Grid governance event management method and system based on event circulation mechanism and medium

Abstract

The invention provides a grid governance event management method and system based on an event circulation mechanism, and a medium, and relates to the technical field of data processing, and the method comprises the steps: arranging edge grid nodes for a plurality of grid regions in a target governance region; after receiving and analyzing the first task scheduling strategy uploaded by the first edge grid node, the grid governance holder judges resource scheduling conflicts of concurrent events and calls out M conflict scheduling strategies of M grid governance events; and carrying out resource allocation optimization on the first task scheduling strategy and the M conflict scheduling strategies based on the event processing priority, outputting M + 1 optimized scheduling strategies, and carrying out processing task cross-department distribution. The technical problems that in the prior art, due to the fact that resource allocation cannot be dynamically adjusted, resource scheduling conflicts are likely to be generated under the condition of multi-task concurrence, the event treatment efficiency is low, and the response speed is low are solved, and the technical effects of improving the event treatment efficiency and enhancing the collaborative treatment capacity are achieved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and specifically to a grid governance event management method, system and medium based on an event flow mechanism. Background Art

[0002] As a refined management model, the grid governance event management method has been widely used in urban communities, grassroots governance and other fields. Its core is to divide the city into multiple grid units and use information technology and management methods to achieve rapid response and efficient processing of various events. Existing methods mostly rely on fixed resource scheduling rules or simple scheduling algorithms. All task scheduling and resource allocation are performed by a centralized control platform. This architecture lacks flexible dynamic adjustment capabilities when dealing with large-scale, multi-task concurrent events. When multiple event tasks require the same resources at the same time, resource scheduling conflicts are prone to occur due to improper resource allocation, resulting in inefficient event processing, slow response speed, and even task delays or governance failures, seriously affecting the overall governance effect. Summary of the invention

[0003] The present application provides a grid governance event management method, system and medium based on an event flow mechanism, which solves the technical problems that the prior art cannot dynamically adjust resource allocation due to resource scheduling relying on fixed rules or simple scheduling algorithms, and is prone to resource scheduling conflicts in the case of multiple tasks concurrently, resulting in low event governance efficiency and slow response speed. It achieves the technical effect of optimizing resource allocation, thereby improving event processing efficiency and enhancing cross-departmental collaborative governance capabilities.

[0004] In view of the above problems, on the one hand, the present application provides a grid governance event management method based on an event flow mechanism, the method comprising: arranging multiple edge grid nodes for multiple grid areas in the target governance area, wherein the multiple edge grid nodes are connected to a grid governance cloud platform; after the grid governance cloud platform receives a first task scheduling policy uploaded by a first edge grid node, it performs resource scheduling conflict judgment of concurrent events by parsing the first task scheduling policy, and calls out M conflict scheduling policies of M grid governance events, wherein the M conflict scheduling policies are uploaded by M edge grid nodes, and the first task scheduling policy has a first grid governance event identifier; the grid governance cloud platform optimizes resource allocation of the first task scheduling policy and the M conflict scheduling policies based on event processing priority, and outputs M+1 optimized scheduling policies; the grid governance cloud platform distributes the processing tasks of the M+1 optimized scheduling policies across departments based on the event flow mechanism, so as to perform event parallel governance of the first grid governance event and the M grid governance events.

[0005] On the other hand, the present application also provides a grid governance event management system based on an event flow mechanism, the system comprising: a grid node setting module, used to arrange multiple edge grid nodes for multiple grid areas in the target governance area, wherein the multiple edge grid nodes are connected to the grid governance cloud platform; a scheduling conflict judgment module, used to perform resource scheduling conflict judgment of concurrent events by parsing the first task scheduling strategy after the grid governance cloud platform receives the first task scheduling strategy uploaded by the first edge grid node, and call out M conflict scheduling strategies for M grid governance events, wherein the M conflict scheduling strategies are uploaded by M edge grid nodes, and the first task scheduling strategy has a first grid governance event identifier; a resource allocation optimization module, used to use the grid governance cloud platform to optimize the resource allocation of the first task scheduling strategy and the M conflict scheduling strategies based on the event processing priority, and output M+1 optimized scheduling strategies; a task distribution module, used to use the grid governance cloud platform to distribute the processing tasks of the M+1 optimized scheduling strategies across departments based on the event flow mechanism, so as to perform event parallel governance of the first grid governance event and the M grid governance events.

[0006] In a third aspect, the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned grid governance event management method based on the event flow mechanism are implemented.

[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: By deploying multiple edge grid nodes in the target governance area and connecting them to the grid governance cloud platform, a distributed governance framework is constructed, which improves the real-time performance of event management, reduces data transmission delay, and can perform partial data processing and decision-making locally, thereby effectively improving the event response speed. After receiving the task scheduling strategy uploaded by the first edge grid node, the grid governance cloud platform concurrently processes the event and makes resource conflict judgments. By parsing the task scheduling strategy, it is possible to judge the resource conflict problem in the case of multiple concurrent tasks, avoiding the resource competition problem caused by static rules in traditional methods, thereby improving the resource utilization efficiency and the coordination of governance tasks. The grid governance cloud platform optimizes the task scheduling strategy based on the priority of event processing, improves the resource utilization efficiency through priority sorting, and ensures that key tasks are given priority in emergency or complex situations, thereby greatly improving the event processing efficiency. Based on the event flow mechanism, the grid governance cloud platform distributes the optimized scheduling strategy across departments, realizes the collaborative work of different departments or grid nodes, enables multiple tasks to be processed in parallel, and improves the collaborative ability and processing speed of each event processing department.

[0008] In summary, this application achieves efficient resource scheduling and conflict management in a multi-task concurrent environment by deploying a distributed processing mechanism on multiple edge grid nodes and combining the intelligent scheduling and resource optimization of the grid governance cloud platform. In multiple concurrent scenarios, it can respond quickly and allocate resources reasonably, effectively improve the real-time and collaborative capabilities of event processing, significantly improve the efficiency of grid governance event management, enhance cross-departmental collaboration capabilities, and optimize resource utilization, thereby achieving a more efficient, flexible, and real-time event governance system.

