Intelligent city multi-modal emergency management system and method based on internet of things large model

Abstract

The specification provides a smart city multi-modal emergency management system and method based on an Internet of Things large model, which is executed by an emergency supervision management platform and includes: based on a preset period, for each sub-data center, based on the remaining computing resources, the reference computing resources and the first target data set, determining the second target data set and the target processing order; based on the first historical data, predicting the amount of data to be processed; based on the reference computing resources and the amount of data to be processed, predicting the resource occupation situation and generating the overload situation; and based on the target processing order of the plurality of sub-data centers, determining the data transmission order, and transmitting the second target data set based on the target transmission order.

Description

TECHNICAL FIELD

[0001] The present specification relates to the technical field of emergency management, and in particular to a smart city multi-modal emergency management system and method based on an Internet of Things large model. BACKGROUND

[0002] City emergency management provides various emergency rescue services for citizens by integrating various city emergency service resources, and provides protection for the public safety of the city. At present, the existing technology has problems such as unbalanced resource allocation, low coordination efficiency of each platform, and inability to obtain the data volume of emergency management data of different modalities at future time points in multi-modal emergency management data processing.

[0003] Therefore, it is hoped to propose a smart city multi-modal emergency management system and method based on an Internet of Things large model to timely and effectively process data and obtain the data volume of emergency management data at future time points, and ensure the stable operation of city emergency management. SUMMARY

[0004] Some embodiments of the present specification provide a smart city multi-modal emergency management system based on an Internet of Things large model, which includes an emergency supervision management platform, an emergency supervision sensor network platform, and an emergency supervision object platform; the emergency supervision management platform includes an emergency supervision general platform, a plurality of emergency supervision sub-platforms, and a plurality of sub-data centers; the emergency supervision management platform is configured to: based on a preset period, for each sub-data center, based on the remaining computing resources, the reference computing resources, and the first target data set, determine the second target data set and the target processing order; based on the first historical data, predict the to-be-processed data volume; based on the reference computing resources and the to-be-processed data volume, predict the resource occupation situation, and generate an overload situation; based on the target processing order of the plurality of sub-data centers, determine the data transmission order, and transmit the second target data set based on the data transmission order.

[0005] Some embodiments of the present specification provide a smart city multi-modal emergency management method based on an Internet of Things large model, which is executed by an emergency supervision management platform, and the method includes: based on a preset period, for each sub-data center, based on the remaining computing resources, the reference computing resources, and the first target data set, determining the second target data set and the target processing order; based on the first historical data, predicting the to-be-processed data volume; based on the reference computing resources and the to-be-processed data volume, predicting the resource occupation situation, and generating an overload situation; based on the target processing order of the plurality of sub-data centers, determining the data transmission order, and transmitting the second target data set based on the data transmission order.

[0006] Some embodiments of this specification provide a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes the aforementioned smart city multimodal emergency management method based on an Internet of Things (IoT) big data model.

[0007] Beneficial effects include: 1) Through a well-trained trend prediction model, the disaster development trend corresponding to the target handling sequence can be predicted relatively accurately. Based on the disaster development trend, the type and number of rescue vehicles can be determined, thereby generating rescue vehicle control instructions to control the corresponding type and number of rescue vehicles for rescue, which can save manpower and resources as much as possible while ensuring rescue efficiency; 2) By acquiring road condition data of rescue vehicles during their journey along the rescue route, risk areas can be identified and transmitted back to the emergency monitoring and management platform, enabling the emergency monitoring and management platform to have a more accurate understanding of the actual risk situation, thereby improving the targeting and timeliness of rescue. Attached Figure Description

[0008] This specification will be further described by way of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting; in these embodiments, the same reference numerals denote the same structures, wherein:

[0009] Figure 1 This is an exemplary structural diagram of a smart city multimodal emergency management system based on an Internet of Things (IoT) big data model, as shown in some embodiments of this specification.

[0010] Figure 2 This is an exemplary flowchart of a smart city multimodal emergency management method based on an IoT big data model, as shown in some embodiments of this specification.

[0011] Figure 3 This is an exemplary schematic diagram of a method for generating rescue vehicle control commands according to some embodiments of this specification;

[0012] Figure 4 This is an exemplary structural diagram of a trend prediction model shown in some embodiments of this specification. Detailed Implementation

[0013] To more clearly illustrate the technical solutions of the embodiments in this specification, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are merely some examples or embodiments of this specification. For those skilled in the art, these drawings can be applied to other similar scenarios without creative effort. Unless obvious from the context or otherwise specified, the same reference numerals in the drawings represent the same structures or operations.

[0014] Figure 1 This is an exemplary structural diagram of a smart city multimodal emergency management system based on an Internet of Things (IoT) big data model, as shown in some embodiments of this specification.

[0015] This specification provides some embodiments of a smart city multimodal emergency management system (hereinafter referred to as the "emergency management system") based on a large-scale Internet of Things (IoT) model. For example... Figure 1 As shown, the emergency management system 100 includes an emergency monitoring user platform 110, an emergency monitoring service platform 120, an emergency monitoring management platform 130, an emergency monitoring sensor network platform 140, and an emergency monitoring object platform 150.

