Smart city emergency supervision method and system based on Internet of Things large model, and medium

Abstract

The embodiment of the invention provides a smart city emergency supervision method and system based on an Internet of Things large model and a medium, and the method comprises the steps: responding to an emergency management demand received from a sub-platform, and determining a data calling sequence of the emergency management demand according to a first emergency degree of the emergency management demand; and calling emergency management data corresponding to the emergency management demand from a database based on the data calling sequence. According to the method, multiple data sources are fused, dynamic scheduling and decision support are achieved based on an intelligent algorithm, the timeliness and effectiveness of emergency management are improved, it is ensured that emergencies can be efficiently coped with in various environments, and construction and development of a smart city are promoted.

Description

Technical Field

[0001] This specification relates to the field of smart cities, and in particular to a smart city emergency supervision method, system and medium based on an Internet of Things big model. Background Art

[0002] Traditional emergency management systems often have problems such as low data collection efficiency, slow response speed, and unreasonable resource scheduling, which makes it difficult to meet the needs of modern cities for rapid and accurate handling of emergency events. Especially in the face of emergencies, how to efficiently integrate multi-source data, dynamically adjust emergency resources, and achieve intelligent decision support has become a key issue that needs to be urgently solved in the smart city emergency supervision system.

[0003] Therefore, we hope to propose a smart city emergency supervision system based on the Internet of Things big model to solve the problems of the singleness and management blind spots of the traditional management model. The system will integrate multiple data sources, realize dynamic scheduling and decision support through intelligent algorithms, improve the timeliness and effectiveness of emergency management, ensure efficient response to emergencies in various environments, and promote the construction and development of smart cities. Summary of the invention

[0004] An embodiment of this specification provides a smart city emergency supervision method based on an Internet of Things large model. The method is executed by an emergency supervision management platform, and the method includes: responding to receiving an emergency management requirement from a sub-platform, determining the data retrieval sorting of the emergency management requirement according to the first emergency level of the emergency management requirement; based on the data retrieval sorting, retrieving emergency management data corresponding to the emergency management requirement from a database, including: retrieving a plurality of emergency management data corresponding to the emergency management requirement; according to the plurality of emergency management data and the second emergency level of each emergency management data in the plurality of emergency management data, retrieving a preset processing model corresponding to each emergency management data from a data processing model library; based on the preset processing model corresponding to each emergency management data, processing each emergency management data; sending the processed plurality of emergency management data to the corresponding sub-platform; within a preset period: obtaining the plurality of second emergency levels of the plurality of emergency management data in a plurality of geographical regions; for each geographical region in the plurality of geographical regions: determining the collection parameters of different emergency management data within the geographical region according to the plurality of second emergency levels, where the collection parameters include the patrol time and / or patrol frequency of emergency vehicles for different emergency management data, and the shooting angle and / or shooting frequency of a camera installed on the emergency vehicle at a patrol point; generating a patrol instruction according to the collection parameters of different emergency management data within the geographical region, and sending the patrol instruction to an emergency supervision object platform to control the emergency vehicles located within the geographical region to patrol within the geographical region according to the patrol time and / or patrol frequency, and controlling the camera at the patrol point to shoot according to the shooting angle and / or the shooting frequency to collect the corresponding emergency management data; when the emergency vehicle is patrolling, controlling an in-vehicle terminal installed on the emergency vehicle to detect an image taken by the emergency vehicle at the patrol point; receiving a warning instruction returned by the in-vehicle terminal, and sending the warning instruction to a display device for display.

[0005] An embodiment of this specification provides a smart city emergency supervision system based on an Internet of Things large model. The system includes: an emergency supervision management platform, an emergency supervision sensing network platform, and an emergency supervision object platform; the emergency supervision management platform is communicatively connected to the emergency supervision object platform through the emergency supervision sensing network platform; the emergency supervision management platform includes a sub-platform and a data center, and the sub-platform includes at least one of an emergency prevention sub-platform, an emergency monitoring sub-platform, a risk prevention sub-platform, and an emergency response sub-platform; the data center includes a database, a data processing model library, and a computing unit; the emergency supervision management platform is configured to execute a smart city emergency supervision method based on an Internet of Things large model.

[0006] One or more embodiments of this specification provide a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions in the storage medium, the computer executes a smart city emergency supervision method based on an Internet of Things large model.