[0009] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 A flow chart of a grid governance event management method based on an event flow mechanism provided in an embodiment of the present application.

[0011] Figure 2 In the grid governance event management method based on the event flow mechanism provided in the embodiment of the present application, task resource scheduling analysis is performed on the first registration event, and a flow chart of the first task scheduling strategy is output.

[0012] Figure 3 A flow chart of performing backtracking verification on a first registration event and outputting a first event verification result in a grid governance event management method based on an event flow mechanism provided in an embodiment of the present application.

[0013] Figure 4 A structural diagram of a grid governance event management system based on an event flow mechanism provided in an embodiment of the present application.

[0014] Description of reference numerals: grid node setting module 10 , scheduling conflict judgment module 20 , resource allocation optimization module 30 , task distribution module 40 . DETAILED DESCRIPTION

[0015] The embodiments of the present application provide a grid governance event management method, system and medium based on an event flow mechanism, thereby solving the technical problems in the prior art that resource scheduling relies on fixed rules or simple scheduling algorithms, cannot dynamically adjust resource allocation, is prone to resource scheduling conflicts in the case of multiple tasks concurrently, and leads to low event governance efficiency and slow response speed. The technical effect of optimizing resource allocation, thereby improving event processing efficiency and enhancing cross-departmental collaborative governance capabilities is achieved.

[0016] Embodiment 1, as Figure 1As shown, the embodiment of the present application provides a grid governance event management method based on an event flow mechanism, the method comprising: Step S1: Arrange multiple edge grid nodes for multiple grid areas in the target management area, wherein the multiple edge grid nodes are connected to the grid management cloud platform.

[0017] Specifically, the target governance area refers to the area that needs to be managed and governed, which can be a certain area of ​​the city, a community or a specific geographical unit. In multiple grid areas of the target governance area, multiple edge grid nodes are deployed, and one edge grid node is deployed for each grid area. These edge grid nodes are devices or computing nodes distributed in each grid area, such as smart sensors, cameras, handheld terminals, etc., which are used to collect event data in real time and perform preliminary processing. The edge grid nodes are connected to the grid governance cloud platform through network communication (such as 5G, NB-IoT, Wi-Fi, etc.) to form a distributed data collection and processing network. Among them, the grid governance cloud platform, as the core management platform, is responsible for receiving, parsing, scheduling and optimizing event processing tasks, and coordinating communications between edge grid nodes and management departments.

[0018] By deploying edge grid nodes within the grid area, real-time collection and preliminary processing of event data are achieved, providing basic data support for subsequent resource scheduling and event processing.

[0019] Step S2: After the grid governance cloud platform receives the first task scheduling policy uploaded by the first edge grid node, it performs resource scheduling conflict judgment of concurrent events by parsing the first task scheduling policy, and calls out M conflict scheduling policies of M grid governance events, wherein the M conflict scheduling policies are uploaded by M edge grid nodes, and the first task scheduling policy has a first grid governance event identifier.

[0020] Specifically, the first edge grid node may be any one of the multiple edge grid nodes in step S1, corresponding to the first grid area. When the first edge grid node detects the local (first grid area) event and generates the first task scheduling strategy, it is uploaded to the grid governance cloud platform. After receiving this strategy, the grid governance cloud platform uses a parsing algorithm (for example, a rule-based parsing algorithm or a machine learning algorithm to parse various parameters and identifiers in the strategy) to parse the first task scheduling strategy. During the parsing process, the grid governance cloud platform compares and analyzes this strategy with the M conflicting scheduling strategies uploaded by the other M edge grid nodes to determine whether there is a resource scheduling conflict. The resources here may include manpower, material resources (such as emergency rescue equipment), etc. Among them, M is a positive integer, which represents the number of grid areas where events occur simultaneously with the first grid area. Exemplarily, the first edge grid node uploads a task scheduling strategy to request the call of clean vehicle resources in a certain area. At the same time, the other M edge grid nodes also upload similar scheduling strategies to request clean vehicle resources in the same area. The grid governance cloud platform parses these strategies and finds that multiple strategies request the same batch of clean vehicle resources, which means there is a resource scheduling conflict.

[0021] Through centralized task scheduling strategy analysis and conflict judgment, potential resource scheduling problems can be discovered in advance, avoiding inefficient processing caused by resource conflicts during event processing.

[0022] Step S3: The grid governance cloud platform optimizes resource allocation for the first task scheduling strategy and M conflict scheduling strategies based on event processing priority, and outputs M+1 optimized scheduling strategies.

[0023] Specifically, after identifying resource scheduling conflicts, the grid governance cloud platform uses resource allocation optimization algorithms (such as resource allocation algorithms based on linear programming, genetic algorithms, etc.) to adjust the first task scheduling strategy and M conflict scheduling strategies based on the event processing priority (this priority can be pre-set based on factors such as the urgency and harm of the event). By reallocating resources (such as adjusting the number of manpower and materials allocated to different events), M+1 optimized scheduling strategies are output.

[0024] Through resource allocation optimization, each event can receive reasonable resource support according to its priority, thus improving the efficiency of overall event processing.

[0025] Step S4: The grid governance cloud platform distributes the processing tasks of the M+1 optimization scheduling strategies across departments based on the event flow mechanism to perform parallel event governance of the first grid governance event and the M grid governance events.

[0026] Specifically, the grid governance cloud platform uses the event flow mechanism (this mechanism can be set up based on workflow engine technology, defining the flow rules and processes of events between different departments) to distribute the processing tasks corresponding to M+1 optimized scheduling strategies across departments, and distribute each processing task and the corresponding optimized scheduling strategy to the designated management department or responsible person. For example, according to the event type and processing process, illegal parking incidents are transferred to the traffic management department, while environmental sanitation issues are transferred to the sanitation department. Through cross-departmental task distribution, parallel governance of multiple events is achieved, which improves the ability of the entire grid governance system to deal with complex events, and also promotes collaborative work between different departments.