[0016] The Internet of Things (IoT) big data model is a model capable of data processing and multimodal interaction. In some embodiments, the emergency management system 100 can implement urban multimodal emergency management methods through the IoT big data model.

[0017] The emergency monitoring user platform 110 is a platform for interacting with users. In some embodiments, the emergency monitoring user platform 110 can be configured as a terminal device.

[0018] In some embodiments, the emergency monitoring user platform 110 can send query instructions for emergency management data to the emergency monitoring management platform 130 via the emergency monitoring service platform 120, and receive data and / or information uploaded by the emergency monitoring service platform 120.

[0019] The emergency monitoring service platform 120 is a platform for receiving and transmitting data and / or information. In some embodiments, the emergency monitoring service platform 120 can receive query instructions issued by the emergency monitoring user platform 110 and forward them to the emergency monitoring management platform 130.

[0020] The emergency monitoring and management platform 130 can coordinate and integrate the connections and collaborations between various functional platforms, and gather all the information of the emergency management system 100, providing a platform for perception management and control management functions for the operation system of the emergency management system 100.

[0021] In some embodiments, such as Figure 1 As shown, the emergency supervision and management platform 130 includes an overall emergency supervision platform 131 and multiple emergency supervision sub-platforms (such as...). Figure 1 The emergency monitoring sub-platforms 132-1, 132-2, ..., 132-n and multiple sub-data centers shown are also mentioned. Figure 1 The sub-data center 1, sub-data center 2, ..., sub-data center n shown are included in the emergency monitoring platform 131, which includes the main data center.

[0022] The Emergency Supervision Platform 131 is a platform that integrates all information and performs data analysis.

[0023] The central data center can aggregate and store all emergency management data from the emergency management system 100. In some embodiments, the central data center can distribute data to each sub-data center and receive relevant data uploaded by the sub-data centers.

[0024] An emergency monitoring sub-platform is a platform that can integrate some information and perform data analysis. In some embodiments, the emergency monitoring management platform 130 includes multiple emergency monitoring sub-platforms, and each emergency monitoring sub-platform corresponds to a sub-data center. In some embodiments, the emergency monitoring sub-platform can interact with the sub-data center.

[0025] Sub-data centers can aggregate and store a portion of the emergency management data from the emergency management system 100. In some embodiments, sub-data centers can interact with the main data center of the emergency monitoring platform 131, the emergency monitoring sub-platforms, and the emergency monitoring sensor network platform 140. For example, a sub-data center can upload data (such as emergency management data) monitored in real time by the emergency monitoring object platform 150 transmitted via the emergency monitoring sensor network platform 140 to the main data center. Another example is that a sub-data center can send emergency management data to the corresponding emergency management sub-platform and obtain the processing results of the emergency management data from the emergency management sub-platform.

[0026] The emergency monitoring sensor network platform 140 is a functional platform for managing sensor communication. In some embodiments, the emergency monitoring sensor network platform 140 can be configured as a communication network and gateway to perform functions such as network management, protocol management, command management, and data parsing. In some embodiments, the emergency monitoring sensor network can receive data (such as emergency management data) monitored in real time by the emergency monitoring object platform 150 and upload it to the emergency monitoring management platform 130.

[0027] The emergency monitoring object platform 150 can be a functional platform for generating sensing information and executing control information. For example, the emergency monitoring object platform 150 includes intelligent gas valves, mobile emergency power vehicles, and display terminals on rescue vehicles. In some embodiments, the emergency monitoring object platform can collect emergency management data through various sensing devices (e.g., image sensors, sound sensors, cameras, etc.) and upload it to each sub-data center through the emergency monitoring sensor network platform 140. In some embodiments, the emergency monitoring object platform 150 can collect road condition data in real time and determine risk areas. In other embodiments, in response to receiving at least one of valve control commands, power vehicle control commands, and rescue vehicle control commands sent by the emergency monitoring management platform (such as the emergency monitoring central platform), the emergency monitoring object platform 150 can control the intelligent gas valve to automatically open or close based on its open / closed state, control the mobile emergency power vehicle to travel based on the driving route and provide power based on the power supply capacity, and / or control the display terminal set on the rescue vehicle to display the rescue route based on the rescue arrival time limit.

[0028] For more information on emergency management systems, please refer to the relevant descriptions below.

[0029] In some embodiments of this specification, the smart city multimodal emergency management system based on the Internet of Things big data model can form an information operation closed loop between the emergency monitoring object platform and the emergency monitoring user platform, and operate in a coordinated and regular manner under the unified management of the emergency monitoring management platform, thereby realizing the informatization and intelligentization of emergency management.

[0030] Figure 2 This is an exemplary flowchart illustrating a smart city multimodal emergency management method based on an IoT big data model, according to some embodiments of this specification. Figure 2 As shown, process 200 includes the following steps. In some embodiments, process 200 may be executed by an emergency monitoring and management platform (such as an emergency monitoring central platform).

[0031] Step 210: Based on a preset period, for each sub-data center, determine the second target dataset and the target processing order based on the remaining computing resources, reference computing resources, and the first target dataset.

[0032] In some embodiments, the preset cycle can be set by a technician based on experience.

[0033] The first target dataset refers to the collection of all pending emergency management data in the data center. In some embodiments, the first target dataset includes multiple emergency management data of different modalities to be processed.