[0007] Beneficial effects: The smart city emergency supervision method, system, and medium based on the Internet of Things large model in this specification can, by setting a first urgency level for emergency management requirements, process emergency management requirements in sequence according to the ranking of the first urgency level in the case of demand congestion, such as when multiple emergency management requirements are waiting to be processed, so that emergency management requirements with a high urgency level can be processed first. Furthermore, by determining the corresponding acquisition parameters according to the second urgency level of multiple emergency management data in multiple geographical regions and generating patrol instructions, the patrol effects of emergency vehicles in different geographical regions can be ensured. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] This specification will be further described by way of exemplary embodiments, which will be described in detail through the accompanying drawings. These embodiments are not restrictive. In these embodiments, the same numbers represent the same structures, where: Figure 1 is a schematic diagram of the system platform of a smart city emergency supervision system based on the Internet of Things large model shown in some embodiments of this specification; Figure 2 is an exemplary flowchart of a smart city emergency supervision method based on the Internet of Things large model shown in some embodiments of this specification; Figure 3 is an exemplary flowchart of retrieving emergency management data corresponding to emergency management requirements from a database shown in some embodiments of this specification; Figure 4 is an exemplary flowchart of determining the first emergency level of emergency management requirements shown in some embodiments of this specification; Figure 5 is an exemplary flowchart of controlling sensors in a geographical region to upload data according to data upload characteristics shown in some embodiments of this specification. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0009] To more clearly illustrate the technical solutions of the embodiments of this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some examples or embodiments of this specification. For those of ordinary skill in the art, without creative efforts, this specification can also be applied to other similar scenarios based on these drawings. Unless obvious from the language context or otherwise stated, the same reference numerals in the figures represent the same structures or operations.

[0010] Figure 1 It is a schematic diagram of the system structure of a smart city emergency supervision system based on the Internet of Things large model shown in some embodiments of this specification.

[0011] In some embodiments, as Figure 1 shown, the smart city emergency supervision system 100 based on the Internet of Things large model may include an emergency supervision management platform 110, an emergency supervision sensing network platform 120, and an emergency supervision object platform 130.

[0012] The emergency supervision management platform 110 refers to a digital monitoring and management platform for supervising emergency events across the region. Among them, an emergency event refers to a sudden, destructive, and harmful event. For example, natural disasters, accident disasters, public health events, social security events, etc.

[0013] In some embodiments, the emergency supervision management platform 110 may be configured in a processor and / or a server. The processor / server may process data and / or information obtained from other platforms. The processor / server may execute program instructions based on these data, information, and / or processing results to perform one or more functions described in this application.

[0014] The emergency supervision sensing network platform 120 refers to a platform for sensing and communicating smart city emergency supervision information, and is used for communication transmission to realize two-way data interaction between the emergency supervision management platform 110 and the emergency supervision object platform 130. For example, the emergency supervision sensing network platform may include communication devices, servers, and various gateway devices, etc.

[0015] The emergency supervision object platform 130 refers to an information processing platform for safely supervising various supervision objects related to emergency management. Among them, the supervision objects may be chemical factories manufacturing flammable and explosive products, transportation hubs, public activity places, etc. The emergency supervision object platform 130 may include a variety of monitoring, sensing, and interaction devices, such as cameras, fire alarms, hazardous gas leakage monitors, environmental monitoring sensors, etc.

[0016] In some embodiments, the emergency supervision management platform 110 is communicatively connected to the emergency supervision object platform 130 through the emergency supervision sensing network platform 120.

[0017] In some embodiments, the emergency supervision management platform 110 includes sub-platforms and a data center.

[0018] In some embodiments, the sub-platforms include at least one of an emergency prevention sub-platform, an emergency monitoring sub-platform, a risk prevention sub-platform, and an emergency response sub-platform.

[0019] The emergency prevention sub-platform refers to a management platform for the assessment and prevention of emergency events.

[0020] The emergency supervision sub-platform refers to a platform for monitoring, collecting, and analyzing emergency event data.

[0021] The risk prevention sub-platform refers to a platform for identifying potential risks, assessing risk levels, and implementing risk reduction strategies.

[0022] The emergency response sub-platform refers to a platform for coordinating, dispatching, and executing emergency plans after an emergency event occurs.

[0023] In some embodiments, the data center includes a database, a data processing model library, and a computing unit.

[0024] The database is used to collect, store, and manage a large amount of data related to emergency management. For example, MySQL, PostgreSQL, InfluxDB, and Prometheus, etc.

[0025] The data processing model library refers to a collection of data processing models for emergency management data processing.

[0026] The computing unit refers to a functional module for executing arithmetic, logical, and other instruction operations. The computing unit may include, but is not limited to, a central processing unit (CPU), etc.

[0027] In some embodiments, the smart city emergency supervision system based on the Internet of Things large model further includes an emergency supervision user platform and an emergency supervision service platform.

[0028] The emergency supervision user platform refers to an interactive platform for emergency management personnel and the public. In some embodiments, the emergency supervision user platform may include at least one personnel interaction device. For example, mobile phones, computers, etc.

[0029] The emergency supervision service platform refers to a platform that provides emergency supervision services. In some embodiments, the emergency supervision service platform may be configured as a server and can perform data interaction with the emergency supervision user platform and the emergency supervision management platform. For example, after the public discovers an emergency event, they use the emergency supervision user platform to report the event situation to the emergency supervision service platform, and the emergency supervision service platform reports the relevant event and its impact assessment to the emergency supervision management platform for the management department to respond and make decisions.