[0027] Furthermore, before the grid governance cloud platform receives the first task scheduling strategy uploaded by the first edge grid node in step S2, the following steps are also included: Step S21: After receiving the first registration event, the first edge mesh node identifies the registration subject of the first registration event to obtain a first registration subject, wherein the first edge mesh node is an edge mesh node configured in a first mesh area among the multiple edge mesh nodes.

[0028] Step S22: When the first registration subject is a non-grid member, perform task resource scheduling analysis on the first registration event and output the first task scheduling strategy.

[0029] Step S23: When the first registration subject is a grid worker, after retrospective verification of the first registration event, the first task scheduling strategy is output through task resource scheduling analysis.

[0030] Specifically, a registration event is event information that needs to be processed when it is detected or received by an edge grid node. These events can come from various sources, such as camera intelligent monitoring, sensor monitoring, and grid worker handheld terminal upload. The registration subject is the initiator or source identifier associated with the registration event, which is divided into two types: grid worker and non-grid worker. The first edge grid node is located in the first grid area and monitors and receives various event information in the first grid area. When the first edge grid node receives the first registration event, it first identifies the registration subject of the event, that is, identifies the source of the event to determine whether it is a grid worker or other subject. The identification process can be performed by querying and judging the identity information attached to the event (such as user account information and device number contained in the event information), or distinguishing according to the specific channel of the event source (such as a dedicated reporting channel for grid workers, a sensor feedback channel, etc.). Clarifying the registration subject helps to adopt different processing procedures according to different subjects in the future, and improve the accuracy and pertinence of event processing.

[0031] When it is determined that the first registrant is a non-grid member, the first edge grid node needs to perform task resource scheduling analysis on the first registration event, including: analyzing the type of event (such as environmental sanitation, safety hazard, etc.), the severity of the event (minor, medium, serious, etc.) and the resources required to handle the event (manpower, material resources, etc.), and then generating the first task scheduling strategy based on the event type, urgency and resource requirements. For example, based on a pre-set event classification and resource requirement mapping table, combined with the specific circumstances of the event, an algorithm (such as a rule-based decision algorithm) can be used to determine the resource scheduling strategy, and finally output the first task scheduling strategy.

[0032] When the first registration subject is a grid member, since the grid member's report depends on the subjective judgment of the person, it is easy to make judgment errors. It is necessary to perform a retrospective verification in combination with the historical records of monitoring equipment (such as cameras, smart sensors) to determine the authenticity of the event. For example, a grid member found an environmental sanitation problem during the inspection and reported it through a handheld terminal. The authenticity of the event was confirmed by comparing the grid member's historical records and on-site photos. Then, the task resource scheduling analysis is performed based on the retrospective verification results. Similar to step S22, the required resources are determined based on the current status, type, severity, etc. of the event, and the first task scheduling strategy is output. Through retrospective verification, it can ensure that the events reported by the grid member are true and reliable, avoid false alarms and repeated reports, and improve the accuracy and efficiency of event processing.

[0033] The above steps S21 to S24 effectively improve the responsiveness and processing efficiency of the grid governance platform through intelligent event registration, task scheduling and resource optimization. By distinguishing and processing events from different sources (reported by grid members and those reported by non-grid members), the efficiency and accuracy of event processing can be ensured.

[0034] Further, such as Figure 2 As shown, step S22 includes: Step S221: interactively obtain multiple groups of sample resource allocation strategies for multiple sample governance events, wherein each sample resource allocation strategy includes a main resource scheduling requirement and F collaborative resource scheduling requirements.

[0035] Step S222: The plurality of sample governance events and the plurality of groups of sample resource allocation strategies are stored in association with each other based on the knowledge graph to complete the data filling of the event flow information base.

[0036] Step S223: The first edge grid node uses the first registration event to traverse the event flow information library to perform task resource scheduling and matching, and obtains a first set of sample resource allocation strategies.

[0037] Step S224: Packing the first group of sample resource allocation strategies and outputting them as the first task scheduling strategy.

[0038] Specifically, sample governance events are representative grid governance event cases collected in advance. These cases are used to build a reference model for event processing. For example, in urban traffic grid governance, past traffic accident handling cases and traffic congestion relief cases can all be used as sample governance events. The sample resource allocation strategy is a resource allocation plan formulated for each sample governance event, including the main resource scheduling requirements (i.e., the main resource requirements for event processing, such as manpower, equipment, etc.) and F collaborative resource scheduling requirements (auxiliary resource requirements that cooperate with the main resources, such as technical support, material supply, etc.). Among them, F is a positive integer, and the value of F is different according to the different collaborative resource scheduling requirements required for different sample governance events. The first edge grid node obtains multiple groups of sample resource allocation strategies for multiple sample governance events by interacting with other devices or systems (such as a central database, other edge nodes, or a dedicated sample collection system). This interaction process can use network communication protocols (such as HTTP protocols, etc.) for data transmission. For example, in a city's community grid governance system, the first edge grid node obtains sample resource allocation strategies for various events in the community in the past (such as community facility maintenance, garbage cleaning, etc.) by interacting with the big data center of the city grid governance. Obtaining multiple sets of sample resource allocation strategies provides rich references for subsequent event processing and helps to formulate task scheduling strategies more accurately.

[0039] The first edge grid node uses knowledge graph technology to associate and store multiple sample governance events and multiple groups of sample resource allocation strategies. For example, through natural language processing tools, the resource type, management department, and the demand relationship between events and resources, and the collaborative relationship between events and processing departments are identified from sample governance events and sample resource allocation strategies. Then, a graph database (such as Neo4j, etc.) is used to build a knowledge graph, with sample governance events, resource types, and management departments as nodes, and the demand relationship between events and resources (such as "Event X requires resource Y"), and the collaborative relationship between events and processing departments (such as "Event X is jointly handled by the transportation department and the emergency department") as the relationship between nodes. The constructed knowledge graph is filled into the event flow information library. This event flow information library is a data warehouse that stores the flow process, processing process, and resource scheduling strategy of events. It contains detailed records of the entire process of each event from registration to processing to reporting, which is used to support task resource scheduling matching.

