Smart city processing method and system based on multi-modal large model

Abstract

The invention discloses a smart city processing method and system based on a multi-modal large model, and relates to the technical field of smart cities. The method comprises the following steps: acquiring a target facility from a target area; determining a first prediction object according to the target facility; determining a plurality of collection devices according to the first prediction object, obtaining a plurality of monitoring data collected by the plurality of collection devices, and obtaining a first position corresponding to the target area; inputting the plurality of monitoring data and the first position into a multi-modal large model for processing to obtain an analysis result; determining a predicted safety accident according to the analysis result, and determining a first emergency processing scheme based on the predicted safety accident; and determining a safety level according to the processing duration and the predicted safety accident, and sending the first emergency processing scheme to the first department based on the safety level, so that personnel corresponding to the first department can check the first emergency processing scheme. By implementing the provided technical scheme, the problems that a traditional analysis framework depends on expert experience and is low in processing efficiency can be remarkably solved.

Description

Technical Field

[0001] This application relates to the technical field of smart cities, and specifically relates to a smart city processing method and system based on a multimodal large model. Background Art

[0002] As a new urban development model relying on advanced information and communication technologies (ICT), a smart city aims to improve urban operation efficiency, enhance the quality of residents' lives, and achieve sustainable development goals through cross-domain data integration, system integration, and technological innovation.

[0003] A smart city integrates the operation data of urban infrastructure such as transportation, healthcare, education, and energy through technologies such as the Internet of Things (IoT), cloud computing, and big data, and constructs a unified data governance framework. On this basis, artificial intelligence algorithms such as machine learning and deep learning are used to perform real-time analysis on massive heterogeneous data, mine potential patterns and risks, and provide intelligent decision-making support for urban governance. For example, an intelligent transportation system dynamically generates a traffic situation map by integrating road condition sensors, in-vehicle terminals, and mobile device data, and realizes congestion prediction and route optimization. Although data integration has brought convenience to urban governance, in the existing analysis process, data anomaly detection relies on the experience of domain experts. With the exponential growth of the data scale, the traditional analysis framework takes a long time, resulting in low analysis efficiency.

[0004] Therefore, there is an urgent need for a smart city processing method and system based on a multimodal large model that can solve the above technical problems. Summary of the Invention

[0005] This application provides a smart city processing method and system based on a multimodal large model. The method inputs the monitoring data automatically obtained by the acquisition device into the multimodal large model for correlation analysis, and generates analysis results in real time, which can significantly solve the problems of relying on expert experience and low processing efficiency in the traditional analysis framework.

[0006] In a first aspect, the present application provides a smart city processing method based on a multimodal large model. The method includes: obtaining target facilities from a target area, where the target facilities include any one or more of public service facilities, commercial facilities, residential facilities, productive building facilities, and industrial building facilities, and the target area is the area corresponding to any city; determining a first prediction object according to the target facilities, where the first prediction object includes traffic accidents, medical accidents, natural disaster accidents, construction accidents, public security accidents, and park security accidents; determining a plurality of collection devices according to the first prediction object, obtaining a plurality of monitoring data collected by the plurality of collection devices, and obtaining a first location corresponding to the target area; inputting the plurality of monitoring data and the first location into the multimodal large model for processing to obtain an analysis result; determining a predicted safety accident according to the analysis result, and determining a first emergency treatment plan based on the predicted safety accident; determining a safety level according to the processing duration and the predicted safety accident, and sending the first emergency treatment plan to a first department based on the safety level, so that the personnel corresponding to the first department can view the first emergency treatment plan, and the processing duration is the total duration for processing the first emergency treatment plan.

[0007] By adopting the above technical solution, the first prediction object is determined according to the type of target facilities in the target area, and a plurality of collection devices are determined according to the first prediction object to automatically obtain monitoring data, so as to cover all the information related to the first prediction object. Then, the obtained monitoring data and the first location are input into the multimodal large model for extraction and correlation analysis, without manual intervention, to obtain an analysis result. Then, a predicted safety accident is determined according to the analysis result, a corresponding emergency treatment plan is quickly generated according to the predicted safety accident, and the safety level is automatically divided according to the predicted safety accident and the processing duration. The emergency resources are dynamically allocated based on the safety level, and the corresponding department is notified in real time through message push. Using the multimodal large model for prediction can replace expert experience to solve the problem that the traditional method relies on domain experts to manually label abnormal data and needs to analyze multi-source data one by one, resulting in low efficiency.

[0008] Optionally, after determining the first prediction object according to the target facilities, the method further includes: obtaining a target factor currently corresponding to the target area, where the target factor includes a time factor and a weather factor; determining a second prediction object according to the target factor and the target facilities; determining whether there is a second prediction object in the first prediction object; when there is a second prediction object in the first prediction object, preferentially monitor the target area according to the second prediction object.

[0009] By adopting the above technical solution, the priority of the prediction object is dynamically adjusted according to the current time and weather. The traditional solution fixedly monitors the preset accident types and cannot cope with sudden environmental changes, resulting in waste of resources. By integrating multi-dimensional data of time, weather, and facility type, more accurate risk prediction is realized, high-risk objects are preferentially monitored, early warning is achieved, and the accuracy of accident prediction is significantly improved.

[0010] Optionally, before obtaining multiple monitoring data collected by multiple collection devices and obtaining the first position corresponding to the target area, the method further includes: determining a data type according to a first prediction object, where the data type includes an electricity consumption data type, a traffic data type, a personnel location data type, an environmental data type, and a construction data type; inputting the data type into a preset frequency database for matching to obtain a target collection frequency; determining a collection device according to the data type, and sending the target collection frequency to each collection device, so that the collection device performs data collection according to the target collection frequency.

[0011] By adopting the above technical solution, the collection frequency is dynamically adjusted according to the data sensitivity of different prediction objects. Since the traditional solution uses a fixed collection frequency, it results in insufficient collection of highly sensitive data or redundancy of low-sensitive data, increasing the storage and computing costs. The collection devices are automatically allocated according to the data type, and the collection resources are dynamically adjusted based on the scenario requirements, significantly improving the efficiency, accuracy, and resource utilization rate of data collection.

[0012] Optionally, determining a safety level according to the processing duration and predicting a safety accident specifically includes: Extracting core data from the first emergency treatment plan; determining a preset sensitive data table according to the predicted safety accident, and judging whether there is core data in the preset sensitive data table and whether the processing duration is less than or equal to a preset duration; when there is core data in the preset sensitive data table and the processing duration is less than or equal to the preset duration, determining that the first emergency treatment plan is at a high risk level, and outputting the high risk level as the safety level; when there is core data in the preset sensitive data table and the processing duration is greater than the preset duration, determining that the first emergency treatment plan corresponds to a medium risk level, and outputting the medium risk level as the safety level.

[0013] By adopting the above technical solution, the risk level is automatically determined according to the preset sensitive data table and the processing duration, avoiding the subjectivity and delay of manual evaluation. Through the preset sensitive data table, only key data is evaluated for risk, avoiding interference from irrelevant data. The allocation of emergency resources is dynamically adjusted according to the risk level. The traditional solution may cause resource waste or resource shortage due to misjudgment of the risk level, significantly improving the evaluation accuracy and response efficiency of the emergency treatment plan.

[0014] Optionally, the first emergency response plan is sent to the first department based on the security level, which specifically includes: when the first emergency response plan corresponds to a high-risk level, the first emergency response plan is first encrypted according to the high-risk level to obtain a second emergency response plan, and the first encryption includes quantum encryption; obtain the second department, where the second department is the department involved in the first emergency response plan; search for multiple preset departments identical to the second department in the target area, with one preset department corresponding to one location; obtain the target permission level corresponding to the third department, where the third department is any one of the multiple preset departments; determine whether the target permission level is consistent with the preset permission level, and the preset permission level is the permission level corresponding to the high-risk level; when the target permission level is consistent with the preset permission level, determine to allocate the second emergency response plan to the third department and output the third department as the first department.