[0030] Figure 2 is an exemplary flowchart of the smart city emergency supervision method based on the Internet of Things large model shown in some embodiments of this specification. As Figure 2 shown, process 200 includes the following steps. In some embodiments, process 200 may be executed in the emergency supervision management platform.

[0031] Step 210: In response to receiving an emergency management requirement from a sub-platform, determine the data retrieval order of the emergency management requirement according to the first emergency level of the emergency management requirement. In some embodiments, step 210 is performed by the data center.

[0032] The emergency management requirement is an instruction sent by a sub-platform to retrieve one or more emergency management data. For example, the emergency management requirement is an instruction sent by the emergency prevention sub-platform to retrieve emergency management data related to accident and disaster prevention.

[0033] The first emergency level is the importance level of the emergency management requirement. The first emergency level may include multiple levels. Emergency management requirements with different levels of the first emergency level correspond to different importance levels.

[0034] In some embodiments, the first emergency level of the emergency management requirement can be preset by technicians or staff.

[0035] In some embodiments, the first emergency level of the emergency management requirement can also be determined according to one or more second emergency levels of one or more emergency management data corresponding to the emergency management requirement. For specific descriptions, see Figure 4 and its related content.

[0036] The data retrieval order is the order of retrieving one or more emergency management data corresponding to multiple emergency management requirements.

[0037] In some embodiments, the data center sorts according to the multiple first emergency levels of multiple emergency management requirements. For example, determine the data retrieval order of the emergency management requirement according to the order of the importance levels corresponding to the first emergency level from high to low.

[0038] Step 220: Based on the data retrieval order, retrieve the emergency management data corresponding to the emergency management requirement from the database. In some embodiments, step 220 is performed by the data center.

[0039] The emergency management data is various types of monitoring data required for emergency management. By way of example only, the emergency management data may include at least one or a combination of environmental temperature, environmental humidity, wind speed, whether there is an open fire in the environment, concentration of combustible gas / toxic gas, population size, quantity of inflammable and explosive items, and duration of gas outage, etc.

[0040] During the transmission of emergency management data, it will be marked with an electronic tag. The electronic tag includes the data type of the emergency management data and the geographical area to which it belongs. The electronic tag can also include the unique number of the device that collected the data, such as the unique numbers of sensors, cameras, etc. The data types include text, images, videos, audio, etc. The geographical area to which the emergency management data belongs is the geographical area where the device that collected the emergency management data is located. The geographical area can include administrative regions or grid regions, etc. The devices that collect emergency management data can include various sensors, cameras, etc.

[0041] In some embodiments, the data center directly retrieves one or more emergency management data corresponding to the emergency management requirements from the database based on the data retrieval sorting. In some embodiments, the data center can also process the retrieved emergency management data and send the processed emergency management data to the corresponding sub-platforms. For specific descriptions, please refer to Figure 3 and its related content.

[0042] Step 230: Obtain multiple second emergency levels of multiple emergency management data in multiple geographical areas within the previous preset period. In some embodiments, step 230 is executed by the emergency supervision and management platform.

[0043] The second emergency level is the importance level of the emergency management data. Emergency management data with different levels of the second emergency level correspond to different importance levels.

[0044] In some embodiments, the second emergency level of the emergency management data can be pre-set by technicians. In some embodiments, the second emergency level of the emergency management data can also be determined according to the retrieval frequency of the emergency management data within the previous preset period, the data type of the emergency management data, and the geographical area to which it belongs. For specific descriptions, please refer to Figure 4 and its related content.

[0045] The preset period is the period for determining and executing the data collection frequency. In some embodiments, the preset period is set by technicians or staff according to experience. For example, the preset period can be set to one day, one week, one month, etc.

[0046] The geographical area is the geographical area where the collection device corresponding to the emergency management data is located. The geographical area can be an administrative region, etc.

[0047] In some embodiments, the second emergency levels of the same type of emergency management data corresponding to multiple geographical areas can be different. Only as an example, for the emergency management data of the population scale category, the level of the second emergency level of the emergency management data in the urban central area is higher than the level of the second emergency level of the emergency management data in other areas outside the urban central area.

[0048] Step 240: For each of the multiple geographical regions, determine the acquisition parameters of different emergency management data within the geographical region according to the multiple second emergency levels. In some embodiments, step 240 is performed by the emergency supervision and management platform.

[0049] The acquisition parameters include the patrol time and / or patrol frequency of emergency vehicles for different emergency management data, and the shooting angle and / or shooting frequency of the cameras installed on the emergency vehicles at the patrol points.

[0050] Emergency vehicles are means of transportation for collecting emergency management data and / or handling emergency incidents. In some embodiments, emergency vehicles may include driverless vehicles, manually driven vehicles, drones, etc. In some embodiments, emergency vehicles may include fire trucks, medical vehicles, emergency rescue vehicles, power supply vehicles, mobile gas stations, etc.