[0034] Emergency management data refers to data related to emergency management. For example, emergency management data includes ambient temperature, flammable gas concentration, population size, number of flammable and explosive materials, and fire intensity. The modality of emergency management data refers to the data type of emergency management data. For example, the modalities of emergency management data include images, videos, audio, and text.

[0035] In some embodiments, the emergency monitoring object platform can collect emergency management data through various sensing devices (e.g., image sensors, sound sensors, cameras, etc.) and upload it to each sub-data center through the emergency monitoring sensor network platform.

[0036] It is worth noting that each piece of emergency management data carries an electronic tag during transmission. This tag includes the modality and geographical region corresponding to each data point. The geographical region of the emergency management data refers to the geographical area where the sensing device collecting the data is located.

[0037] Remaining computing resources refer to the remaining computing resources in a sub-data center. In some embodiments, the sub-data center can monitor and calculate resource usage in real time, thereby determining the remaining computing resources and sending this information to the central data center of the emergency monitoring platform. For more information on computing resource usage, please refer to step 230 and related instructions.

[0038] Reference computing resources refer to the computing resources required by the data centers to process different modalities of emergency management data in the first target dataset. In some embodiments, the emergency monitoring platform can calculate the average of the computing resources required by the data centers to process different modalities of emergency management data multiple times in historical data, and use this average as the reference computing resources.

[0039] The second target dataset refers to a collection of multiple emergency management data that need to be processed in a certain order. In some embodiments, the second target dataset includes multiple emergency management data that are to be sorted.

[0040] In some embodiments, for each sub-data center, the emergency monitoring platform may directly determine the first target dataset as the second target dataset in response to the fact that the remaining computing resources of each sub-data center are less than the reference computing resources.

[0041] In some embodiments, the emergency monitoring platform may also obtain a priority dataset from the first target dataset; and determine a second target dataset based on the priority dataset, the first target dataset, remaining computing resources, and reference computing resources.

[0042] A priority dataset refers to a collection of emergency management data that requires priority processing. In some embodiments, the priority dataset includes emergency management data with a hazard level higher than a preset risk threshold. For example, the priority dataset includes emergency management data corresponding to fires at gas stations or chemical storage sites. In some embodiments, the hazard level of emergency management data can be set by technicians within the electronic tag corresponding to each piece of emergency management data. The preset risk threshold can be set by technicians based on experience.

[0043] In some embodiments, the emergency monitoring platform can filter emergency management data with a hazard level higher than a preset risk threshold from a first target dataset and combine them to obtain a priority dataset.

[0044] In some embodiments, the emergency monitoring platform may, in response to the remaining computing resources being less than the reference computing resources, remove the priority dataset from the first target dataset and combine the remaining emergency management data to obtain the second target dataset. The emergency monitoring platform generates processing instructions for the priority dataset and the second target dataset and issues them to the corresponding sub-data centers, which then process the emergency management data in the priority dataset first, and then process the emergency management data in the second target dataset.

[0045] Understandably, emergency management data with a high risk level requires immediate processing. Some embodiments in this specification obtain a priority dataset by filtering emergency management data with a risk level exceeding a preset risk threshold, and then combining the remaining emergency management data to obtain a second target dataset. This lays the foundation for determining a more reasonable data processing order, thereby improving the speed of emergency management response.

[0046] The target processing order refers to the order in which all emergency management data in the second target dataset are processed. In some embodiments, technicians can preset priorities for emergency management data of different modalities. The overall emergency monitoring platform can, in response to the situation where the remaining computing resources of the sub-data centers are less than the reference computing resources, sort all emergency management data in the second target dataset according to the preset priorities from high to low, thereby obtaining the target processing order. The preset priorities can be set within the electronic tags corresponding to each piece of emergency management data.

[0047] Step 220: Based on the first historical data, predict the amount of data to be processed.

[0048] First historical data refers to unprocessed data from multiple time points within a preset period, representing multiple different modalities. In some embodiments, the emergency monitoring platform can retrieve first historical data from sub-data centers.

[0049] Data to be processed refers to emergency management data of different modalities that need to be processed at multiple points in time within the next preset period of the data center. In some embodiments, data to be processed includes the amount of data to be processed and its corresponding points in time.

[0050] The amount of data to be processed refers to the computing resources required to process the data. In some embodiments, for each modality, the emergency monitoring platform can obtain a prediction curve by fitting first historical data, and then determine the amount of data to be processed for that modality at multiple time points in the next preset period based on the prediction curve. The prediction curve is a curve showing the change of the amount of data to be processed for a certain modality over time at multiple time points within the preset period. In some embodiments, the fitting method includes, but is not limited to, linear regression models, multinomial regression models, etc.

[0051] In some embodiments, the emergency monitoring platform can also predict the amount of data to be processed for each sub-data center based on first historical data, second historical data, and regional characteristics using a data volume prediction model.

[0052] Secondary historical data refers to unprocessed data from multiple time points and modalities within the previous preset period of historical data. In some embodiments, the emergency monitoring platform can retrieve secondary historical data from sub-data centers. The previous preset period refers to the period preceding the current preset period.

[0053] Regional characteristics refer to the features related to the geographical area to which the emergency monitoring sub-platform corresponding to the sub-data center belongs. For example, regional characteristics include geographical features, population flow, and industrial type. Geographical features include whether the area is prone to geological activity; industrial type includes heavy industry, light industry, etc. Regional characteristics can be preset by technical personnel.