[0015] By adopting the above technical solution, when the emergency response plan is determined to be at a high-risk level, quantum encryption is automatically triggered to generate a second emergency response plan. The characteristics of quantum encryption ensure the security of the transmission process and the storage process. Traditional encryption may lead to data leakage due to key leakage or algorithm cracking. Through the dynamic comparison of the target permission level and the preset permission level, it is ensured that only authorized departments can access high-risk plans. By searching for preset departments identical to the second department in the target area and combining permission level matching, the dynamic allocation of emergency resources is achieved. The entire process is automatically executed by the system, reducing human intervention. Through the dynamic matching of preset departments and permission levels, multi-department collaboration and cross-regional linkage are supported, significantly improving the security, execution efficiency, and collaboration ability of the emergency response plan.

[0016] Optionally, the first emergency response plan is sent to the first department based on the security level, which specifically includes: when the first emergency response plan corresponds to a medium-risk level, the first emergency response plan is second encrypted according to the medium-risk level to obtain a third emergency response plan, and the second encryption is to encrypt only the core words in the first emergency response plan. The processing plan consists of core words and non-core words, and the core words are the words that appear in the preset sensitive information table; obtain the second location corresponding to the predicted safety accident, where the second location is the specific location of the predicted safety accident in the target area; calculate the distances between multiple preset departments and the second location in sequence to obtain multiple distance values; sort the multiple distance values from smallest to largest to obtain the target sorting result; obtain the fourth department, where the fourth department is the department corresponding to the distance value ranked first in the target sorting result, and send the third emergency response plan to the fourth department.

[0017] By adopting the above technical solution, when the emergency response plan is at the medium risk level, only the core words that appear in the preset sensitive information table in the emergency response plan are encrypted, rather than encrypting the entire emergency response plan. Through targeted encryption, while ensuring the security of sensitive information, the encryption calculation amount is reduced, the response efficiency is improved. By calculating the distances from multiple preset departments to the accident location and sorting them from small to large, the third emergency response plan is preferentially sent to the department closest to the accident location. According to the distance sorting result, emergency resources are dynamically allocated.

[0018] Optionally, after determining the first prediction object according to the target facility, the method further includes: obtaining the target activities in the target area within a preset time, where the target activities include concert activities, sports meeting activities, and exhibition activities; combining the target activities with the target facility to obtain target combination information; performing simulation analysis on the target combination information to obtain a third prediction object, and preferentially monitoring the target area according to the third prediction object.

[0019] By adopting the above technical solution, the target activities (such as concerts, sports meetings, exhibitions) are combined with the target facility, and potential high-risk scenarios are identified through simulation analysis. The traditional solution may ignore the combined risks due to the lack of correlation analysis between activities and facilities, or respond passively after the risk occurs. According to the third prediction object, the target area is preferentially monitored. Based on the simulation analysis results, an emergency response plan for high-risk combinations is formulated in advance. By identifying high-risk combinations in advance and formulating response measures, the probability of accidents is reduced.

[0020] In the second aspect of the present application, a smart city processing system based on a multi-modal large model is provided. The system includes an acquisition unit, a processing unit, and a sending unit. The acquisition unit acquires a target facility from the target area, where the target facility includes any one or more of public service facilities, commercial facilities, residential facilities, productive building facilities, and industrial building facilities, and the target area is the area corresponding to any city; the processing unit determines a first prediction object according to the target facility, where the first prediction object includes traffic accidents, medical accidents, natural disaster accidents, construction accidents, public security accidents, and park security accidents; determines a plurality of acquisition devices according to the first prediction object, acquires a plurality of monitoring data collected by the plurality of acquisition devices, and acquires the first position corresponding to the target area; inputs the plurality of monitoring data and the first position into the multi-modal large model for processing to obtain an analysis result; determines a predicted safety accident according to the analysis result, and determines a first emergency response plan based on the predicted safety accident; the sending unit determines a safety level according to the processing duration and the predicted safety accident, and sends the first emergency response plan to the first department based on the safety level, so that the personnel corresponding to the first department can view the first emergency response plan, and the processing duration is the total duration corresponding to processing the first emergency response plan.

[0021] Optionally, the obtaining unit is configured to obtain a target factor currently corresponding to the target area, where the target factor includes a time factor and a weather factor; the processing unit is configured to determine a second prediction object according to the target factor and the target facility; determine whether there is a second prediction object in the first prediction object; when there is a second prediction object in the first prediction object, preferentially monitor the target area according to the second prediction object.

[0022] Optionally, the processing unit is configured to determine a data type according to the first prediction object, where the data type includes an electricity consumption data type, a traffic data type, a personnel location data type, an environmental data type, and a construction data type; input the data type into a preset frequency database for matching to obtain a target collection frequency; determine a collection device according to the data type, and send the target collection frequency to each collection device so that the collection device performs data collection according to the target collection frequency.

[0023] Optionally, the obtaining unit is configured to extract core data from the first emergency treatment plan; the processing unit is configured to determine a preset sensitive data table according to the predicted safety accident, and determine whether there is core data in the preset sensitive data table and whether the processing duration is less than or equal to a preset duration; when there is core data in the preset sensitive data table and the processing duration is less than or equal to the preset duration, determine that the first emergency treatment plan is at a high risk level, and output the high risk level as the safety level; when there is core data in the preset sensitive data table and the processing duration is greater than the preset duration, determine that the first emergency treatment plan corresponds to a medium risk level, and output the medium risk level as the safety level.

[0024] Optionally, when the first emergency treatment plan corresponds to a high risk level, the processing unit is configured to perform first encryption on the first emergency treatment plan according to the high risk level to obtain a second emergency treatment plan, where the first encryption includes quantum encryption; the obtaining unit is configured to obtain a second department, where the second department is a department involved in the first emergency treatment plan; the processing unit is configured to search for a plurality of preset departments in the target area that are the same as the second department, and one preset department corresponds to one location; the obtaining unit is configured to obtain a target permission level corresponding to a third department, where the third department is any one of the plurality of preset departments; the processing unit is configured to determine whether the target permission level is consistent with a preset permission level, where the preset permission level is the permission level corresponding to the high risk level; when the target permission level is consistent with the preset permission level, the sending unit is configured to determine to allocate the second emergency treatment plan to the third department and output the third department as the first department.

[0025] Optionally, when the first emergency response plan corresponds to a medium risk level, the processing unit is configured to perform second encryption on the first emergency response plan according to the medium risk level to obtain a third emergency response plan. The second encryption is to encrypt only the core words in the first emergency response plan. The emergency response plan consists of core words and non-core words, and the core words are the words that appear in the preset sensitive information table. The obtaining unit is configured to obtain a second location corresponding to the predicted safety accident, where the second location is the specific location of the predicted safety accident in the target area. Calculate the distances between multiple preset departments and the second location in sequence to obtain multiple distance values. Sort the multiple distance values from smallest to largest to obtain a target sorting result. The sending unit is configured to obtain a fourth department, where the fourth department is the department corresponding to the distance value ranked first in the target sorting result, and send the third emergency response plan to the fourth department.

[0026] Optionally, the obtaining unit is configured to obtain target activities in the target area within a preset time. The target activities include concert activities, sports meeting activities, and exhibition activities. The processing unit is configured to combine the target activities with the target facilities to obtain target combination information. Perform simulation analysis on the target combination information to obtain a third prediction object, and preferentially monitor the target area according to the third prediction object.