[0051] The patrol time is one or more time points when the emergency vehicle starts the patrol action within a preset period, for example, starting patrol at 10:00 and 16:00 every day. The patrol frequency is the number of times the emergency vehicle patrols on the planned route within a preset period and the interval between the times, for example, patrolling 5 times a day with an interval of 2 hours each time.

[0052] The patrol points are the locations where the emergency vehicle needs to collect data on the planned route, for example, areas that need to be focused on.

[0053] The shooting angle includes the angles of the camera in the horizontal and vertical directions, the horizontal movement angle, the vertical movement angle, etc. The shooting frequency is the frame rate, for example, it can be 25 frames per second, etc.

[0054] In some embodiments, the acquisition parameters of different emergency management data within the geographical region can be determined according to the mean value of the second emergency level of the corresponding emergency management data in the previous preset period and a first preset table. The first preset table includes the acquisition parameters corresponding to each item of emergency management data under the mean values of different second emergency levels. The first preset table can be set by those skilled in the art according to experience.

[0055] Step 250: Generate a patrol instruction according to the acquisition parameters of different emergency management data within the geographical region, and send it to the emergency supervision object platform. In some embodiments, step 250 is performed by the emergency supervision and management platform.

[0056] Send the patrol instruction to the emergency supervision object platform so as to control the emergency vehicles located within the geographical region to patrol within the geographical region according to the patrol time and / or patrol frequency, and control the cameras at the patrol points to shoot according to the shooting angle and / or shooting frequency to collect the corresponding emergency management data.

[0057] The patrol instruction is an instruction including patrol time and / or patrol frequency, shooting angle and / or shooting frequency. For example, the patrol instruction directly controls the patrol vehicle and the camera in the form of a computer instruction.

[0058] Step 260, when the emergency vehicle is on patrol, control the built-in terminal set on the emergency vehicle to detect the images taken by the emergency vehicle at the patrol points. In some embodiments, step 260 is executed by the emergency supervision and management platform.

[0059] The built-in terminal is a processor for image detection, such as an image processor or a central processor.

[0060] In some embodiments, after the image processor of the built-in terminal obtains the image data, it can output the detection result of whether the image is an accident image. If the output detection result is an accident image, the built-in terminal can determine that an accident has occurred.

[0061] In some embodiments, after the built-in terminal detects an accident, it generates a warning instruction and sends it to the emergency supervision and management platform.

[0062] Step 270, receive the warning instruction returned by the built-in terminal and send the warning instruction to the display device for display. In some embodiments, step 260 is executed by the emergency supervision and management platform.

[0063] The warning instruction includes the accident type, the place of occurrence, and the time of occurrence. The accident type may include finding a fire point, a crowd density exceeding the threshold, etc.

[0064] In some embodiments, the emergency supervision and management platform sends the received warning instruction to the display device for display. The display device includes a display device associated with the user, including one or any combination of the warning center display screen, mobile phone terminal, computer terminal, etc.

[0065] By setting the first emergency level for emergency management requirements, in the case of demand congestion, for example, when multiple emergency management requirements are waiting to be processed, the emergency management requirements can be processed in sequence according to the ranking of the first emergency level, so that the emergency management requirements with a high emergency level can be given priority. Furthermore, by determining the corresponding acquisition parameters according to the second emergency level of multiple emergency management data in multiple geographical regions and generating patrol instructions, the patrol effect of emergency vehicles in different geographical regions can be ensured.

[0066] Figure 3 It is an exemplary flowchart for retrieving emergency management data corresponding to emergency management requirements from the database according to some embodiments of the present specification. As Figure 3 shown, process 300 includes the following steps. In some embodiments, process 300 can be executed by the data center.

[0067] Step 310: Retrieve multiple emergency management data corresponding to the emergency management requirements from the database.

[0068] In some embodiments, the data center retrieves multiple emergency management data corresponding to the emergency management requirements from the database according to the data retrieval sorting of the emergency management requirements.

[0069] Step 320: Retrieve the preset processing model corresponding to each emergency management data from the data processing model library according to the multiple emergency management data and the second emergency level of each emergency management data in the multiple emergency management data.

[0070] The data processing model library is a model library storing multiple preset processing models with different output accuracies and parameter scales.

[0071] The output accuracy is the degree of accuracy of the model output result. Taking the emergency management data as the population scale as an example, the different output accuracies of the corresponding data processing model can include retaining one decimal place, retaining two decimal places, etc.

[0072] The parameter scale is the number of model input parameters and / or the accuracy of the input parameters themselves. When the emergency management data includes emergency management data in the form of images, the input parameters include the accuracy of the images themselves, such as the resolution of the images.

[0073] In some embodiments, the data center sorts the multiple emergency management data according to the second emergency level of each emergency management data in descending order of the second emergency level, and then retrieves the preset processing model corresponding to each emergency management data in sequence according to the sorting.

[0074] In some embodiments, the preset processing models corresponding to the emergency management data belonging to different ranges of the second emergency level are different. The output accuracies and / or parameter scales of different preset processing models are different.