[0040] The first edge grid node uses the first received registration event as the query condition and traverses the event flow information library. During the traversal process, the event type, severity and other characteristics are matched with the sample governance events in the library to find multiple sample resource allocation strategies corresponding to the most similar sample governance events, that is, the first set of sample resource allocation strategies. This matching process can use a similarity algorithm (such as the cosine similarity algorithm, etc.) to measure the similarity between events. By traversing the event flow information library for task resource scheduling matching, we can quickly draw on the experience of handling similar events in the past and formulate a more reasonable task scheduling strategy.

[0041] The first set of matched sample resource allocation strategies are packaged, and the related resource allocation scheduling strategies are combined to form a unified format, which is then output to the next processing link as the first task scheduling strategy. For example, the information of the main resource scheduling requirements and the collaborative resource scheduling requirements is arranged in a certain order, and the necessary identification information is added to form a task scheduling strategy that can be recognized by the grid governance cloud platform.

[0042] The above steps S221 to S224 optimize the task scheduling process by introducing sample event resource allocation strategies and knowledge graph technology. By extracting resource scheduling strategies from historical events and matching them with new events, a reasonable and efficient resource allocation plan can be provided for each event, thereby improving the response speed and processing efficiency of event processing.

[0043] Further, step S23 includes: Step S231: When the first registration subject is a grid worker, extract the first event occurrence location and the first event timestamp from the first registration event.

[0044] Step S232: performing a retrospective check on the first registered event with the first event occurrence location and the first event timestamp as constraints, and outputting a first event verification result.

[0045] Step S233: If the first event verification result is passed, perform task resource scheduling analysis on the first registration event and output the first task scheduling strategy.

[0046] Specifically, when it is determined that the first registration subject is a grid worker, the first edge grid node will extract the location of the first event and the timestamp of the first event from the relevant information of the first registration event. The location of the event refers to the specific geographical coordinates or area identifier where the registration event occurs. The location of the event can be obtained by parsing the geographical positioning information attached when the grid worker reports the event (for example, the coordinates obtained by GPS positioning); the event timestamp is an identifier that records the time when the registration event occurs, and the event timestamp can be directly obtained from the time record field of the reported event.

[0047] The first registered event is retrospectively verified with the extracted first event location and first event timestamp as constraints. Retrospective verification can verify whether the event is repeated or contradictory by querying and comparing historical event records, or by verifying the on-site situation of the event through image recognition, video surveillance and other means, and judging whether the event has passed the verification based on spatiotemporal information and historical data. For example, with the location of the event as a constraint, query whether similar events have occurred at the location before; with the event timestamp as a constraint, check whether other related events have occurred near the time. The first event verification result is output based on these query results, and the verification result may be passed or failed.

[0048] If the first event verification result is passed, a task resource scheduling analysis is performed on the first registered event. Based on factors such as the type of event (such as infrastructure damage) and severity (such as the loss of a sewage manhole cover being a medium safety hazard), the resources required to handle the event are determined, and then the first task scheduling strategy is output.

[0049] Through step S231 to step S233, the processing flow of grid workers' registration events is further refined. By extracting the location and timestamp of the event, accurate spatiotemporal information can be provided for retrospective verification; the authenticity and reliability of the event are ensured through retrospective verification; and reasonable scheduling strategies are generated through task resource scheduling analysis. This process not only improves the accuracy and reliability of event processing, but also improves the efficiency of event processing through intelligent resource scheduling, ensuring that the events reported by grid workers can be efficiently processed.

[0050] Further, such as Figure 3 As shown, step S232 includes: Step S232-1: Positioning a first surveillance camera in the first grid area according to the location where the first event occurred.

[0051] Step S232-2: Perform video backtracking on the first surveillance camera according to the first event timestamp to obtain a first video frame sequence.

[0052] Step S232-3: Output a first identified event by performing governance event recognition on the first video frame sequence.

[0053] Step S232-4: Perform consistency verification on the first registration event and the first identification event, and output the first event verification result.

[0054] Specifically, when performing a retrospective verification, based on the location of the first event extracted from the first registered event, with the help of a pre-established grid area surveillance camera layout diagram (which can be a map data containing information such as camera location coordinates and monitoring range), the surveillance camera is located in the first grid area to determine the first surveillance camera located near the location of the event. For example, if the location of the first event indicates that the event occurred near a street light pole on a certain street, then by querying the layout diagram, the first surveillance camera that can cover the street light pole area is found. Accurately locating the first surveillance camera provides a data source for subsequent video retrospection, which helps to further verify the authenticity of the registered event.

[0055] According to the first event timestamp, a video backtracking instruction is sent to the first surveillance camera. After receiving the instruction, the first surveillance camera extracts the video data of the time period near the first event from the local storage (if the camera has its own storage function) or the storage device connected to the camera (such as a network video recorder, NVR) to form a first video frame sequence. The extraction process can query the video file with the first event timestamp as the time index, and extract video frames at a certain frame rate (for example, 25 frames per second). For example, if the time feature of the first event is 9:30~9:40 in the morning, the first surveillance camera will extract 25 frames of video data per second in this time period from its storage device to form the first video frame sequence.

[0056] Use image recognition technology (such as a target detection algorithm based on deep learning, such as the YOLO algorithm) to identify governance events on the first video frame sequence, and look for targets or behaviors related to the first registration event in the video frame. For example, if the registration event is illegal parking in a certain area, then detect whether the vehicle is illegally parked in the video frame. Output the first recognition event based on the recognition result. The first recognition event is an event situation related to governance identified from the video frame sequence. By identifying governance events on the video frame sequence, the event situation based on video evidence can be obtained, providing comparative data for subsequent consistency verification.

[0057] Compare the first registration event and the first identification event for consistency check. For example, compare whether the event type (such as both are about specific behaviors in a certain area), the subject of the event (such as both involve specific people or vehicles), the severity of the event, etc. are consistent. If they are basically consistent in all aspects, then the first event verification result is output as passed; if there are large differences, then the output is failed. Consistency check can ensure that the registration events reported by grid workers are consistent with the actual situation, improve the accuracy of event information, and thus provide a reliable basis for subsequent resource scheduling and other operations.