[0027] In the third aspect of the present application, an electronic device is provided. The electronic device includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions. The user interface and the network interface are used to communicate with other devices. The processor is configured to execute the instructions stored in the memory, so that an electronic device executes the method according to any one of the above in the present application.

[0028] In the fourth aspect of the present application, a computer-readable storage medium is provided. The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of the above in the present application is executed.

[0029] In summary, one or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Determine the first prediction object according to the type of target facilities in the target area. Determine to use multiple acquisition devices to automatically obtain monitoring data according to the first prediction object to cover all information related to the first prediction object. Then input the obtained monitoring data and the first location into the multi-modal large model for extraction and correlation analysis. Without manual intervention, obtain the analysis result. Then determine the predicted safety accident according to the analysis result, quickly generate a corresponding emergency response plan according to the predicted safety accident, automatically divide the safety level according to the predicted safety accident and the processing duration, dynamically allocate emergency resources based on the safety level, and notify the corresponding department in real time through message push. Using the multi-modal large model for prediction can replace expert experience to solve the problem that traditional frameworks rely on domain experts to manually label abnormal data and need to analyze multi-source data one by one, resulting in low efficiency.

[0030] 2. Dynamically adjust the priority of prediction objects according to the current time and weather. The traditional solution fixedly monitors preset accident types and cannot cope with sudden environmental changes, resulting in waste of resources. By integrating multi-dimensional data of time, weather, and facility types, more accurate risk prediction is achieved, high-risk objects are preferentially monitored, early warnings are realized, and the accuracy of accident prediction is significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 is a schematic flowchart of a smart city processing method based on a multi-modal large model provided by an embodiment of the present application; Figure 2 is a schematic structural diagram of a smart city processing system based on a multi-modal large model provided by an embodiment of the present application; Figure 3 is a schematic structural diagram of an electronic device disclosed by an embodiment of the present application.

[0032] Description of reference numerals: 201, acquisition unit; 202, processing unit; 203, sending unit; 300, electronic device; 301, processor; 302, memory; 303, user interface; 304, network interface; 305, communication bus. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments.

[0034] In the description of the embodiments of the present application, words such as "for example" or "for illustration" are used to indicate examples, illustrations, or explanations. Any embodiment or design solution described as "for example" or "for illustration" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Exactly speaking, the use of words such as "for example" or "for illustration" is intended to present related concepts in a specific manner.

[0035] In the description of the embodiments of the present application, the term "plural" means two or more. For example, plural systems refer to two or more systems, and plural screen terminals refer to two or more screen terminals. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance or implicitly specifying the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. The terms "include", "comprise", "have" and their variants all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0036] As a new urban development model relying on advanced information and communication technology (ICT), Smart City aims to improve urban operation efficiency, enhance the quality of residents' lives and achieve sustainable development goals through cross-domain data fusion, system integration and technological innovation.

[0037] Smart City integrates the operation data of urban infrastructure such as transportation, medical care, education, and energy through technologies such as the Internet of Things (IoT), cloud computing, and big data, and constructs a unified data governance framework. On this basis, using artificial intelligence algorithms such as machine learning and deep learning, it conducts real-time analysis on massive heterogeneous data, mines potential rules and risks, and provides intelligent decision-making support for urban governance. For example, the intelligent transportation system dynamically generates a traffic situation map by integrating road condition sensors, in-vehicle terminals and mobile device data, and realizes congestion prediction and route optimization. Relying on a unified government service platform, Smart City promotes the full process digitization of government service matters and realizes "one network for all services". Citizens can complete operations such as identity authentication, material submission, and progress query online through mobile terminals or self-service terminals, breaking through time and space limitations. For example, the electronic certificate system based on blockchain technology can achieve cross-departmental data sharing and business collaboration, significantly shorten the approval cycle, and improve the accessibility of public services. Although data fusion brings convenience to urban governance, in the existing analysis process, data anomaly detection relies on the experience of domain experts. With the exponential growth of data volume, the traditional analysis framework takes a long time, resulting in low analysis efficiency.

[0038] Therefore, how to solve the problems of relying on expert experience and low processing efficiency in the traditional analysis framework. A Smart City processing method based on a multimodal large model provided by the embodiments of the present application is applied to a server. The server of the present application is a platform that provides Smart City services. Figure 1 It is a schematic flowchart of a Smart City processing method based on a multimodal large model provided by the embodiments of the present application. Refer to Figure 1 , and this method includes the following steps S101 - step S106.

[0039] S101: Obtain target facilities from the target area.

[0040] In the above S101, when monitoring the real-time situation of each city, due to the different building facilities and geographical locations within different cities, when predicting various situations of each city, it is necessary to combine the building facilities and geographical location of the city to determine the possible safety accidents, and then make predictions in a timely manner. This application takes the monitoring of various data of a certain city as an example to illustrate. First, determine the monitoring area, that is, the target area. The target area refers to any city or a certain area in the city. Then, through the urban geographic information system (GIS), building drawings, satellite images, Internet of Things sensors, etc., obtain the information of the target facilities in the target area. The target facilities include any one or more of public service facilities, commercial facilities, residential facilities, productive building facilities, and industrial building facilities. According to the facility functions, various buildings in real life are classified into five categories. Public service facilities include schools, hospitals, libraries, fire stations, etc.; commercial facilities include shopping malls, office buildings, hotels, restaurants, etc.; residential facilities include residential communities, apartments, villas, etc.; productive building facilities include factories, warehouses, logistics centers, etc.; industrial building facilities include chemical plants, power plants, refineries, etc. The GIS platform (such as ArcGIS, QGIS) can be used for spatial analysis, and the facility attributes can be extracted in combination with the building information model (BIM).

[0041] For example, the target area refers to Area A of a certain city. Then obtain what target facilities exist in Area A of the city. If Area A of the city includes the People's Hospital, primary school, Garden Community, and chemical plant, etc., then classify them into the corresponding five major facility categories according to the facility functions at this time.

[0042] S102: Determine the first prediction object according to the target facilities. The first prediction object includes traffic accidents, medical accidents, natural disaster accidents, construction accidents, public security accidents, and park safety accidents.

[0043] In the above S102, after obtaining the target facilities from the target area, according to the facility type, analyze the possible types of safety accidents it may cause. At this time, the type of safety accident refers to the first prediction object. For example, public service facilities may cause medical accidents and natural disaster accidents, while commercial facilities may cause fire accidents and public security accidents. Residential facilities may cause gas leakage accidents and high-altitude falling object accidents, etc. Productive building facilities may cause construction accidents and equipment failure accidents, etc. Industrial building facilities may cause chemical leakage accidents and explosion accidents, etc. For example, if the target facility is a chemical plant, the first prediction object is a chemical leakage accident or an explosion accident.