[0075] In some embodiments, the higher the emergency level, the smaller the output accuracy and / or parameter scale of the preset processing model.

[0076] The data center can select the corresponding preset processing model according to the range where the second emergency level of the emergency management data is located.

[0077] The preset processing model is a model for processing the emergency management data. The preset processing model can be a machine learning model, for example, a neural network model (NN). The input of the preset processing model is the emergency management data. The emergency management data can be text, images, etc. The output of the preset processing model is the information required by the emergency management requirements. Only as an example, when the input emergency management data is an image related to the population scale, the preset processing model processes the image and outputs the result related to the population scale.

[0078] In some embodiments, the data center may obtain a plurality of training samples, each training sample including emergency management data and its corresponding label, where the label is information required for emergency management needs, such as the number of the crowd size. The emergency management data in the training samples may be sourced from historical emergency management data, and its corresponding label may be the manually annotated information corresponding to the historical emergency management data.

[0079] In some embodiments, the data center may input the emergency management data in the training samples into an initial preset processing model to obtain the model prediction output corresponding to the training samples; according to the model prediction output and the label of the corresponding training sample, substitute them into the formula of a predefined loss function to calculate the value of the loss function; according to the value of the loss function, update the model parameters in the preset processing model in reverse, and this step can be performed using various methods. For example, it can be updated based on the gradient descent method; when the iteration end condition is met, end the iteration to obtain the preset processing model that has completed training.

[0080] In some embodiments, the data center may perform incremental training on a plurality of preset processing models in the data processing model library based on the emergency management data retrieved within a preset period. The training order of the plurality of preset processing models is determined according to the plurality of emergency management needs corresponding to the plurality of preset processing models within the preset period.

[0081] In some embodiments, the data center may extract a plurality of emergency management data retrieved within a preset period and their corresponding processed data. The processed data may directly serve as the label corresponding to the emergency management data, or the processed data may be used as the label corresponding to the emergency management data after manual correction. The data center may use the plurality of emergency management data and their corresponding labels as incremental training samples for training. The training process is similar to the above model training process and will not be elaborated here.

[0082] Merely as an example, if a preset processing model processes a plurality of emergency management data corresponding to multiple emergency management needs within a preset period, then the multiple emergency management needs are the multiple emergency management needs corresponding to the preset processing model within the preset period. For each preset processing model, the data center may determine the training order of the preset processing model according to the average value of the first emergency levels corresponding to the multiple emergency management needs of the preset processing model. For example, the greater the average value of the first emergency levels, the higher the training priority of the corresponding preset processing model.

[0083] Determining the incremental training order of different preset processing models according to emergency supervision requirements can improve the output effect of the preset processing models while improving the data processing efficiency of the system.

[0084] Step 330: Process each piece of emergency management data based on the preset processing model corresponding to it.

[0085] For more explanations about the preset processing model, refer to the content of Step 320.

[0086] Step 340: Send the processed multiple pieces of emergency management data to the corresponding sub-platforms.

[0087] In some embodiments, the data center sends each piece of processed emergency management data to the sub-platform that issued the emergency management requirement corresponding to the emergency management data. The corresponding sub-platform can then further prepare for the corresponding emergency management based on the received data information, such as mobilizing a corresponding number of crowd evacuation devices based on the scale of the accident population.

[0088] Processing the emergency management data according to the ranking of the second emergency level of the multiple pieces of emergency management data can enable the emergency management data with a higher emergency level to be processed first, obtain the results first, and improve the efficiency of emergency management.

[0089] Figure 4 It is an exemplary flowchart for determining the first emergency level of emergency management requirements shown in some embodiments of this specification. As Figure 4 shown, process 400 includes the following steps. In some embodiments, process 400 can be executed by the data center.

[0090] Step 410: Determine the retrieval frequency of the emergency management data based on the historical retrieval data corresponding to the emergency management data.

[0091] The historical retrieval data is the relevant data of the retrieval data history corresponding to the emergency management data in the past period of time. The past period of time can be, for example, the past month. The historical retrieval data includes the historical retrieval time and historical retrieval order of each retrieval data.

[0092] The retrieval frequency is the historical retrieval times of the emergency management data within a unit of time. The unit of time can be one day or one week, etc.

[0093] In some embodiments, the historical retrieval times of each piece of emergency management data can be determined based on the historical retrieval time of each retrieval of the emergency management data in the historical retrieval data, and then the historical retrieval times of each piece of emergency management data within a unit of time can be determined as the retrieval frequency of the emergency management data.

[0094] Step 420: Determine the second emergency level of the emergency management data based on the retrieval frequency, the data type of the emergency management data, and the geographical area to which it belongs.

[0095] In some embodiments, the data center can determine the second emergency level corresponding to each emergency management data through a vector database. The vector database includes feature vectors and labels corresponding to the feature vectors. The feature vectors are constructed from the retrieval frequencies, data types, and geographical regions of historical emergency management data in multiple historical retrievals. The label corresponding to the feature vector is the second emergency level corresponding to the historical emergency management data.