[0054] A data volume prediction model is a model used to predict the amount of data to be processed. In some embodiments, the data volume prediction model is a machine learning model. For example, the data volume model can be one or more combinations of a neural network (NN) model or other user-defined models.

[0055] In some embodiments, the inputs to the data volume prediction model include first historical data, second historical data, and regional features, and the output of the data volume prediction model includes the amount of data to be processed.

[0056] In some embodiments, the emergency monitoring platform can train a data volume prediction model based on multiple first training samples with first labels. For example, the platform can input the first training samples into the initial data volume prediction model, construct a loss function based on the output of the initial model and the first labels, iteratively update the parameters of the initial model based on the loss function, and terminate the iteration when the iteration termination condition is met, thus obtaining the trained data volume prediction model. Iterative update methods include, but are not limited to, gradient descent, and the iteration termination condition can be the convergence of the loss function or the reaching of a threshold number of iterations.

[0057] In some embodiments, the first training sample includes first historical data of samples within a first historical preset period, second historical data of samples within a second historical preset period, and sample region features. The first label includes the actual amount of data to be processed corresponding to the first training sample. The first label and the first training sample can be obtained based on historical data. The second historical preset period refers to the preset period preceding the first historical preset period.

[0058] Some embodiments in this specification utilize a trained data volume prediction model to predict the amount of data to be processed, which can effectively ensure the accuracy of the model's output results, thereby obtaining a more accurate amount of data to be processed.

[0059] In some embodiments, the input to the data volume prediction model may also include the disaster development trend corresponding to the target processing order.

[0060] Disaster development trend refers to the changing trend of a disaster over a future period of time. For example, disaster development trend includes potential disasters, the scope of their impact, and the potential losses they may cause. In some embodiments, when emergency management data is processed according to different target processing sequences, different disasters may occur over the future, and their development trends may also differ; therefore, different target processing sequences correspond to different disaster development trends.

[0061] In some embodiments, a disaster development trend can be characterized by one of the following: the probability of a disaster occurring in the future, the extent of the disaster's impact, and the potential losses caused by the disaster. For example, it can be characterized by the potential losses caused by the disaster.

[0062] For more information on disaster development trends, please refer to [link / reference]. Figures 3-4 And its related descriptions.

[0063] In some embodiments, when the input to the data volume prediction model includes the disaster development trend corresponding to the target processing order, the first training samples also include the disaster development trend corresponding to the target processing order of samples within a first historical preset period. The emergency monitoring platform can determine the disaster development trend corresponding to the target processing order of samples within the first historical preset period based on the actual disaster situation within a preset time period after the first historical preset period.

[0064] Understandably, disasters of varying severity can affect the frequency of emergency management data collection, thereby impacting the amount of data to be processed in the next preset cycle. Some embodiments in this specification, by using the disaster development trend corresponding to the target processing order as input to the data volume prediction model, further consider the impact of the disaster development trend on the amount of data to be processed, thereby further improving the accuracy of the model's predictions and making the predicted amount of data to be processed more accurate.

[0065] Step 230: Based on the reference computing resources and the amount of data to be processed, predict the resource usage and generate overload conditions.

[0066] Resource utilization refers to the computing resource utilization of a data center at multiple points in time within the next preset period. Resource utilization can be expressed as the ratio of the computing resources used by the data center to process multiple different modalities of data at multiple points in time within the next preset period to its total computing resources.

[0067] In some embodiments, the emergency monitoring platform can calculate the computing resources occupied by each sub-data center when processing multiple different modes of data at multiple time points within the next preset period, based on the amount of data to be processed and reference computing resources. Then, it can calculate the ratio of the data occupied by each sub-data center to the total computing resources of the corresponding sub-data center, thereby obtaining the resource occupancy of each sub-data center at multiple time points within the next preset period.

[0068] An overload situation refers to a situation where a data center consumes more computing resources than its total computing resources when processing multiple different modalities of data at multiple points in time within the next preset cycle. In some embodiments, an overload situation includes overload points and overload amounts.

[0069] An overload point refers to a point in time when the computing resources used by a sub-data center to process multiple different modalities of data to be processed within a preset period exceed its total computing resources. In some embodiments, the emergency monitoring platform can determine one or more points in time when the resource occupancy of each sub-data center exceeds 100% within the next preset period, based on the resource occupancy status of each sub-data center at multiple points in the next preset period.

[0070] Overload refers to the portion of computing resources used by a data center that exceeds its total computing resources when processing multiple data points of different modes within a preset period. In some embodiments, the emergency monitoring platform can determine the overload amount corresponding to the overload point as the difference between the computing resources corresponding to the overload point and the total computing resources of the data center.

[0071] Step 240: Based on the target processing order of multiple sub-data centers, determine the data transmission order, and transmit the second target dataset based on the data transmission order.

[0072] The data transmission order refers to the sequence in which all emergency management data in the second target dataset are transmitted. In some embodiments, the overall emergency monitoring platform can determine the target processing order of each sub-data center as the corresponding data transmission order based on the target processing order of multiple sub-data centers, and the corresponding sub-data center will transmit the second target dataset based on this data transmission order.