[0044] In addition to directly analyzing potential safety accidents based on the facility type, safety accidents can also be dynamically adjusted according to the time and weather in the target area. Priority is given to monitoring relatively urgent safety accidents, detecting them in a timely manner and issuing early warnings. Specifically, it includes: obtaining the target factors corresponding to the current target area, where the target factors include time factors and weather factors; determining the second prediction object based on the target factors and the target facilities; judging whether there is a second prediction object among the first prediction objects; when there is a second prediction object among the first prediction objects, priority is given to monitoring the target area according to the second prediction object. Specifically, first obtain the current time in the target area. The current time is used to determine whether it is during the morning and evening rush hours, holidays, and special periods. Then obtain the real-time weather data corresponding to the target area. The real-time weather data includes heavy rain, high temperature, strong wind, haze, and sunny days, etc. The current time can be obtained through the system clock, or the special periods (such as the Spring Festival travel rush, peak tourist season) can be judged by combining historical data. The peak tourist season can be analyzed in combination with the tourist attractions in the target area. The real-time weather data can be obtained through the meteorological bureau API. For example, the target area is Area A of a certain city. The current time is 18:00 on October 1, 202x (National Day holiday), and the current weather is heavy rain (precipitation 50mm / hour). Then, in combination with the target factors (time and weather) and the target facilities, analyze the possible types of safety accidents. For example, if the current time is during a holiday and the shopping mall is crowded with people, a stampede accident may occur. If the current time is during the evening rush hour after work, a traffic accident may occur. When the weather in the target area is heavy rain, waterlogging may occur in low-lying areas, and chemical leakage may occur in chemical plants. When the weather is high temperature, heatstroke accidents may occur in open-air construction areas. According to the correlation analysis results, determine the possible types of safety accidents. The target facilities are the square and the chemical plant. Among the target factors, the time is a holiday and the weather is heavy rain. At this time, the second prediction object corresponding to the square is a stampede accident, and the second prediction object of the chemical plant is a chemical leakage accident. Since the first prediction objects (such as traffic accidents, medical accidents, etc.) have been determined in advance according to the target facilities. Then compare the first prediction object with the second prediction object to judge whether there is an overlap between the two. When there is an overlap between the two, take the second prediction object as the priority monitoring object. When there is no overlap, continue to monitor according to the first prediction object. For example, when the target facility is a square, the corresponding first prediction object is a public safety accident. When the target is a chemical plant, the corresponding first prediction object is a chemical leakage accident. Introduce the time and weather factors, and then conduct a correlation analysis between the time and weather factors and the target facilities. When the target facility is a square, the second prediction object is a stampede accident. When the target facility is a chemical plant, the second prediction object is a chemical leakage accident at this time. Then compare the first prediction object with the second prediction object. When the target facility is a square, since the first prediction object (public safety accident) includes the second prediction object (stampede accident), it is necessary to give priority to monitoring the second prediction object.When the target facility is a chemical plant, the first prediction object is the same as the second prediction object, and continuous monitoring is carried out. When the second prediction object exists, adjust the priority of the monitoring equipment, and focus on monitoring the relevant areas affected by time and weather factors, so as to avoid the situation where areas with safety accidents cannot be detected in time without setting priorities, ensure that high-risk prediction objects are monitored first, and improve the emergency response efficiency. For example, in the heavy rain on National Day, chemical leakage accidents in chemical plants can be detected and warned in time by increasing the monitoring frequency of gas sensors.

[0045] Furthermore, spatial association can be performed with the target facilities according to the activities carried out in the target area to avoid missing high-risk areas during prediction, specifically including: obtaining the target activities in the target area within a preset time, where the target activities include concert activities, sports meeting activities, and exhibition activities; combining the target activities with the target facilities to obtain target combination information; performing simulation analysis on the target combination information to obtain a third prediction object, and preferentially monitoring the target area according to the third prediction object. Specifically, first determine the current first position corresponding to the target area, and then use web crawler technology to capture the activity information carried out within the first position within the preset time from relevant platforms. The activity information can be obtained from the official activity approval records of the government or the venue, or the activity information can be viewed through the ticketing platform, or the activity information can be obtained through the activity promotion information on platforms such as social media. At this time, the preset time can be the next 1 week, 1 month, or a specific time period (such as holidays). After obtaining the activity data, remove duplicate and invalid activity information. The activities include concert activities, sports meeting activities, and exhibition activities. Then obtain the target facilities in the target area. The commercial facilities include venues (such as stadiums), and the public service facilities also include roads (roads around the stadium), parking lots, and public transportation stations, etc. Associate each activity with its corresponding facility to form an activity-facility combination. At this time, the corresponding facility refers to the location where the activity is carried out. The activity location and the facility location can be matched through a Geographic Information System (GIS). For example, the activity name is a science and technology exhibition, and the location where the science and technology exhibition is carried out is the science and technology museum, that is, the target facility is the science and technology museum. At this time, combine the activity information with the target facility to obtain science and technology exhibition - science and technology museum. Then predict the possible risks or impacts of each activity-facility combination to generate a third prediction object (such as the number of people, traffic congestion, safety hazards), and the data of past similar activities (such as the number of people, accident records) can be referred to. Traffic simulation software (such as VISSIM) can also be used to simulate the flow of people and vehicles during the activity. Since the activity carried out at this time is a science and technology exhibition, the equipment of the science and technology exhibition and the science and technology museum is associated and analyzed to analyze the possible safety accidents. Therefore, the third prediction object is to identify risks such as facility aging and fire passage occupancy. After obtaining the third prediction object, the target area can be preferentially monitored according to the third prediction object, and real-time inspections can be carried out on the fire passage and power facilities. For example, during a star concert, it is determined that the current third prediction object is a traffic accident. Therefore, the traffic and the number of people on the surrounding roads where the concert is held are preferentially predicted, and then the security measures are automatically adjusted according to the predicted peak number of people to ensure the safety of the activity.

[0046] S103: Determine multiple collection devices according to the first prediction object, obtain multiple monitoring data collected by the multiple collection devices, and obtain the first position corresponding to the target area.

[0047] In the above S103, after determining the first prediction object based on the actual situation of the target area, a suitable monitoring device is selected according to the first prediction object. For example, when the first prediction object is a chemical leakage accident, the collection devices for the chemical leakage accident are gas sensors, liquid level sensors, etc. Then, through the Internet of Things platform or directly connecting to the device, the monitoring data collected by the collection device is obtained in real time. When collecting the monitoring data, the position corresponding to the first prediction object in the target area needs to be obtained, and then a connection is established with the collection device installed at this position.

[0048] In addition, before using the acquisition device to obtain the monitoring data, since the acquisition frequency will affect the subsequent data analysis results, after determining the acquisition device for data acquisition, it is also necessary to adjust the acquisition frequencies of each acquisition device to ensure that the collected monitoring data can be used for subsequent analysis and avoid the situation where there is insufficient data for analysis. Specifically, it includes: determining the data type according to the first prediction object, where the data type includes electricity consumption data type, traffic data type, personnel location data type, environmental data type, and construction data type; inputting the data type into the preset frequency database for matching to obtain the target acquisition frequency; determining the acquisition device according to the data type and sending the target acquisition frequency to each acquisition device so that the acquisition device can acquire data according to the target acquisition frequency. Specifically, based on the target facilities in the target area, predict the possible types of safety accidents, that is, the first prediction object. Associate the first prediction object with the data types that need to be monitored. For traffic accidents, the required traffic data types (traffic flow, vehicle speed, road congestion index), for fires, the required environmental data types (temperature, smoke concentration, oxygen content), for chemical leaks, the required environmental data types (gas concentration, liquid level, pressure), and for risks in crowded areas, the required personnel location data types (crowd density, residence time). For example, if the first prediction object is a chemical leakage accident in an industrial park, the data types resulting in the chemical leakage accident include environmental data type and construction data type. At this time, the environmental data includes gas concentration, liquid level, pressure, and the construction data type includes pipeline pressure and valve status. Build a preset frequency database in advance, which contains the corresponding relationship between each data type and the acquisition frequency. For example, for the environmental data type, the gas concentration can be set to once per minute or once per hour, and the temperature can be set to once per 5 minutes; for the traffic data type, the traffic flow can be set to once per second or once per 10 seconds. Select different acquisition frequencies based on different data types. Select the corresponding acquisition frequency from the database according to the first prediction object to ensure that the monitoring data collected at the acquisition frequency can be used for subsequent data analysis. The first prediction object can also be classified according to the risk level. For example, a chemical leakage accident can be classified as a high-risk level, and at this time, the acquisition frequency can be set to once per minute according to the risk level. Then select the corresponding hardware device according to the data type. The acquisition devices for the environmental data type are gas sensors and temperature sensors. The acquisition devices for the traffic data type are cameras and geomagnetic sensors. The acquisition devices for the personnel location data type are Wi-Fi probes and Bluetooth beacons. Send the target acquisition frequency to the acquisition device through the Internet of Things protocol (such as MQTT, CoAP) or API interface. The acquisition device adjusts the sampling period according to the received frequency. For example, the gas sensor is adjusted from collecting once per hour to collecting once per minute.