[0096] The label corresponding to the feature vector can be determined in the following manner: Obtain the actual retrieval order of the historical emergency management data corresponding to the feature vector in multiple historical retrievals, determine the average value of the multiple actual retrieval orders, and use this average value as the second emergency level of the emergency management data.

[0097] The data center can determine the vector to be matched corresponding to the current emergency management data according to the retrieval frequency, data type, and geographical region of the emergency management data, and determine the feature vector with the highest vector similarity to the vector to be matched in the vector database as the target vector, and use the label of the target vector as the second emergency level corresponding to the current emergency management data.

[0098] In some embodiments, the data center can determine the second emergency level of the emergency management data through an emergency model according to the retrieval frequency, data type, and geographical region of the emergency management data.

[0099] In some embodiments, the emergency model is a machine learning model. The emergency model is a Neural Network (NN) model.

[0100] The emergency model includes an associated feature extraction layer and an emergency level prediction layer. The associated feature extraction layer and the emergency level prediction layer can be trained separately or jointly.

[0101] The input of the associated feature extraction layer includes each emergency management data, the data type of each emergency management data, and the geographical region. The output of the associated feature extraction layer includes the associated features of each emergency management data.

[0102] The associated features include multiple associated data associated with the emergency management data. For example, when the associated data of emergency management data A are emergency management data B and emergency management data C, the associated feature of emergency management data A is emergency management data B and emergency management data C.

[0103] In separate training, the data feature extraction layer can be trained using the first training samples and the first labels. The data center can obtain, for each historical retrieval in multiple historical retrievals, multiple historical emergency management data, the data types of the multiple historical emergency management data, and the geographical regions to which they belong, and construct multiple first training samples. The first label is the associated feature corresponding to the multiple historical emergency management data corresponding to the first training sample.

[0104] The data center can obtain several historical retrievals before and after the historical retrieval corresponding to the historical emergency management data in the first training sample. If some of the historical emergency management data are retrieved in the same historical retrieval or the number of times they are retrieved in two adjacent historical retrievals is greater than the second preset threshold, then the historical emergency management data in this part of the historical emergency management data are associated data with each other.

[0105] The input of the emergency level prediction layer includes the retrieval frequency, associated features, data types, and geographical regions to which each emergency management data belongs. The output of the emergency level prediction layer includes the second emergency level of each emergency management data.

[0106] In separate training, the emergency level prediction layer can be trained using the second training samples and the second labels. The data center can obtain historical retrievals corresponding to emergency management with accident losses less than the preset threshold. For each historical retrieval, it constructs multiple second training samples based on the retrieval frequency, associated features, data types, and geographical regions of the multiple historical emergency management data. In some embodiments, the data center or other related devices can count the losses ultimately caused by the historical accidents corresponding to the historical retrievals as the accident losses corresponding to the historical retrievals. The preset threshold is set according to requirements. The second label is the second emergency level of the historical emergency management data in the second training sample. The data center can use the actual retrieval order of the historical emergency management data in the historical retrieval corresponding to the second training sample as the label corresponding to the second training sample.

[0107] The training processes of the associated feature extraction layer and the emergency level prediction layer in separate training are similar to the training process of the preset processing model in step 320, and will not be elaborated here.

[0108] By introducing the retrieval frequency and associated features of the emergency management data into the emergency model, the model can consider future retrieval situations when predicting the current input data, improving the accuracy of model prediction.

[0109] Step 430, determine the first emergency level of the emergency management requirement according to the multiple second emergency levels of the multiple emergency management data corresponding to the emergency management requirement.

[0110] In some embodiments, the data center may calculate the weighted sum of the second emergency levels of multiple emergency management data corresponding to the emergency management requirements, and use it as the first emergency level of the emergency management requirements. The weighting coefficients of the second emergency levels of each emergency management data are set according to experience.

[0111] In some embodiments, the data center determines the demand coincidence rate of each emergency management data among the multiple emergency management data corresponding to the emergency management requirements, and adjusts the first emergency levels of the multiple emergency management requirements according to the demand coincidence rate.

[0112] The demand coincidence rate is the retrieval coincidence rate of the emergency management data among multiple emergency management requirements.

[0113] In some embodiments, the data center receives multiple emergency management requirements within a period of time. The duration of this period can be set according to requirements. The multiple emergency management data among the multiple emergency management requirements can be used as a batch of emergency management data. The data center can determine the demand coincidence rate of the emergency management data retrieved in this batch. In some embodiments, for each emergency management data, the ratio of the number of emergency management requirements that retrieve this emergency management data in this batch retrieval to the total number of emergency management requirements retrieved in this batch is used as the demand coincidence rate of this emergency management data.

[0114] In some embodiments, for each emergency management requirement, the data center may calculate the mean value of the demand coincidence rates of the multiple emergency management data corresponding to it, and multiply the first emergency level by this mean value to obtain the adjusted first emergency level of the emergency management requirement.