[0058] Through steps S232-1 to S232-4, the retrospective verification process of grid worker registration events is further refined. By positioning surveillance cameras, video retrospection, event identification and consistency verification, the authenticity and accuracy of events can be efficiently verified, thereby improving the reliability of event processing.

[0059] Further, step S232-3 includes: Step 1: Interactively obtain multiple sample time image sets of the multiple sample governance events.

[0060] Step 2: After extracting identification features from the multiple sample time image sets, construct multiple governance event identification channels based on the extraction results.

[0061] Step three: Complete the construction of the first edge event identification network in the first edge grid node by connecting the multiple governance event identification channels in parallel.

[0062] Step 4: Input the first video frame sequence into the first edge event recognition network, perform governance event recognition through the multiple governance event recognition channels, and output multiple alternative recognition event sequences.

[0063] Step 5: Output the first recognition event by performing time aggregation on the multiple candidate recognition event sequences.

[0064] Specifically, when identifying governance events, we first interact with other devices or systems (such as sample data storage servers, other edge grid nodes, etc.) to obtain multiple sample time image sets of multiple sample governance events. These sample time image sets are image sets related to multiple sample governance events and arranged in chronological order. They are used to build a governance event identification channel. The set contains various image features that may appear at different times for different sample governance events that have been collected and sorted in advance, such as image sets of traffic events, community events, etc. in different scenarios. Obtaining sample time image sets provides a data basis for building a governance event identification channel, and rich sample images help improve the accuracy of identification.

[0065] Use image feature extraction algorithms (such as feature extraction layer algorithms in convolutional neural networks) to extract identification features from multiple sample time image sets. These features will serve as the basis for subsequent model training. Next, multiple governance event identification channels are constructed based on the extracted features. The construction of each governance event identification channel is based on the clustering and abstraction of features related to a certain type of governance event, and is specifically used to identify a specific type of governance event. For example, for community governance scenarios, features are extracted from sample time image sets containing sample governance events such as community environmental sanitation and community facility damage, and environmental sanitation event identification channels are constructed based on environmental sanitation-related features, and facility damage event identification channels are constructed based on facility damage-related features. These channels are trained through deep learning frameworks (such as TensorFlow or PyTorch), and the model parameters are optimized using labeled sample image data to ensure that each channel can accurately identify its corresponding event type, thereby achieving efficient and accurate identification of governance events.

[0066] The multiple governance event identification channels that have been constructed are operated in parallel, that is, a parallel data processing process is set up at the software level, so that each governance event identification channel can independently process the input data, and through the parallel connection of multiple governance event identification channels, the first edge event identification network in the first edge grid node is formed. This parallel structure can be regarded as a parallel computing architecture, similar to multiple processors processing different tasks at the same time. The first edge event identification network can process multiple types of governance event identification at the same time, which improves the overall event identification speed and processing capabilities.

[0067] Input the first video frame sequence into the constructed first edge event recognition network. Inside the network, the first video frame sequence will flow through multiple governance event recognition channels at the same time, and each channel will perform governance event recognition operations on the video frame sequence according to its own recognition rules and algorithms. Then, each channel will output an alternative recognition event sequence, which contains possible governance event information identified under each channel. By identifying the first video frame sequence in parallel through multiple governance event recognition channels, a variety of possible governance event information can be fully obtained, reducing the possibility of missing important events.

[0068] Governance time aggregation is performed on multiple candidate recognition event sequences, and each candidate recognition event is integrated according to factors such as the time sequence of the events and the types of events. For example, multiple related events occurring in the same time period and at the same location are merged into a comprehensive event description, and the first recognition event is finally output. This event is an integrated and optimized governance event description. For example, the alternative recognition event sequence output by the traffic accident recognition channel shows that a vehicle collision occurred at a certain intersection between 9:00 and 9:10 in the morning, and the alternative recognition event sequence output by the traffic congestion recognition channel shows that traffic congestion occurred at the intersection between 9:05 and 9:20 in the morning. Through governance time aggregation, the first recognition event output is traffic congestion caused by a vehicle collision at a certain intersection between 9:00 and 9:20 in the morning. Governance time aggregation can integrate multiple scattered alternative recognition event sequences into a complete and accurate first recognition event, improving the accuracy and practicality of event recognition.

[0069] The above steps 1 to 5 use a parallel multi-channel recognition network to efficiently process the video frame sequence, extract multi-dimensional event information, and output the final recognized event through a time aggregation method, which greatly improves the efficiency and accuracy of event recognition while ensuring the comprehensiveness and consistency of event recognition.

[0070] Further, step S3 includes: Step S31: interactively obtain the event governance priority sequence of the multiple sample governance events.

[0071] Step S32: parsing the M conflicting scheduling strategies to obtain M groups of conflicting resource allocation strategies.

[0072] Step S33: Taking the event governance priority sequence as a constraint, perform a policy selection conflict analysis on the first group of sample resource allocation strategies and the M groups of conflicting resource allocation strategies, and output the M+1 optimized scheduling strategies.

[0073] Specifically, the event governance priority sequence is the result of sorting multiple sample governance events according to factors such as importance or urgency. The event governance priority sequence of multiple sample governance events is obtained by interacting with other devices or systems (such as sample governance event database, central management system, etc.). The event governance priority sequence provides an important basis for the subsequent processing of resource allocation conflicts and helps to reasonably allocate resources to deal with events of different urgency.

[0074] Parse M conflicting scheduling strategies, analyze and organize the resource requirements (such as manpower, material resources, equipment, etc.) in multiple scheduling strategies, and extract M groups of conflicting resource allocation strategies. For example, by parsing the text description or data structure in the scheduling strategy, the information about resource allocation is extracted to form an independent conflicting resource allocation strategy group. Parsing the conflicting scheduling strategy to obtain the conflicting resource allocation strategy clarifies the specific content of the resource allocation conflict and provides accurate input for subsequent conflict analysis.