[0049] After determining the acquisition frequency, the acquisition device acquires relevant monitoring data according to the acquisition frequency and sends the acquired multiple monitoring devices to the server. It is also necessary to obtain the first location corresponding to the target area, where the first location refers to the location of the facilities corresponding to the first prediction object in the target area. For example, when the target facility in the target area is a chemical plant, the first prediction object for the chemical plant is determined to be a chemical plant leakage accident. According to the chemical plant leakage accident, the acquisition device is determined. The acquisition device includes a gas sensor (monitoring chemical leakage) and a pressure sensor (monitoring explosion risk). Then, the gas sensor is used to obtain the ammonia concentration of 15 ppm in a certain area, and the pressure sensor is used to obtain the pressure of 2.5 Mpa in a certain storage tank. The acquired multiple monitoring data are transmitted. At this time, it is also necessary to obtain the specific location of the chemical plant in the target area, that is, the first location.

[0050] S104: Input the multiple monitoring data and the first location into a multi-modal large model for processing to obtain an analysis result.

[0051] In the above S104, after receiving the multiple monitoring data sent by the acquisition device and before inputting the multiple monitoring data and the first location into the multi-modal large model for processing, it is necessary to construct a multi-modal large model. The multi-modal large model uses large models such as BERT and combines multi-modal inputs such as text (monitoring data), images (facility images), and geographical locations (GIS data) for comprehensive analysis. In the early stage, various historical monitoring data, historical accidents, and geographical locations need to be collected to obtain a data sample set. Then, the initial model is used to train the data sample set. When the loss value between the training result output by the model training and the actual result meets the convergence condition, the training is ended, and the initial model at the end of the training is output as the multi-modal large model. Then, the multiple monitoring data and the first location information are converted into a format that can be understood by the multi-modal large model. Then, the key features in the data (such as ammonia concentration, pressure value, etc.) are extracted. According to the features, the possible types and probabilities of safety accidents are predicted. At this time, the occurrence probability refers to the type and occurrence probability of the safety accident. For example, when the ammonia concentration in the input data is 15 ppm and the pressure of the storage tank is abnormal, and then the layout plan and location of the chemical plant are also input into the multi-modal large model for processing to obtain an analysis result. The analysis result includes that there is a chemical leakage risk in the chemical plant, with a probability of 70%.

[0052] S105: Determine the predicted safety accident according to the analysis result and determine the first emergency treatment plan based on the predicted safety accident.

[0053] In the above S105, obtain the accident type and the occurrence probability from the analysis result, and then compare the accident type and the occurrence probability with the preset occurrence probability. When the occurrence probability is less than the preset occurrence probability, it is determined that the accident occurrence probability is low, and continue to monitor this area, predicting that the safety accident is none. When the occurrence probability is equal to or greater than the preset occurrence probability, determine that the accident type in this analysis result is the predicted safety accident. At this time, the predicted safety accident refers to a safety accident that is likely to occur with a high probability. In the above example, there is a risk of chemical leakage in the chemical plant, with a probability of 70%. At this time, the predicted safety accident is a chemical leakage accident. According to the predicted safety accident type, formulate a corresponding emergency treatment plan. When the predicted safety accident is a chemical leakage accident, the first emergency treatment plan includes evacuating the surrounding residents; activating the emergency plan and notifying departments such as fire and environmental protection; closing the leakage source and carrying out emergency repairs. When the predicted safety accident is an explosion accident, the first emergency treatment plan includes activating the emergency evacuation procedure; notifying departments such as medical and fire to standby; cutting off the power supply to prevent secondary accidents.

[0054] S106: Determine the safety level according to the processing duration and the predicted safety accident, and send the first emergency treatment plan to the first department based on the safety level, so that the corresponding personnel in the first department can view the first emergency treatment plan.

[0055] In the above S106, the processing duration is the total duration corresponding to the first emergency treatment plan. After determining the first emergency treatment plan according to the predicted safety accident, it is necessary to obtain the response time of the first emergency treatment plan, that is, how long each department needs to complete this treatment plan. The safety level can be divided according to the severity and impact degree of the predicted safety accident, and the safety level can also be determined according to the processing duration and the predicted safety accident. Specifically, it includes: extracting the core data from the first emergency treatment plan; determining the preset sensitive data table according to the predicted safety accident, and judging whether the core data exists in the preset sensitive data table, and judging whether the processing duration is less than or equal to the preset duration; when the core data exists in the preset sensitive data table and the processing duration is less than or equal to the preset duration, it is determined that the first emergency treatment plan is at a high risk level, and the high risk level is output as the safety level; when the core data exists in the preset sensitive data table and the processing duration is greater than the preset duration, it is determined that the first emergency treatment plan corresponds to a medium risk level, and the medium risk level is output as the safety level.

[0056] Specifically, first obtain the first emergency response plan, which includes the issuing department, monitoring time, collection frequency, accident occurrence probability, predicted accident occurrence time, etc. Then, through natural language processing (NLP) technology, extract structured data from the emergency response plan text, match key fields, identify high-risk keywords, and summarize the key information in the first emergency response plan to obtain core data. At this time, the key information includes the processing duration, processing measures, affected areas, and resource requirements. The processing duration refers to the time required to complete the emergency response plan, such as 30 minutes. The processing measures refer to specific emergency measures, such as evacuating personnel. The affected area refers to the scope of the accident impact, such as a certain floor. The resource requirements refer to the required manpower and materials, such as 3 firefighters and 2 fire extinguishers. After obtaining the core data, establish a sensitive field table based on the predicted safety accidents, that is, a database associating each historical accident with a risk level, that is, which fields correspond to high risks and which fields do not correspond to high risks. Then, match the core data with the preset sensitive data table to check whether the core data contains the fields in the preset sensitive data table. For example, the core data includes a processing duration of 45 minutes, closing the valve, evacuating personnel, and a certain workshop, while the preset sensitive database includes gas concentration, personnel density, and affected dangerous areas, etc. At this time, the preset sensitive data table contains the core data, and then compare the extracted processing duration with the preset duration. The preset duration is the allowable processing duration for normal chemical leakage accidents. When the processing duration is 45 minutes and the preset duration is 60 minutes, at this time the processing duration is less than or equal to the preset duration, and it is determined that the first emergency response plan corresponds to a high risk. When the processing duration is greater than the preset duration, it is determined that the first emergency response plan corresponds to a medium risk. When the core data does not exist in the preset sensitive data table and the processing duration is greater than the preset duration, it is determined that the first emergency response plan corresponds to a low risk. When the core data does not exist in the preset sensitive data table and the processing duration is less than or equal to the preset duration, it is determined that the first emergency response plan is at a medium risk level.