[0073] Some embodiments in this specification determine the target processing order based on the priority of emergency management data. By transmitting and processing emergency management data according to the data transmission order and the target processing order, more important emergency management data can be processed first, while improving data processing efficiency and reducing data processing time.

[0074] In some embodiments, the emergency monitoring platform can also generate resource control instructions based on the overload status and the amount of data to be processed in multiple sub-data centers, and send the resource control instructions to multiple sub-data centers to control the multiple sub-data centers to clear cache space and / or adjust transmission bandwidth.

[0075] Resource control commands are generated based on the current resource usage and performance indicators of the emergency management system, and are used to guide multiple sub-data centers to perform resource cleanup and optimization operations. In some embodiments, the emergency monitoring platform can determine resource control parameters based on the overload status and the amount of data to be processed in multiple sub-data centers, and then automatically generate resource control commands based on the resource control parameters through a preset program. The preset program can be set in advance by those skilled in the art.

[0076] Resource control parameters are parameters used for dynamically managing and allocating resources for sub-data centers. In some embodiments, resource control parameters include the cache size that each sub-data center needs to clear at multiple points in time within the next preset period, and / or the transmission bandwidth allocated to each sub-data center.

[0077] In some embodiments, for each overload point, the emergency monitoring center can determine the amount of cache to be cleared from the sub-data center based on the overload level, ensuring that the remaining computing resources of the sub-data center are not less than the reference computing resources. The amount of cache to be cleared must be no less than the overload level. For more information on remaining computing resources and reference computing resources, see [link to relevant documentation]. Figure 2 The relevant description of step 210.

[0078] In other embodiments, the emergency monitoring platform can adjust the transmission bandwidth of each sub-data center at multiple points in the next preset period so that the transmission bandwidth of each sub-data center at each point in time is not less than the corresponding amount of data to be processed.

[0079] Transmission bandwidth can be used to describe the amount of data that a data transmission channel (such as a network link, communication line, etc.) can transmit per unit of time. In some embodiments, the emergency monitoring platform can adjust the transmission bandwidth of multiple sub-data centers at multiple points in time within the next preset period based on resource control parameters and the transmission bandwidth allocated to each sub-data center.

[0080] It should be noted that the emergency monitoring platform can control multiple sub-data centers to clear cache space or adjust transmission bandwidth according to the actual situation, or simultaneously control multiple sub-data centers to clear cache space and adjust transmission bandwidth.

[0081] In some embodiments of this specification, resource regulation parameters are generated based on accurate predictions of overload time points and multimodal data volumes. Redundant caches can be proactively cleared and transmission bandwidth allocated before the next preset cycle, which can effectively prevent system crashes caused by sudden data congestion.

[0082] In some embodiments, the emergency monitoring platform can also generate at least one of valve control instructions, power supply vehicle control instructions, and rescue vehicle control instructions based on the second target dataset, target processing order, and overload conditions of multiple sub-data centers, and send them to the emergency monitoring object platform to control the intelligent gas valve to automatically open or close based on its open / closed state, control the mobile emergency power supply vehicle to travel based on its route and provide power based on its power supply capacity, and / or control the display terminal set on the rescue vehicle to display the rescue route based on the rescue arrival time limit.

[0083] Valve control commands are used to automatically open or close intelligent gas valves. Power vehicle control commands are used to control the mobile emergency power vehicle to travel to the corresponding geographical area based on the driving route and to supply power based on the available power. Rescue vehicle control commands are used to control the display terminal installed on the rescue vehicle to display the rescue route based on the rescue arrival time limit. The display terminal is the display device for interaction between the rescue vehicle and rescue personnel. For example, the display terminal includes a vehicle-mounted display screen.

[0084] In some embodiments, the emergency monitoring platform can determine emergency management parameters based on a second target dataset from multiple sub-data centers, the target processing order, and overload conditions. Then, based on these emergency management parameters, it can automatically generate at least one of valve control commands, power supply vehicle control commands, and rescue vehicle control commands through a preset program. The preset program can be pre-set by someone skilled in the art.

[0085] Emergency management parameters refer to parameters related to emergency management. In some embodiments, emergency management parameters include the open / closed status of intelligent gas valves, the travel route and power output of mobile emergency power vehicles, and the arrival time and route of rescue vehicles. Rescue vehicles include gas emergency repair vehicles, fire trucks, ambulances, and search and rescue vehicles. The arrival time refers to the latest time a rescue vehicle can arrive at the geographic area to which each emergency management data belongs. The rescue route refers to the route taken by the rescue vehicle to the geographic area to which each emergency management data belongs.

[0086] The emergency monitoring platform can determine emergency management parameters through various methods based on the second target dataset from multiple sub-data centers, the target processing order, and the overload time points and corresponding overload amounts within the next preset cycle. For more information on overload time points and overload amounts, please refer to [link to relevant documentation]. Figure 2 The relevant description of step 230.

[0087] In some embodiments, the emergency monitoring platform can, based on a second target dataset from multiple sub-data centers, determine the opening and closing status of the smart gas valve, the driving route of the mobile emergency power vehicle, and the power supply capacity of the emergency management data corresponding to the data by querying a first preset table for emergency management data whose processing time in the second target dataset exceeds a first preset threshold.

