of environmental monitoring indicator in each triangular mesh is compared with the corresponding qualified threshold interval, and the cumulative time that the numerical time series corresponding to each type of environmental monitoring indicator is not located in the corresponding qualified threshold interval is obtained. When the corresponding cumulative time of a type of environmental monitoring indicator is greater than the preset time threshold, the alarm information of the type of environmental monitoring indicator is generated; The data analysis module is used to set grid inspectors in each triangular grid according to the grassroots risk warning data set, including: Obtain the number of alarm information of each type of environmental monitoring indicator, the number of alarm information of each type of meteorological monitoring indicator, building and infrastructure data, average human flow, and average vehicle flow of each triangular mesh, use the number of alarm information of each type of environmental monitoring indicator, the number of alarm information of each type of meteorological monitoring indicator, building and infrastructure data, average human flow, and average vehicle flow as evaluation indicators, preset an indicator weight matrix of the evaluation indicators, and obtain the membership matrix of each triangular mesh for the preset importance level through fuzzy comprehensive evaluation; Obtain the importance level of each triangular mesh according to the membership matrix and the indicator weight matrix, compare the importance level of each triangular mesh with the preset importance level threshold, set a grid inspector for the triangular mesh with an importance level greater than the importance level threshold, and obtain the number, inspection frequency and inspection range distribution of the grid inspectors of the triangular mesh according to the coverage of the triangular mesh with an importance level greater than the importance level threshold in combination with the importance level. The grid inspectors are used to feed back manual inspection abnormal data to the monitoring center, and the manual inspection abnormal data includes abnormal problem type and abnormal problem location; The process of obtaining the number of grid inspectors, inspection frequency, and inspection range distribution of each triangular grid includes: Construct an inspection requirement comparison table, the inspection requirement comparison table includes the number of grid inspectors and inspection frequencies corresponding to different importance levels, set the number of grid inspectors and inspection frequencies of the triangular mesh according to the importance level of the triangular mesh and the inspection requirement comparison table, and simultaneously obtain the coverage of the triangular mesh and the maximum inspection coverage of the grid inspectors, obtain the comprehensive coverage rate of the grid inspectors according to the maximum inspection coverage of the grid inspectors, the number of grid inspectors, and the coverage of the triangular mesh, use the comprehensive coverage rate as the fitness function of the particle swarm algorithm, obtain the best inspection position of each grid inspector in the triangular mesh through the particle swarm algorithm, and set the inspection area of ​​each grid inspector in the triangular mesh according to the best inspection position and the maximum inspection coverage of each grid inspector; The data prediction module is used to construct a grassroots risk warning data prediction model to obtain a predicted grassroots risk warning data set for each triangular mesh, including: Construct a grassroots risk warning data prediction model based on deep learning, obtain environmental data, meteorological data and multi-source heterogeneous data in several historical collection cycles of each triangular grid as a training set and a test set, input the training set into the grassroots risk warning data prediction model for training until the loss function training is stable, and save the model parameters, test the grassroots risk warning data prediction model through the test set until it meets the preset requirements, and output the grassroots risk warning data prediction model; Output the predicted grassroots risk warning data set of each triangular grid in the current collection period according to the grassroots risk warning data prediction model, wherein the predicted grassroots risk warning data set includes the predicted numerical time series sequence corresponding to each type of environmental monitoring indicator in the environmental data, the predicted numerical time series sequence corresponding to each type of meteorological monitoring indicator in the meteorological data, the predicted pedestrian flow time series sequence and the predicted vehicle flow time series sequence; The data processing module is used to perform early warning operations on each triangular grid and dynamically update the inspection frequency of the inspectors according to the predicted grassroots risk warning data set, including: Compare the predicted numerical time series corresponding to each type of environmental monitoring indicator in the environmental data of the triangular grid and the predicted numerical time series corresponding to each type of meteorological monitoring indicator in the meteorological data with the corresponding qualified threshold intervals, and determine whether to generate warning information of a type of meteorological monitoring indicator or a type of environmental monitoring indicator according to the comparison results; if warning information of a type of meteorological monitoring indicator or a type of environmental monitoring indicator is generated, arrange relevant departments to implement grassroots warning measures according to the warning information; Construct a feature sharing database, retrieve and analyze the predicted grassroots risk warning data set of the triangular mesh in the feature sharing database, obtain the probability of occurrence of abnormal data of manual inspection of the triangular mesh, and determine whether the probability of occurrence of abnormal data of manual inspection is greater than a preset probability threshold. If so, extract the abnormal problem type and abnormal problem location in the abnormal data of manual inspection, obtain the grid inspector whose inspection area overlaps with the abnormal problem location in the triangular mesh, mark the grid inspector as a key grid inspector, and update the inspection frequency of the key grid inspector according to the abnormal problem type and the probability of occurrence of abnormal data of manual inspection; Among them, the process of constructing a feature sharing database, searching and analyzing the predicted grassroots risk warning data set of the triangular mesh in the feature sharing database, and obtaining the probability of occurrence of manual inspection abnormal data of the triangular mesh includes: The numerical time series corresponding to each type of environmental monitoring indicator in the environmental data within the acquisition period of the triangular mesh, the numerical time series corresponding to each type of meteorological monitoring indicator in the meteorological data, the time series of human flow, the time series of vehicle flow, and the building and infrastructure data are marked as feature data sets. At the same time, if there is abnormal manual inspection data fed back to the monitoring center by the grid inspector within the acquisition period of the triangular mesh, the abnormal manual inspection data is added to the feature data set, a feature sharing database is constructed, and the feature data sets of each triangular mesh within several acquisition periods are stored in the feature sharing database; Perform similarity matching on the predicted grassroots risk warning dataset and the feature dataset, obtain the similarity between each feature dataset in the feature sharing database and the predicted grassroots risk warning dataset, screen out feature datasets with a similarity greater than a preset similarity threshold, mark the feature dataset as a key feature dataset, obtain the number of key feature sets including manual inspection abnormal data in the key feature dataset and the total number of key feature datasets, and obtain the probability of occurrence of manual inspection abnormalities based on the number of key feature sets including manual inspection abnormal data in the key feature dataset and the total number of key feature datasets.

Images