[0075] Taking the obtained event governance priority sequence as the constraint condition, the first group of sample resource allocation strategies and the M groups of conflicting resource allocation strategies are analyzed for policy selection conflicts. The event priorities corresponding to different strategies are compared, and the resource allocation strategies corresponding to high-priority events are prioritized. The resource allocation strategies for low-priority events are adjusted or abandoned according to the actual situation, and finally M+1 optimized scheduling strategies are output. For example, if the priority of the power facility emergency repair event is higher than that of the network line maintenance event, in the policy selection conflict analysis, the resource allocation strategy for the power facility emergency repair is prioritized, and the resource allocation strategy for the network line maintenance is replanned. Through the policy selection conflict analysis, the optimized scheduling strategy is output, which can reasonably allocate resources according to the priority of the event when resources are limited, improve resource utilization efficiency, and ensure that high-priority events are handled in a timely manner.

[0076] Steps S31 to S33 interactively obtain event management priority sequences, parse conflicting resource allocation strategies, and conduct policy selection conflict analysis, which can dynamically optimize resource allocation, generate reasonable scheduling strategies, improve the scientific nature of resource allocation and management efficiency, and ensure the efficiency and reliability of event processing.

[0077] Furthermore, in step S4, the grid governance cloud platform distributes the processing tasks of the M+1 optimization scheduling strategies across departments based on the event flow mechanism, and then further includes: Step S41: configuring M+1 governance time frames for the M+1 optimization scheduling strategies.

[0078] Step S42: interactively obtain multiple sets of standard processing progress stage identifiers for the multiple sample governance events.

[0079] Step S43: In the grid governance cloud platform, with the M+1 governance time frames as constraints, the processing progress of the first grid governance event and the M grid governance events is visualized according to the multiple sets of standard processing progress stage identifiers.

[0080] Specifically, the governance time frame is the time range set for each optimization scheduling strategy, which clarifies the estimated time interval from the start to the end of the governance event corresponding to the strategy. In the grid governance cloud platform, for M+1 optimization scheduling strategies, the corresponding governance time frame is configured for each strategy based on factors such as the nature of the event, resource allocation, and previous experience in handling similar events. Configuring the governance time frame can give clear time expectations for the handling of each governance event, facilitate further allocation and management of resources, and also help monitor whether the event is handled on time.

[0081] The grid governance cloud platform obtains multiple sets of standard processing progress stage identifiers for multiple sample governance events by interacting with other systems (such as sample governance event management systems, historical data repositories, etc.). This data interaction can use data interfaces or network communication protocols (such as RESTful APIs, message queues, etc.) to achieve data transmission. These standard processing progress stage identifiers define key nodes in the event processing process, such as "event reception", "resource allocation", "on-site processing" and "completion feedback" stages, providing a reference standard for the subsequent processing progress visualization, and can measure and display event processing progress with a unified standard.

[0082] With M+1 governance time frames as constraints, the processing progress of the first grid governance event and the M grid governance events is visualized according to multiple sets of standard processing progress stage identifiers. This can be achieved using visualization tools (such as dashboards, Gantt charts, etc.). For example, in the Gantt chart, the governance time frame is used as the horizontal axis and the standard processing progress stage identifier is used as the vertical axis. By marking the processing progress stage reached by each event at different time points, the progress of event processing can be intuitively displayed. Processing progress visualization allows managers and relevant personnel to intuitively understand the progress of governance events, facilitate timely identification of problems and adjustments, and improve the management efficiency of governance events.

[0083] By configuring the governance time frame and progress visualization, it is possible to achieve refined management of the governance event handling process, allowing managers to intuitively understand the processing progress of each event and adjust resource allocation and processing strategies in a timely manner, thereby improving event governance efficiency and response speed.

[0084] In summary, the grid governance event management method based on the event flow mechanism provided in the embodiment of the present application has the following technical effects: By arranging multiple edge grid nodes in multiple grid areas of the target governance area and networking with the grid governance cloud platform, a distributed data collection and preliminary processing network is constructed, which can collect event data in the grid in real time and upload the data to the cloud platform, providing basic data support for subsequent resource scheduling and event processing. The grid governance cloud platform receives and parses the task scheduling strategy, identifies and understands the resource requirements and event priority of each grid node, and provides a basis for subsequent resource scheduling conflict judgment. The cloud platform analyzes the task scheduling strategy, calls out multiple conflict scheduling strategies, identifies the resource conflicts that may occur in multi-task concurrent scenarios, and ensures that subsequent resource allocation can be optimized for these conflicts. Based on the event processing priority, the task scheduling strategy is optimized for resource allocation to ensure that high-priority events can obtain sufficient resource support while reducing resource waste and conflicts. Through the event flow mechanism, the optimized scheduling strategy is distributed to relevant departments to achieve parallel governance of events, break the information islands between departments, achieve efficient cross-departmental collaboration, and improve overall governance efficiency.

[0085] Overall, the embodiment of the present application forms a complete set of intelligent grid governance event management processes by arranging edge grid nodes, analyzing task scheduling strategies, judging resource conflicts, optimizing resource allocation, and using event flow mechanisms for cross-departmental distribution. This solution not only optimizes resource allocation and improves event processing efficiency, but also enhances cross-departmental collaborative governance capabilities, significantly improves the refinement level and overall effectiveness of grid governance, and provides efficient and scientific decision-making support for complex grid governance events.

[0086] Embodiment 2, as Figure 4 As shown, based on the same inventive concept as the above-mentioned embodiment 1, the embodiment of the present application provides a grid governance event management system based on an event flow mechanism, and the system includes: The grid node setting module 10 is used to arrange multiple edge grid nodes for multiple grid areas in the target management area, wherein the multiple edge grid nodes are connected to the grid management cloud platform.

[0087] The scheduling conflict judgment module 20 is used to judge the resource scheduling conflicts of concurrent events by parsing the first task scheduling strategy after the grid governance cloud platform receives the first task scheduling strategy uploaded by the first edge grid node, and call out M conflict scheduling strategies for M grid governance events, wherein the M conflict scheduling strategies are uploaded by M edge grid nodes, and the first task scheduling strategy has a first grid governance event identifier.