[0057] Further, after determining the security level, select an appropriate transmission method according to the security level to send the first emergency response plan to the relevant departments so that the personnel of the relevant departments can promptly adopt corresponding operations, specifically including: sending the first emergency response plan to the first department based on the security level, specifically including: when the first emergency response plan corresponds to a high-risk level, perform first encryption on the first emergency response plan according to the high-risk level to obtain a second emergency response plan, and the first encryption includes quantum encryption; obtain the second department, where the second department is the department involved in the first emergency response plan; screen multiple preset departments identical to the second department from the target area, with one preset department corresponding to one location; obtain the target permission level corresponding to the third department, where the third department is any one of the multiple preset departments; determine whether the target permission level is consistent with the preset permission level, and the preset permission level is the permission level corresponding to the high-risk level; when the target permission level is consistent with the preset permission level, determine to allocate the second emergency response plan to the third department and output the third department as the first department. Specifically, by evaluating the first emergency response plan, determine the security level corresponding to the first emergency response plan. When the first emergency response plan corresponds to a high-risk level, start the encryption process. For example, when a chemical plant leaks, if it is evaluated that the leakage volume exceeds 5 tons and there is a residential area within 5 kilometers around, it is determined to be a high-risk level. The first encryption can be performed on the first emergency response plan. At this time, the first encryption refers to using a quantum key distribution system to generate a random key and using the characteristics of quantum states (such as photon polarization) to ensure that the key cannot be eavesdropped. Convert the first emergency response plan into a digital format and use a symmetric encryption algorithm (such as AES) combined with the quantum key to encrypt the data. Generate the ciphertext and the corresponding quantum key identifier (used for matching during decryption). At this time, the generated ciphertext can be used as the second emergency response plan, and the key is transmitted securely to the corresponding department through a quantum channel in advance. The corresponding department here refers to the department mentioned in the emergency response plan. By parsing the list of responsible departments in the first emergency response plan (such as the fire department, environmental protection department, and medical team). Store the list as a "second department" list for subsequent permission verification. For example, in the first emergency plan, it is clearly stated that the fire department is responsible for extinguishing fires, the environmental protection department is responsible for monitoring pollution, and the medical team is responsible for treating the wounded. Divide the geographical range according to the accident location (such as all departments within a radius of 10 kilometers). Traverse all department databases in the target area and screen out the departments that match the name or function in the second confidentiality list. Then, assign permission levels to each department in advance (such as the fire department - high, environmental protection department - medium). Then, determine the permission level according to the high risk corresponding to the preset safety accident. At this time, the permission level refers to the qualifications or scale that the department handling the accident needs to have. Then, select the third department one by one from the preset department list. Query the permission level of this department and compare it with the preset permission level (high risk corresponds to high permission).When comparing the permission level corresponding to the third department with the preset permission level, if the permission level of the third department is consistent with the preset permission level, it is determined that the third department can handle the accident, and then the second emergency response plan is allowed to be allocated to the third department. Transmitting the encrypted second emergency response plan to the eligible departments realizes the security encryption, precise allocation, and permission control of the high-risk emergency plan, ensuring the efficiency and security of the emergency response. When the permission level corresponding to the third department does not match the preset permission level, other departments need to be retrieved from the preset departments, and the permission levels of other departments are compared with the preset permission level to find the eligible department to handle the accident. After receiving the second emergency response plan, the third department decrypts the second emergency response plan with the key to obtain the decrypted first emergency response plan, and then executes the corresponding treatment measures according to the decrypted first emergency response plan. Record the entire allocation process in real time for easy auditing and tracing.

[0058] Furthermore, when the first emergency response plan corresponds to the medium risk level, the first emergency response plan is secondarily encrypted according to the medium risk level to obtain a third emergency response plan. The secondary encryption is to encrypt only the core words in the first emergency response plan. The response plan consists of core words and non-core words, and the core words are the words that appear in the preset sensitive information table; obtain the second location corresponding to the predicted safety accident, and the second location is the specific location of the predicted safety accident in the target area; calculate the distances between multiple preset departments and the second location in sequence to obtain multiple distance values; sort the multiple distance values in ascending order to obtain the target sorting result; obtain the fourth department, where the fourth department is the department corresponding to the distance value ranked first in the target sorting result, and send the third emergency response plan to the fourth department. Specifically, if the risk level of the first emergency response plan is determined to be medium risk, the second encryption process is triggered. First, extract the words that appear in the preset sensitive information table from the first emergency response plan. For example, the preset sensitive information table includes leakage, fire, and explosion. The first emergency response plan includes that a gas leakage occurs, and personnel should be evacuated immediately and the valve should be closed. At this time, the core word is "leakage". The second encryption process is to encrypt only the core words, and the non-core words remain unchanged. The keyword matching algorithm can be used to identify the core words, and then the core words are replaced and encrypted, such as replaced with a hash value, to obtain the third emergency response plan. Then obtain the second location corresponding to the predicted safety accident. At this time, if the first predicted object is a chemical leakage accident, the chemical plant in the target area needs to be monitored. At this time, the specific location of the chemical plant in the target area can be obtained, that is, the second location. The accident location can be obtained in real time through Internet of Things sensors. Then determine multiple preset departments corresponding to the response from the first emergency response plan. The preset departments refer to the relevant departments that need to participate in the emergency response plan this time. The relevant departments include the fire department, hospital, environmental protection bureau, work safety supervision bureau, and traffic management department. For example, if the department mentioned in the first emergency response plan is the fire department, all fire departments in the target area are obtained at this time, and then multiple preset departments are obtained. Then obtain the locations corresponding to each of the multiple preset departments in sequence, and then calculate the distances between the multiple preset departments and the second location in sequence. The straight-line distance can be calculated using longitude and latitude (such as the Haversine formula), or the actual road distance can be obtained by calling a map API (such as x-de Map, x-du Map). For each preset department, calculate its distance to the second location to obtain multiple distance values. Sort the multiple distance values in ascending order to obtain the target sorting result. Use a sorting algorithm (such as quicksort, mergesort) to sort the distance values. Obtain the department ranked first in the target sorting result (i.e., the department closest in distance) as the fourth department, and automatically push the second emergency response plan through the emergency command system. Send the third emergency response plan to the person in charge of the fourth department.When multiple processing departments are mentioned in the emergency response plan, the relevant departments closest to the preset safety accident can be selected in sequence according to the above content, and then the encrypted third emergency response plan can be sent to these departments. After receiving the third emergency response plan, the fourth department decrypts the encrypted data involved in the third emergency response plan to obtain the decrypted emergency response plan, and then executes the corresponding processing measures according to the decrypted emergency response plan.

[0059] When the first emergency response plan corresponds to a low-risk level, for example, the first emergency response plan is weather anomaly information, and it is necessary to remind people in a certain area to pay attention to safety when traveling. At this time, the first emergency response plan does not involve sensitive information, and the first emergency response plan can be directly sent to the meteorological department and the traffic management department, so that the personnel of the meteorological department and the traffic management department can perform subsequent operations according to the first emergency response plan. In addition to the above-mentioned safety accidents, the real-time situation of traffic congestion sections can also be provided for users, so that users can adjust their travel time or routes in a timely manner. The historical scenic spots or buildings in the target area can also be combined to provide travel arrangements for users or other tourists. The safety accidents in each factory or park can also be monitored daily, early warnings can be issued in advance, the probability of equipment failures can be reduced, and the lives of workers can be ensured.

[0060] By adopting the above method, the first prediction object is determined according to the target facility type in the target area, and multiple acquisition devices are used to automatically obtain monitoring data according to the first prediction object to cover all the information related to the first prediction object. Then, the obtained monitoring data and the first location are input into the multi-modal large model for extraction and correlation analysis. Without manual intervention, the analysis result is obtained. Then, the predicted safety accident is determined according to the analysis result, and the corresponding emergency response plan is quickly generated according to the predicted safety accident. Then, the predicted safety accident and the processing duration are used to automatically divide the safety level, and the emergency resources are dynamically allocated based on the safety level, and the corresponding department is notified in real time through message push. Using the multi-modal large model for prediction can replace expert experience to solve the problem that traditional frameworks rely on domain experts to manually label abnormal data, and multiple source data need to be analyzed one by one, resulting in low efficiency.