[0088] Processing time refers to the time spent processing a specific piece of emergency management data. In some embodiments, the emergency monitoring platform can determine the time point for processing a specific piece of emergency management data and the time point for processing the next piece of emergency management data based on the target processing order, and then calculate the processing time by calculating the difference between the two time points.

[0089] The first preset table can reflect the difference between emergency management data, processing time, and a first preset threshold, as well as the correlation between the opening and closing status of intelligent gas valves, the travel route of mobile emergency power vehicles, and power supply. In some embodiments, the first preset table can be constructed by technicians based on experience. The first preset threshold can be determined by technicians based on experience.

[0090] In some embodiments, the emergency monitoring platform can also determine the rescue arrival time and rescue route of the rescue vehicle corresponding to each emergency management data in the second dataset by querying a second preset table, based on the second target dataset of multiple sub-data centers and the overload time points and their corresponding overload amounts in the next preset period.

[0091] The second preset table can reflect emergency management data, the number of overload points and the average of their corresponding overload amounts, as well as the correlation between the arrival time of rescue vehicles and rescue routes. In some embodiments, the second preset table can be constructed by technicians based on experience.

[0092] For more information on how the emergency monitoring platform generates rescue vehicle control commands, please refer to [link / reference needed]. Figure 3 And its related descriptions.

[0093] In some embodiments, in response to receiving at least one of the valve control command, power vehicle control command, and rescue vehicle control command sent by the emergency monitoring platform, the emergency monitoring target platform may control the intelligent gas valve to automatically open or close based on its open / closed state, control the mobile emergency power vehicle to travel based on its route and provide power based on its power supply capacity, and / or control the display terminal set on the rescue vehicle to display the rescue route based on the rescue arrival time limit.

[0094] In some embodiments, in response to receiving a valve control command, the emergency monitoring platform can automatically generate a corresponding control signal and send it to the smart gas valve to control the smart gas valve to automatically open or close based on its open / closed state. For example, when the smart gas valve is closed, the emergency monitoring platform can control the smart gas valve to automatically open in response to receiving the valve control command.

[0095] In some embodiments, in response to receiving a power supply vehicle control command, the emergency monitoring platform can automatically generate a corresponding control signal and send it to the mobile emergency power supply vehicle to control the vehicle to travel to the corresponding geographical area based on a route and provide power based on the available power. For example, when a disaster causes a power outage, the mobile emergency power supply vehicle can provide power to the affected area.

[0096] In some embodiments, in response to receiving a rescue vehicle control command, the emergency monitoring platform can automatically generate a corresponding control signal and send it to the display terminal to control the display terminal to automatically display the rescue route based on the rescue arrival time limit.

[0097] For more information on how the emergency monitoring platform controls the display terminal to automatically display rescue routes, please refer to the relevant descriptions below.

[0098] Understandably, when a data center is overloaded, taking timely emergency management measures based on emergency management parameters to prevent problems in advance can not only improve the timeliness of the emergency management system's response to various emergencies, but also effectively avoid greater losses caused by untimely data processing.

[0099] In some embodiments, while the rescue vehicle is traveling along a rescue route, the rescue vehicle is configured to identify risk areas based on road condition data and transmit the data back to the emergency monitoring and management platform.

[0100] Road condition data refers to data related to the road conditions encountered by rescue vehicles during their journey. For example, road condition data includes the types and concentrations of hazardous gases at accident scenes along the road, thermal imaging images, or color images. In some embodiments, road condition data can be collected using sensing devices (such as gas sensors, thermal imagers, and cameras) mounted on the rescue vehicle.

[0101] In some embodiments, a terminal (such as an in-vehicle information terminal, an in-vehicle computer, etc.) installed on a rescue vehicle can determine the risks on the rescue route based on road condition data by querying a third preset table, and identify areas with risks as risk areas.

[0102] The third preset table reflects the correlation between road condition data and risk type and risk level. Risk types include, but are not limited to, crowd gatherings, building collapses, and smoke dispersal. Risk level reflects the severity of the risk. A risk level greater than 0 indicates the presence of a risk. In some embodiments, the third preset table can be constructed based on historical data.

[0103] In some embodiments, the extent of a risk area is directly related to the degree of risk. The higher the degree of risk, the larger the risk area. Here, the extent of a risk area refers to the range or impact of the risk. The extent of a risk area can be determined based on road condition data collection points. Different risk types can be represented by different colors.

[0104] Some embodiments in this specification acquire road condition data of rescue vehicles traveling along rescue routes to identify risk areas and transmit the data back to the emergency monitoring and management platform. This enables the emergency monitoring and management platform to have a more accurate understanding of the actual risk situation, thereby improving the targeting and timeliness of rescue efforts.

[0105] It should be noted that the above description of process 200 is for illustrative purposes only and does not limit the scope of this specification. Those skilled in the art can make various modifications and changes to the above process under the guidance of this specification. However, these modifications and changes remain within the scope of this specification.

[0106] Figure 3 This is an exemplary schematic diagram of a method for generating control commands for a rescue vehicle according to some embodiments of this specification.