[0088] The resource allocation optimization module 30 is used to optimize the resource allocation of the first task scheduling strategy and M conflict scheduling strategies based on event processing priority using the grid governance cloud platform, and output M+1 optimized scheduling strategies.

[0089] The task distribution module 40 is used to utilize the grid governance cloud platform to distribute the processing tasks of the M+1 optimization scheduling strategies across departments based on the event flow mechanism, so as to perform event parallel governance of the first grid governance event and M grid governance events.

[0090] Furthermore, the scheduling conflict determination module 20 in the embodiment of the present application is also used to perform the following steps: After receiving the first registration event, the first edge grid node identifies the registration subject of the first registration event to obtain the first registration subject, wherein the first edge grid node is an edge grid node configured in the first grid area among the multiple edge grid nodes; when the first registration subject is a non-grid member, a task resource scheduling analysis is performed on the first registration event, and the first task scheduling strategy is output; when the first registration subject is a grid member, after performing a backtracking check on the first registration event, the first task scheduling strategy is output through task resource scheduling analysis.

[0091] Furthermore, the scheduling conflict determination module 20 in the embodiment of the present application is also used to perform the following steps: Interactively obtain multiple groups of sample resource allocation strategies for multiple sample governance events, wherein each sample resource allocation strategy includes a main resource scheduling requirement and F collaborative resource scheduling requirements; based on the knowledge graph, the multiple sample governance events and the multiple groups of sample resource allocation strategies are associated and stored to complete the data filling of the event flow information library; the first edge grid node uses the first registration event to traverse the event flow information library for task resource scheduling matching to obtain a first group of sample resource allocation strategies; and the first group of sample resource allocation strategies are packaged as the first task scheduling strategy output.

[0092] Furthermore, the scheduling conflict determination module 20 in the embodiment of the present application is also used to perform the following steps: In the case where the first registration subject is a grid worker, the first event occurrence location and the first event timestamp are extracted from the first registration event; the first registration event is retrospectively verified with the first event occurrence location and the first event timestamp as constraints, and the first event verification result is output; if the first event verification result is passed, the first registration event is subjected to task resource scheduling analysis, and the first task scheduling strategy is output.

[0093] Furthermore, the scheduling conflict determination module 20 in the embodiment of the present application is also used to perform the following steps: The first surveillance camera is located in the first grid area according to the location where the first event occurred; the video of the first surveillance camera is traced back according to the first event timestamp to obtain a first video frame sequence; a first recognition event is output by performing governance event recognition on the first video frame sequence; the first registration event and the first recognition event are consistency verified, and the first event verification result is output.

[0094] Furthermore, the scheduling conflict determination module 20 in the embodiment of the present application is also used to perform the following steps: Interactively obtain multiple sample time image sets of the multiple sample governance events; extract identification features from the multiple sample time image sets, and then construct multiple governance event identification channels based on the extraction results; complete the construction of the first edge event identification network in the first edge grid node by connecting the multiple governance event identification channels in parallel; input the first video frame sequence into the first edge event identification network, perform governance event identification via the multiple governance event identification channels, and output multiple alternative identification event sequences; output the first identification event by performing governance time aggregation on the multiple alternative identification event sequences.

[0095] Furthermore, the resource allocation optimization module 30 in the embodiment of the present application is also used to perform the following steps: Interactively obtain the event governance priority sequence of the multiple sample governance events; parse the M conflict scheduling strategies to obtain M groups of conflicting resource allocation strategies; based on the event governance priority sequence as a constraint, perform strategy selection conflict analysis on the first group of sample resource allocation strategies and the M groups of conflicting resource allocation strategies, and output the M+1 optimized scheduling strategies.

[0096] Furthermore, the task distribution module 40 in the embodiment of the present application is also used to perform the following steps: Configure M+1 governance time frames for the M+1 optimized scheduling strategies; interactively obtain multiple sets of standard processing progress stage identifiers for the multiple sample governance events; and on the grid governance cloud platform, use the M+1 governance time frames as constraints to visualize the processing progress of the first grid governance event and the M grid governance events according to the multiple sets of standard processing progress stage identifiers.

[0097] Through the above-mentioned detailed description of the grid governance event management method based on the event flow mechanism in this specification, those skilled in the art can clearly understand the grid governance event management system based on the event flow mechanism in this embodiment. For the system disclosed in Example 2, since it corresponds to the method disclosed in Example 1 and has corresponding functional modules and beneficial effects, the relevant parts can be referred to the description of the method part.

[0098] Embodiment three, based on the same inventive concept of the grid governance event management method based on the event flow mechanism in the aforementioned embodiment one, the present application also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by the processor, the various steps of the above-mentioned embodiment of the grid governance event management method based on the event flow mechanism are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0099] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A grid governance event management method based on event flow mechanism, characterized in that: The method comprises: Arranging a plurality of edge grid nodes for a plurality of grid areas in a target governance area, wherein the plurality of edge grid nodes are connected to a grid governance cloud platform; When the first registration subject of the received first registration event is a grid member, the first edge grid node performs a backtracking verification with the first event occurrence location and the first event timestamp of the first registration event as constraints, and outputs a first task scheduling strategy according to the verification result; after the grid governance cloud platform receives the first task scheduling strategy uploaded by the first edge grid node, it performs resource scheduling conflict judgment of concurrent events by parsing the first task scheduling strategy, and calls out M conflict scheduling strategies of M grid governance events, wherein the M conflict scheduling strategies are uploaded by the M edge grid nodes, and the first task scheduling strategy has a first grid governance event identifier; The grid governance cloud platform optimizes resource allocation for the first task scheduling strategy and M conflict scheduling strategies based on event processing priority, and outputs M+1 optimized scheduling strategies; The grid governance cloud platform distributes the processing tasks of the M+1 optimization scheduling strategies across departments based on the event flow mechanism to perform parallel event governance of the first grid governance event and the M grid governance events.