[0061] The embodiment of the present application also provides a smart city processing system based on a multi-modal large model. Figure 2 It is a schematic structural diagram of a smart city processing system based on a multi-modal large model provided by the embodiment of the present application. Refer to Figure 2 The system includes an acquisition unit 201, a processing unit 202, and a sending unit 203.

[0062] The acquisition unit 201 acquires target facilities from the target area, and the target facilities include any one or more of public service facilities, commercial facilities, residential facilities, productive building facilities, and industrial building facilities. The target area is an area corresponding to any city.

[0063] The processing unit 202 determines a first prediction object according to the target facility. The first prediction object includes traffic accidents, medical accidents, natural disaster accidents, construction accidents, public security accidents, and park security accidents. Multiple acquisition devices are determined according to the first prediction object, and multiple monitoring data collected by the multiple acquisition devices are obtained, and a first position corresponding to the target area is obtained. The multiple monitoring data and the first position are input into a multimodal large model for processing to obtain an analysis result. A predicted safety accident is determined according to the analysis result, and a first emergency treatment plan is determined based on the predicted safety accident.

[0064] The sending unit 203 determines a safety level according to the processing duration and the predicted safety accident, and sends the first emergency treatment plan to the first department based on the safety level, so that the personnel corresponding to the first department can view the first emergency treatment plan. The processing duration is the total duration corresponding to the processing of the first emergency treatment plan.

[0065] In a possible implementation manner, the obtaining unit 201 is used to obtain the current target factors corresponding to the target area. The target factors include time factors and weather factors. The processing unit 202 is used to determine a second prediction object according to the target factors and the target facility. It is judged whether there is a second prediction object in the first prediction object. When there is a second prediction object in the first prediction object, the target area is preferentially monitored according to the second prediction object.

[0066] In a possible implementation manner, the processing unit 202 is used to determine the data type according to the first prediction object. The data type includes power consumption data type, traffic data type, personnel location data type, environmental data type, and construction data type. The data type is input into a preset frequency database for matching to obtain a target acquisition frequency. Acquisition devices are determined according to the data type, and the target acquisition frequency is sent to each acquisition device, so that the acquisition devices perform data acquisition according to the target acquisition frequency.

[0067] In a possible implementation manner, the obtaining unit 201 is used to extract core data from the first emergency treatment plan. The processing unit 202 is used to determine a preset sensitive data table according to the predicted safety accident, judge whether there is core data in the preset sensitive data table, and judge whether the processing duration is less than or equal to a preset duration. When there is core data in the preset sensitive data table and the processing duration is less than or equal to the preset duration, it is determined that the first emergency treatment plan is at a high risk level, and the high risk level is output as the safety level. When there is core data in the preset sensitive data table and the processing duration is greater than the preset duration, it is determined that the first emergency treatment plan corresponds to a medium risk level, and the medium risk level is output as the safety level.

[0068] In a possible implementation manner, the processing unit 202 is configured to, when the first emergency response plan corresponds to a high-risk level, perform first encryption on the first emergency response plan according to the high-risk level to obtain a second emergency response plan, where the first encryption includes quantum encryption; the acquisition unit 201 is configured to acquire a second department, and the second department is the department involved in the first emergency response plan; the processing unit 202 is configured to screen out multiple preset departments identical to the second department from the target area, and one preset department corresponds to one location; the acquisition unit 201 is configured to acquire the target permission level corresponding to a third department, and the third department is any one of the multiple preset departments; the processing unit 202 is configured to determine whether the target permission level is consistent with the preset permission level, and the preset permission level is the permission level corresponding to the high-risk level; the sending unit is configured to, when the target permission level is consistent with the preset permission level, determine to allocate the second emergency response plan to the third department and output the third department as the first department.

[0069] In a possible implementation manner, the processing unit 202 is configured to, when the first emergency response plan corresponds to a medium-risk level, perform second encryption on the first emergency response plan according to the medium-risk level to obtain a third emergency response plan, where the second encryption is to encrypt only the core words in the first emergency response plan, and the processing plan consists of core words and non-core words, and the core words are the words appearing in the preset sensitive information table; the acquisition unit 201 is configured to acquire a second location corresponding to the predicted safety accident, and the second location is the specific location of the predicted safety accident in the target area; calculate the distances between multiple preset departments and the second location in sequence to obtain multiple distance values; sort the multiple distance values from small to large to obtain a target sorting result; the sending unit 203 is configured to acquire a fourth department, and the fourth department is the distance value ranked first in the target sorting result, and send the third emergency response plan to the fourth department.

[0070] In a possible implementation manner, the acquisition unit 201 is configured to acquire the target activities in the target area within a preset time, and the target activities include concert activities, sports meeting activities, and exhibition activities; the processing unit 202 is configured to combine the target activities with the target facilities to obtain target combination information; perform simulation analysis on the target combination information to obtain a third prediction object, and preferentially monitor the target area according to the third prediction object.

[0071] It should be noted that: when the system provided in the above embodiments implements its functions, only the above-mentioned division of each functional module is used for illustration. In practical applications, the above functions can be allocated to different functional modules according to needs, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the system and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be seen in the method embodiments, which will not be elaborated here.

[0072] The present application also discloses an electronic device. Referring to Figure 3 , Figure 3 FIG. is a schematic structural diagram of an electronic device provided by an embodiment of the present application. The electronic device 300 may include: at least one processor 301, at least one network interface 304, a user interface 303, a memory 302, and at least one communication bus 305.

[0073] Among them, the communication bus 305 is used to implement connection communication between these components.

[0074] Among them, the user interface 303 may include a display screen (Display) and a camera (Camera). Optionally, the user interface 303 may further include a standard wired interface and a wireless interface.

[0075] Among them, the network interface 304 may optionally include a standard wired interface and a wireless interface (such as a WI-FI interface).

[0076] Among them, the processor 301 may include one or more processing cores. The processor 301 connects various parts within the entire server through various interfaces and lines, and by running or executing instructions, programs, code sets, or instruction sets stored in the memory 302, as well as calling data stored in the memory 302, it executes various functions of the server and processes data. Optionally, the processor 301 may be implemented in at least one of the following hardware forms: digital signal processing (DSP), field-programmable gate array (FPGA), and programmable logic array (PLA). The processor 301 may integrate one or a combination of several of a central processing unit (CPU), a graphics processing unit (GPU), and a modem. Among them, the CPU mainly processes operating systems, user interfaces, and application requests, etc.; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 301 and may be implemented separately by a single chip.

[0077] Among them, the memory 302 may include a Random Access Memory (RAM), or may include a Read-Only Memory. Optionally, the memory 302 includes a non-transitory computer-readable storage medium. The memory 302 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 302 may include a program storage area and a data storage area. Among them, the program storage area can store instructions for implementing the operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area can store the data involved in the above-mentioned method embodiments. Optionally, the memory 302 may also be at least one storage device located far from the aforementioned processor 301.

[0078] As Figure 3 shown, the memory 302 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program for smart city processing based on a multimodal large model.

[0079] In Figure 3 the electronic device 300 shown, the user interface 303 is mainly used to provide an input interface for the user and obtain the data input by the user; while the processor 301 can be used to call the application program for smart city processing stored in the memory 302, and when executed by one or more processors, the electronic device is enabled to execute the method described in one or more of the above embodiments.