[0107] In some embodiments, the rescue vehicle control command includes the type and quantity of rescue vehicles. In some embodiments, such as... Figure 3 As shown, the emergency monitoring and management platform (such as the emergency monitoring general platform) can determine the disaster development trend 350 corresponding to the target processing order based on the second target dataset 310, target processing order 320, and overload situation 330 of multiple sub-data centers, through the trend prediction model 340; and generate rescue vehicle control instructions 360 based on the disaster development trend 350.

[0108] For more information on the second target dataset, target processing order, overload conditions, disaster development trends, and rescue vehicle control instructions, please refer to [link / reference needed]. Figure 2 And its related descriptions.

[0109] Trend prediction models are used to predict the disaster development trend corresponding to the target treatment sequence. In some embodiments, the trend prediction model can be a machine learning model. For example, the trend prediction model may include one or more combinations of deep neural network (DNN) models or other user-defined models.

[0110] Figure 4 This is an exemplary structural diagram of a trend prediction model shown in some embodiments of this specification. For example... Figure 4 As shown, the trend prediction model 340 includes a time-consuming prediction layer 341 and a loss prediction layer 342.

[0111] The time-consuming prediction layer is used to determine the total time required for the target processing order. In some embodiments, the time-consuming prediction layer is a machine learning model. For example, the time-consuming prediction layer may include a DNN model, etc.

[0112] In some embodiments, such as Figure 4 As shown, the inputs to the time-consuming prediction layer 341 include the target processing order 320, the overload situation 330, and a modality 410 of multiple emergency monitoring data to be sorted. The output of the time-consuming prediction layer 341 includes the total time 420 corresponding to the target processing order. For more information on emergency management data and modalities, please refer to [link to relevant documentation]. Figure 2 And its related descriptions.

[0113] In some embodiments, the emergency monitoring platform can train a time consumption prediction layer based on multiple second training samples with second labels. The training process of the time consumption prediction layer is similar to that of the data volume prediction model; see [link to relevant documentation] for more details. Figure 2 The relevant description of step 220.

[0114] In some embodiments, the second training samples include the sample processing order in the historical processing, the modality of the sample emergency management data in the sample second target dataset, and the sample overload situation. The second label includes the actual total time spent processing the sample second target dataset based on the sample processing order in the historical processing. The second training samples and the second label can be obtained based on historical data.

[0115] The loss prediction layer is used to predict the disaster development trend corresponding to the target treatment sequence. In some embodiments, the loss prediction layer is a machine learning model. For example, the loss prediction model may include a DNN model, etc.

[0116] In some embodiments, such as Figure 4 As shown, the inputs of the loss prediction layer 342 include the second target dataset 310, the target processing order 320, and the total time 420 corresponding to the target processing order. The output of the loss prediction layer 342 includes the disaster development trend 350 corresponding to the target processing order.

[0117] In some embodiments, the emergency monitoring platform can train a loss prediction layer based on multiple third training samples with third labels. The training process of the loss prediction layer is similar to that of the data volume prediction model; for more details, see [link to relevant documentation]. Figure 2 The relevant description of step 220.

[0118] In some embodiments, the third training sample includes the sample second target dataset from the historical processing process, the sample processing order, and the total time consumed corresponding to the sample processing order. The third label includes, after processing the sample second target dataset based on the sample processing order during the historical processing process, the average of the actual disasters that occurred, the average of the actual impact range of the disasters, and the average of the actual losses caused by the disasters. The third training sample and the third label can be obtained based on historical data.

[0119] In some embodiments, the emergency monitoring platform can determine the type and number of rescue vehicles by querying a fourth preset table based on the disaster development trend. The fourth preset table reflects the correlation between potential disasters, their development trends, and the types and numbers of rescue vehicles. This fourth preset table can be constructed by technical personnel based on experience.

[0120] In some embodiments, the emergency monitoring platform can automatically generate rescue vehicle control instructions based on the type and number of rescue vehicles through a preset program, and send them to the emergency monitoring target platform to control the display terminals on the corresponding type and number of rescue vehicles to display the rescue route based on the rescue arrival time limit. The preset program can be set in advance by those skilled in the art.

[0121] Some embodiments in this specification use a trained trend prediction model to process the second target dataset, target processing order, and overload situation of multiple data centers. This can accurately predict the disaster development trend corresponding to the target processing order. Based on the disaster development trend, the type and number of rescue vehicles are determined, thereby generating rescue vehicle control instructions to control the corresponding type and number of rescue vehicles for rescue. This can save manpower and resources as much as possible while ensuring rescue efficiency.

[0122] Some embodiments of this specification also provide a computer-readable storage medium that stores computer instructions. When a computer reads the computer instructions from the storage medium, the computer executes the smart city multimodal emergency management method based on the Internet of Things big model described above.

[0123] The basic concepts have been described above. Obviously, for those skilled in the art, the detailed disclosure above is merely illustrative and does not constitute a limitation of this specification. Although not explicitly stated herein, 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 and therefore remain within the spirit and scope of the exemplary embodiments described herein.