2. The grid governance event management method based on the event flow mechanism as claimed in claim 1 is characterized in that: Before the grid governance cloud platform receives the first task scheduling strategy uploaded by the first edge grid node, the method further includes: After receiving the first registration event, the first edge mesh node identifies the registration subject of the first registration event to obtain the first registration subject, wherein the first edge mesh node is an edge mesh node configured in the first mesh area among the plurality of edge mesh nodes; In the case where the first registration subject is a non-grid member, performing task resource scheduling analysis on the first registration event and outputting the first task scheduling strategy; In the case where the first registration subject is a grid worker, after retrospective verification of the first registration event, the first task scheduling strategy is output through task resource scheduling analysis.

3. The grid governance event management method based on the event flow mechanism as claimed in claim 2 is characterized in that: Performing task resource scheduling analysis on the first registration event and outputting the first task scheduling strategy, the method includes: Interactively obtain multiple groups of sample resource allocation strategies for multiple sample governance events, where each sample resource allocation strategy includes a main resource scheduling requirement and F collaborative resource scheduling requirements; Based on the knowledge graph, the plurality of sample governance events and the plurality of groups of sample resource allocation strategies are stored in association, thereby completing the data filling of the event flow information library; The first edge grid node uses the first registration event to traverse the event flow information library to perform task resource scheduling and matching, and obtains a first set of sample resource allocation strategies; The first group of sample resource allocation strategies is packaged and output as the first task scheduling strategy.

4. The grid governance event management method based on the event flow mechanism as claimed in claim 3 is characterized in that: In the case where the first registration subject is a grid worker, after retrospective verification of the first registration event, the first task scheduling strategy is output through task resource scheduling analysis, and the method includes: In the case where the first registration subject is a grid worker, extracting a first event occurrence location and a first event timestamp from the first registration event; Performing a backtracking check on the first registration event based on the occurrence location of the first event and the timestamp of the first event as constraints, and outputting a first event verification result; If the first event verification result is passed, a task resource scheduling analysis is performed on the first registration event, and the first task scheduling strategy is output.

5. The grid governance event management method based on the event flow mechanism as claimed in claim 4 is characterized in that: Performing a backtracking check on the first registered event based on the occurrence location of the first event and the timestamp of the first event as constraints, and outputting a first event verification result, the method comprising: Positioning a first surveillance camera in the first grid area according to the location where the first event occurred; Perform video backtracking on the first surveillance camera according to the first event timestamp to obtain a first video frame sequence; Outputting a first recognized event by performing governance event recognition on the first video frame sequence; Perform consistency verification on the first registration event and the first identification event, and output the first event verification result.

6. The grid governance event management method based on the event flow mechanism as claimed in claim 5 is characterized in that: By performing governance event recognition on the first video frame sequence and outputting a first recognition event, the method includes: Interactively obtaining a plurality of sample time image sets of the plurality of sample governance events; After extracting identification features from the multiple sample time image sets, constructing multiple governance event identification channels based on the extraction results; By connecting the plurality of governance event identification channels in parallel, constructing a first edge event identification network in the first edge grid node; Inputting the first video frame sequence into the first edge event recognition network, performing governance event recognition via the multiple governance event recognition channels, and outputting multiple candidate recognition event sequences; The first recognition event is output by performing time aggregation on the multiple candidate recognition event sequences.

7. The grid governance event management method based on event flow mechanism as claimed in claim 3 is characterized in that: The grid governance cloud platform optimizes resource allocation for the first task scheduling strategy and M conflict scheduling strategies based on event processing priority, and outputs M+1 optimized scheduling strategies, wherein the method comprises: Interactively obtain an event governance priority sequence of the plurality of sample governance events; Parsing the M conflicting scheduling strategies to obtain M groups of conflicting resource allocation strategies; Taking the event governance priority sequence as a constraint, a policy selection conflict analysis is performed on the first group of sample resource allocation strategies and the M groups of conflicting resource allocation strategies, and the M+1 optimized scheduling strategies are output.

8. The grid governance event management method based on event flow mechanism as claimed in claim 3 is characterized in that: The grid governance cloud platform distributes the processing tasks of the M+1 optimization scheduling strategies across departments based on the event flow mechanism, and then the method further includes: Configuring M+1 governance time frames for the M+1 optimization scheduling strategies; Interactively obtain multiple sets of standard processing progress stage identifiers for the multiple sample governance events; In the grid governance cloud platform, with the M+1 governance time frames as constraints, the processing progress of the first grid governance event and the M grid governance events is visualized according to the multiple sets of standard processing progress stage identifiers.

9. The grid governance event management system based on the event flow mechanism is characterized by: The system is used to execute the grid governance event management method based on the event flow mechanism according to any one of claims 1 to 8, including: A grid node setting module, used to arrange multiple edge grid nodes for multiple grid areas in the target management area, wherein the multiple edge grid nodes are connected to the grid management cloud platform; A backtracking verification execution module, used for the first edge grid node to perform backtracking verification based on the first event occurrence location and the first event timestamp of the first registration event received as constraints when the first registration subject of the first registration event is a grid worker, and output a first task scheduling strategy according to the verification result; A scheduling conflict judgment module is used to perform resource scheduling conflict judgment of concurrent events by parsing the first task scheduling strategy after the grid governance cloud platform receives the first task scheduling strategy uploaded by the first edge grid node, and call out M conflict scheduling strategies of M grid governance events, wherein the M conflict scheduling strategies are uploaded by the M edge grid nodes, and the first task scheduling strategy has a first grid governance event identifier; A resource allocation optimization module, used to optimize resource allocation of the first task scheduling strategy and M conflict scheduling strategies based on event processing priority using the grid governance cloud platform, and output M+1 optimized scheduling strategies; The task distribution module is used to utilize the grid governance cloud platform to distribute the processing tasks of the M+1 optimization scheduling strategies across departments based on the event flow mechanism, so as to perform parallel event management of the first grid governance event and the M grid governance events.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, the steps of the grid governance event management method based on the event flow mechanism as described in any one of claims 1-8 are implemented.

Images