[0080] It should be noted that for the foregoing method embodiments, for the sake of simple description, they are all expressed as a series of action combinations. However, those skilled in the art should know that this application is not limited by the described action sequence, because according to this application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also know that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by this application.

[0081] In the above embodiments, the descriptions of the respective embodiments have their own emphases. For the parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0082] In several embodiments provided in the present application, it should be understood that the disclosed device can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the couplings or direct couplings or communication connections shown or discussed with each other can be through some service interfaces. The indirect couplings or communication connections of the devices or units can be in electrical or other forms.

[0083] The units described as separate components may or may not be physically separated. The components shown as units may or may not be physical units, that is, they can be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] In addition, in each embodiment of the present application, the functional units can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0085] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable memory. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or all or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions for causing a computer device (which can be a personal computer, a server or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. And the aforementioned memory includes: various media such as USB flash drives, mobile hard disks, magnetic disks or optical discs that can store program codes.

[0086] The above are only exemplary embodiments of the present disclosure, and the scope of the present disclosure cannot be limited thereby. That is, all equivalent changes and modifications made in accordance with the teachings of the present disclosure still fall within the scope covered by the present disclosure. Those skilled in the art will easily think of other implementation schemes of the present disclosure after considering the specification and the disclosure of the practical truth. The present application aims to cover any variations, uses or adaptive changes of the present disclosure, and these variations, uses or adaptive changes follow the general principles of the present disclosure and include the common general knowledge or conventional technical means in the technical field not recorded in the present disclosure.

Claims

1. A smart city processing method based on a multimodal large model, characterized in that, The method includes: Obtaining a target facility from a target area, where the target facility includes any one or more of public service facilities, commercial facilities, residential facilities, productive building facilities, and industrial building facilities, and the target area is the area corresponding to any city; Determining a first prediction object according to the target facility, where the first prediction object includes traffic accidents, medical accidents, natural disaster accidents, construction accidents, public security accidents, and park security accidents; Determining a plurality of collection devices according to the first prediction object, obtaining a plurality of monitoring data collected by the plurality of collection devices, and obtaining a first location corresponding to the target area; Inputting the plurality of monitoring data and the first location into a multi-modal large model for processing to obtain an analysis result; Determining a predicted safety accident according to the analysis result, and determining a first emergency treatment plan based on the predicted safety accident; Determining a safety level according to the processing duration and the predicted safety accident, and sending the first emergency treatment plan to a first department based on the safety level, so that the personnel corresponding to the first department can view the first emergency treatment plan, where the processing duration is the total duration for processing the first emergency treatment plan.

2. The method according to claim 1, wherein After determining the first prediction object according to the target facility, the method further includes: Obtaining a target factor currently corresponding to the target area, where the target factor includes a time factor and a weather factor; Determining a second prediction object according to the target factor and the target facility; Judging whether the second prediction object exists in the first prediction object; When the second prediction object exists in the first prediction object, preferentially monitoring the target area according to the second prediction object.

3. The method according to claim 1, characterized in that, Before obtaining a plurality of monitoring data collected by the plurality of collection devices and obtaining a first location corresponding to the target area, the method further includes: Determining a data type according to the first prediction object, where the data type includes an electricity consumption data type, a traffic data type, a personnel location data type, an environmental data type, and a construction data type; Inputting the data type into a preset frequency database for matching to obtain a target collection frequency; Determining the collection devices according to the data type, and sending the target collection frequency to each of the collection devices, so that the collection devices collect data according to the target collection frequency.

4. The method according to claim 1, characterized in that The determining the safety level according to the processing duration and the predicted safety accident specifically includes: Extracting core data from the first emergency treatment plan; Determining a preset sensitive data table according to the predicted safety accident; Judging whether the core data exists in the preset sensitive data table, and judging whether the processing duration is less than or equal to a preset duration; When the core data exists in the preset sensitive data table and the processing duration is less than or equal to the preset duration, determining that the first emergency treatment plan is at a high risk level, and outputting the high risk level as the safety level; When there is the core data in the preset sensitive data table and the processing duration is greater than the preset duration, determine that the first emergency handling plan corresponds to the medium risk level, and output the medium risk level as the security level.

5. The method according to claim 4, characterized in that, Sending the first emergency handling plan to the first department based on the security level specifically includes: When the first emergency handling plan corresponds to the high risk level, perform first encryption on the first emergency handling plan according to the high risk level to obtain a second emergency handling plan, where the first encryption includes quantum encryption; Obtain a second department, where the second department is the department involved in the first emergency handling plan; Search for multiple preset departments identical to the second department in the target area, and one preset department corresponds to one location; Obtain the target permission level corresponding to a third department, where the third department is any one of the multiple preset departments; Judge whether the target permission level is consistent with the preset permission level, where the preset permission level is the permission level corresponding to the high risk level; When the target permission level is consistent with the preset permission level, determine to allocate the second emergency handling plan to the third department, and output the third department as the first department.

6. The method according to claim 5, characterized in that, Sending the first emergency handling plan to the first department based on the security level specifically includes: When the first emergency handling plan corresponds to the medium risk level, perform second encryption on the first emergency handling plan according to the medium risk level to obtain a third emergency handling plan. The second encryption is to encrypt only the core words in the first emergency handling plan. The handling plan consists of core words and non-core words, and the core words are the words that appear in the preset sensitive information table; Obtain the second location corresponding to the predicted safety accident, where the second location is the specific location of the predicted safety accident in the target area; Calculate the distances between the multiple preset departments and the second location in sequence to obtain multiple distance values; Sort the multiple distance values from smallest to largest to obtain a target sorting result; Obtain a fourth department, where the fourth department is the department corresponding to the distance value ranked first in the target sorting result, and send the third emergency handling plan to the fourth department.

7. The method according to claim 1, characterized in that, After determining the first prediction object according to the target facility, the method further includes: Obtain the target activities in the target area within a preset time, where the target activities include concert activities, sports meeting activities, and exhibition activities; Combine the target activities with the target facility to obtain target combination information; Perform simulation analysis on the target combination information to obtain a third prediction object, and preferentially monitor the target area according to the third prediction object.

8. A smart city processing system based on a multimodal large model, characterized in that, The system includes an acquisition unit (201), a processing unit (202), and a sending unit (203), The obtaining unit (201) obtains a target facility from a target area, where the target facility includes any one or more of public service facilities, commercial facilities, residential facilities, productive building facilities, and industrial building facilities, and the target area is the area corresponding to any city; The processing unit (202) determines a first prediction object according to the target facility, where the first prediction object includes traffic accidents, medical accidents, natural disaster accidents, construction accidents, public security accidents, and park security accidents; determines a plurality of collection devices according to the first prediction object, obtains a plurality of monitoring data collected by the plurality of collection devices, and obtains a first position corresponding to the target area; inputs the plurality of monitoring data and the first position into a multimodal large model for processing to obtain an analysis result; determines a predicted safety accident according to the analysis result, and determines a first emergency treatment plan based on the predicted safety accident; The sending unit (203) determines a safety level according to the processing duration and the predicted safety accident, and sends the first emergency treatment plan to a first department based on the safety level, so that the personnel corresponding to the first department can view the first emergency treatment plan, where the processing duration is the total duration corresponding to processing the first emergency treatment plan.

9. An electronic device, characterized in that, It includes a processor (301), a memory (302), a user interface (303), and a network interface (304). The memory (302) is used to store instructions. The user interface (303) and the network interface (304) are used to communicate with other devices. The processor (301) is used to execute the instructions stored in the memory (302) so that the electronic device (300) executes the method according to any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions, and when the instructions are executed, the method according to any one of claims 1-7 is executed.

Images