[0115] When calculating the first emergency level of each emergency management requirement, also considering the demand coincidence rate of the emergency management data corresponding to each requirement, can enable the emergency management data with a higher demand coincidence rate to be retrieved preferentially, thereby improving the overall response speed of this batch of requirements.

[0116] Determining the second emergency level of the emergency management data by integrating the data type and the geographical area to which it belongs can preferentially process the dangerous data or important data in important areas or high-risk areas.

[0117] Figure 5 It is an exemplary flowchart for controlling sensors within a geographical area shown in some embodiments of this specification to upload data according to data upload characteristics. As Figure 5 shown, process 500 includes the following steps. In some embodiments, process 500 can be executed by the emergency supervision management platform.

[0118] Step 510, determine the second emergency level of the emergency management data corresponding to the sensor.

[0119] In some embodiments, the emergency management data includes the unique number information of the sensors. The emergency supervision and management platform can determine the emergency management data corresponding to the sensors according to the unique number information of the sensors included in the emergency management data, and then obtain the second emergency level of the emergency management data from the data center. For the description of obtaining the second emergency level of the emergency management data, see step 230 and its related content.

[0120] Step 520, determine the data upload characteristics of the sensors according to the second emergency level of the emergency management data corresponding to the sensors.

[0121] The data upload characteristics include the data upload amount and / or the data upload frequency. In some embodiments, the acquisition parameters include the data upload characteristics of the sensors within the geographical area. For the description of the acquisition parameters, see step 240 and its related content.

[0122] The data upload amount is the amount of data when the sensor uploads data each time. The data upload amount can be in bytes.

[0123] The data upload frequency is the number of times the sensor uploads data within a unit time.

[0124] In some embodiments, the emergency supervision and management platform calculates the mean value of the second emergency levels of one or more emergency management data corresponding to the sensors, and according to this mean value, through a third preset table, determines the data upload amount and data upload frequency of the sensors. The third preset table includes the data upload amount and data upload frequency corresponding to different emergency management data with the mean value of the second emergency level belonging to different second emergency level ranges. In some embodiments, the third preset table is pre-set by technicians according to experience.

[0125] In some embodiments, the data center can adjust the data upload characteristics of the sensors according to the average demand coincidence rate of the emergency management data corresponding to the sensors in multiple data retrievals within a preset period.

[0126] The average demand coincidence rate is the mean value of the demand coincidence rates of multiple emergency management data. For the description of the demand coincidence rate, see step 430 and its related content.

[0127] In some embodiments, when the average demand coincidence rate of multiple emergency management data corresponding to the sensors is greater than a preset threshold, the data center can increase the data upload frequency and data upload amount of the sensors by a corresponding preset adjustment amount. The preset threshold and the preset adjustment amount are set by technicians according to experience.

[0128] In some embodiments, the preset adjustment amount is related to the associated features of one or more emergency management data corresponding to the sensor. The more associated data there is between the one or more emergency management data and other emergency management data, the larger the preset adjustment amount. When there are more association relationships between the one or more emergency management data and other emergency management data, it indicates that the acquisition accuracy, acquisition volume, and acquisition frequency of this sensor have a greater impact on the entire emergency management. At this time, it is necessary to appropriately increase the preset adjustment amount to ensure that the sensor has sufficient acquisition accuracy and acquisition volume.

[0129] Adjusting the data upload characteristics of the sensor according to the average demand coincidence rate of the emergency management data corresponding to the sensor can enable important sensors (i.e., sensors that have a greater impact on the entire emergency management) to have sufficient acquisition frequency and acquisition volume, thereby facilitating the timeliness and accuracy of emergency management.

[0130] Step 530, generate an upload instruction according to the data upload characteristics of the sensor and send it to the emergency supervision object platform.

[0131] The upload instruction is used to control the sensors within the geographical area to upload data according to the data upload characteristics.

[0132] In some embodiments, after receiving the upload instruction, the emergency supervision object platform sends the upload instruction to the sensor, controls the sensor to upload data to the emergency supervision object platform according to the data upload characteristics, and then sends it to the emergency supervision management platform, such as the database therein.

[0133] Determining the data upload volume of the sensor through the second emergency level of the emergency management data can enable important sensors (i.e., sensors that have a greater impact on the entire emergency management) to have sufficient acquisition frequency and acquisition volume, thereby facilitating the timeliness and accuracy of emergency management.

[0134] The basic concepts have been described above. Obviously, for those skilled in the art, the above detailed disclosure is only an example and does not constitute a limitation to this specification. Although not explicitly stated here, those skilled in the art may make various modifications, improvements, and corrections to this specification. Such modifications, improvements, and corrections are suggested in this specification, so such modifications, improvements, and corrections still fall within the spirit and scope of the exemplary embodiments of this specification.

[0135] Finally, it should be understood that the embodiments described in this specification are only used to illustrate the principles of the embodiments of this specification. Other deformations may also fall within the scope of this specification. Therefore, by way of example rather than limitation, alternative configurations of the embodiments of this specification can be regarded as consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly introduced and described in this specification.