Claims

1. A smart city multimodal emergency management system based on an Internet of Things (IoT) big data model, characterized in that: The system includes an emergency monitoring and management platform, an emergency monitoring sensor network platform, and an emergency monitoring object platform; the emergency monitoring and management platform includes a general emergency monitoring platform, multiple emergency monitoring sub-platforms, and multiple sub-data centers. The emergency monitoring and management platform is configured as follows: Based on a preset period, for each sub-data center, Obtain the preferred dataset from the first target dataset; Based on the priority dataset, remaining computing resources, reference computing resources, and the first target dataset, a second target dataset is determined; The reference computing resources refer to the computing resources required by the sub-data center to process emergency management data of different modalities in the first target dataset; The second target dataset is sorted according to a preset priority to determine the target processing order; Based on the first historical data, the second historical data, and regional characteristics, the amount of data to be processed is predicted using a data volume prediction model, which is a machine learning model; the regional characteristics refer to the features related to the geographical region to which the emergency monitoring sub-platform corresponding to the sub-data center belongs. Based on the reference computing resources and the amount of data to be processed, predict resource usage and generate overload conditions; Based on the target processing order of the multiple sub-data centers, the data transmission order is determined, and the second target dataset is transmitted based on the data transmission order; The emergency monitoring and management platform is also configured as follows: Based on the overload situation of the multiple data centers and the amount of data to be processed, resource control parameters are determined. The resource control parameters include multiple time points in the next preset period, the cache size to be cleared for each data center, and / or the transmission bandwidth allocated to each data center. Based on the resource control parameters, a resource control instruction is generated; The resource control instructions are sent to the multiple sub-data centers to control the multiple sub-data centers to clear cache space and / or adjust the transmission bandwidth.

2. The system as described in claim 1, characterized in that, The emergency monitoring and management platform is also configured as follows: Based on the second target dataset from the multiple sub-data centers, the target processing order, and the overload situation, at least one of valve control instructions, power supply vehicle control instructions, and rescue vehicle control instructions is generated and sent to the emergency monitoring object platform to control the intelligent gas valve to automatically open or close based on its open / closed state, control the mobile emergency power supply vehicle to travel based on its route and provide power based on its power supply capacity, and / or control the display terminal set on the rescue vehicle to display the rescue route based on the rescue arrival time limit.

3. The system as described in claim 2, characterized in that, During the course of the rescue vehicle's journey along the rescue route, the rescue vehicle is configured to identify risk areas based on road condition data and transmit the data back to the emergency monitoring and management platform.

4. The system as described in claim 2, characterized in that, The rescue vehicle control commands include the type and number of rescue vehicles; The emergency monitoring and management platform is also configured as follows: Based on the second target dataset from the multiple data centers, the target processing order, and the overload situation, a trend prediction model is used to determine the disaster development trend corresponding to the target processing order. The trend prediction model is a machine learning model. Based on the development trend of the disaster, the command to control the rescue vehicle is generated.

5. A smart city multimodal emergency management method based on an Internet of Things (IoT) big data model, executed by an emergency monitoring and management platform, characterized in that: include: Based on a preset period, for each sub-data center, Obtain the preferred dataset from the first target dataset; Based on the priority dataset, remaining computing resources, reference computing resources, and the first target dataset, a second target dataset is determined; The reference computing resources refer to the computing resources required by the sub-data center to process emergency management data of different modalities in the first target dataset; The second target dataset is sorted according to a preset priority to determine the target processing order; Based on the first historical data, the second historical data, and regional characteristics, the amount of data to be processed is predicted using a data volume prediction model, which is a machine learning model; the regional characteristics refer to the features related to the geographical region to which the emergency monitoring sub-platform corresponding to the sub-data center belongs. Based on the reference computing resources and the amount of data to be processed, predict resource usage and generate overload conditions; Based on the target processing order of multiple sub-data centers, the data transmission order is determined, and the second target dataset is transmitted based on the data transmission order; Based on the overload situation of the multiple data centers and the amount of data to be processed, resource control parameters are determined. The resource control parameters include multiple time points in the next preset period, the cache size to be cleared for each data center, and / or the transmission bandwidth allocated to each data center. Based on the resource control parameters, a resource control instruction is generated; The resource control instructions are sent to the multiple sub-data centers to control the multiple sub-data centers to clear cache space and / or adjust the transmission bandwidth.

6. The method as described in claim 5, characterized in that, Also includes: Based on the second target dataset from the multiple sub-data centers, the target processing order, and the overload situation, at least one of valve control instructions, power supply vehicle control instructions, and rescue vehicle control instructions is generated and sent to the emergency monitoring platform to control the intelligent gas valve to automatically open or close based on its open / closed state, control the mobile emergency power supply vehicle to travel based on its route and provide power based on its power supply capacity, and / or control the display terminal set on the rescue vehicle to display the rescue route based on the rescue arrival time limit.

7. The method as described in claim 6, characterized in that, During the course of the rescue vehicle's journey along the rescue route, the rescue vehicle is configured to identify risk areas based on road condition data and transmit the data back to the emergency monitoring and management platform.

8. The method as described in claim 6, characterized in that, The rescue vehicle control commands include the type and number of rescue vehicles; The step of generating at least one of valve control commands, power supply vehicle control commands, and rescue vehicle control commands based on the second target dataset from the multiple sub-data centers, the target processing order, and the overload situation includes: Based on the second target dataset from the multiple data centers, the target processing order, and the overload situation, a trend prediction model is used to determine the disaster development trend corresponding to the target processing order. The trend prediction model is a machine learning model. Based on the development trend of the disaster, the command to control the rescue vehicle is generated.

Images