Claims

1. A smart city emergency supervision method based on the Internet of Things large model, characterized in that: The method is executed by the emergency supervision management platform. The method comprises: In response to receiving an emergency management demand from a sub-platform, determining a data retrieval order for the emergency management demand according to a first emergency level of the emergency management demand; Retrieving emergency management data corresponding to the emergency management requirements from a database based on the data retrieval sorting includes: Retrieving multiple emergency management data corresponding to the emergency management requirements; According to the plurality of emergency management data and the second emergency degree of each of the plurality of emergency management data, calling a preset processing model corresponding to each of the emergency management data from a data processing model library, wherein the preset processing model is a machine learning model; Processing each emergency management data based on the preset processing model corresponding to each emergency management data; The processed multiple emergency management data are sent to the corresponding sub-platform.

2. The method according to claim 1, characterized in that The method further comprises: within a preset period: Acquire multiple second emergency levels of multiple emergency management data in multiple geographical areas within a previous preset period; For each geographic region in multiple geographic regions: determining, according to the plurality of second emergency levels, collection parameters for different emergency management data within the geographic area; Generate patrol instructions according to the collection parameters of different emergency management data in the geographical area and send them to the emergency supervision object platform; when the emergency vehicle is patrolling, control the built-in terminal installed on the emergency vehicle to detect the images taken by the emergency vehicle at the patrol point; Receive the warning instruction returned by the built-in terminal, and send the warning instruction to the display device for display.

3. The method according to claim 1, characterized in that The method further comprises: Determining the frequency of retrieval of the emergency management data according to the historical retrieval data corresponding to the emergency management data; Determining a second emergency level of the emergency management data according to the retrieved frequency, the data type of the emergency management data, and the geographical area to which it belongs; The first emergency level of the emergency management requirement is determined according to the multiple second emergency levels of the multiple emergency management data corresponding to the emergency management requirement.

4. The method according to claim 3, characterized in that Determining the second emergency level of the emergency management data according to the retrieved frequency, the data type of the emergency management data, and the geographical area to which it belongs includes: The second emergency level of the emergency management data is determined based on the retrieval frequency of the emergency management data, the data type and the geographical area to which it belongs through an emergency model, and the emergency model is a machine learning model.

5. The method according to claim 1, characterized in that The acquisition parameters also include data upload characteristics of sensors disposed in the geographical area, the data upload characteristics including data upload amount and / or data upload frequency; The method further comprises: During the preset period: determining the second emergency level of the emergency management data corresponding to the sensor; Determining a data upload feature of the sensor according to the second emergency level of the emergency management data corresponding to the sensor; An upload instruction is generated according to the data upload characteristics of the sensor and sent to the emergency supervision object platform to control the sensors in the geographical area to upload data according to the data upload characteristics.

6. A smart city emergency supervision system based on the Internet of Things large model, characterized in that: The system includes: an emergency supervision management platform, an emergency supervision sensor network platform and an emergency supervision object platform; The emergency supervision management platform is connected to the emergency supervision object platform through the emergency supervision sensor network platform; The emergency supervision and management platform includes a sub-platform and a data center, and the sub-platform includes at least one of an emergency prevention sub-platform, an emergency monitoring sub-platform, a risk prevention sub-platform, and an emergency response sub-platform; The data center includes a database, a data processing model library and a computing unit; The emergency supervision management platform is configured to execute the smart city emergency supervision method based on the Internet of Things big model as described in right 1.

7. The system according to claim 6, characterized in that The data center is further configured to: Determining the frequency of retrieval of the emergency management data according to the historical retrieval data corresponding to the emergency management data; Determining a second emergency level of the emergency management data according to the retrieved frequency, the data type of the emergency management data, and the geographical area to which it belongs; The first emergency level of the emergency management requirement is determined according to the multiple second emergency levels of the multiple emergency management data corresponding to the emergency management requirement.

8. The system according to claim 7, characterized in that The data center is further configured to: The second emergency level of the emergency management data is determined based on the retrieval frequency of the emergency management data, the data type and the geographical area to which it belongs through an emergency model, and the emergency model is a machine learning model.

9. The system according to claim 6, characterized in that The acquisition parameters also include data upload characteristics of sensors disposed in the geographical area, the data upload characteristics including data upload amount and / or data upload frequency; The emergency supervision management platform is further configured as follows: During the preset period: determining the second emergency level of the emergency management data corresponding to the sensor; Determining a data upload feature of the sensor according to the second emergency level of the emergency management data corresponding to the sensor; An upload instruction is generated according to the data upload characteristics of the sensor and sent to the emergency supervision object platform to control the sensors in the geographical area to upload data according to the data upload characteristics.

10. A computer-readable storage medium, characterized in that: The storage medium stores computer instructions. When the computer reads the computer instructions in the storage medium, the computer executes the smart city emergency supervision method based on the Internet of Things large model as described in any one of claims 1-5.

Images