Method, system and device for determining regional risk intensity of smart city and medium

Abstract

The invention discloses a smart city regional risk intensity determination method, system and device, and a storage medium. The method comprises the following steps: obtaining risk factor parameters of each device of a smart city at a current sampling moment; determining an equipment risk value of each piece of equipment at the current sampling moment according to the risk factor parameters; when the equipment risk value is greater than a preset threshold value, generating work order alarm data of each piece of equipment; generating a risk category of each device at the current sampling moment according to the work order alarm data; determining a risk coefficient of a target area according to the equipment risk value and the risk category; wherein the target area comprises a plurality of devices, and the risk coefficient is used for representing the risk intensity of the smart city area. The method can be widely applied to the technical field of artificial intelligence.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method, system, device and storage medium for determining regional risk intensity in a smart city. Background Art

[0002] With the rapid development and deep integration of cutting-edge technologies such as the Internet of Things (IoT), cloud computing, and artificial intelligence (AI), the interconnection of all things, intelligent management, and digital operations have gradually become key elements of the core strategies of governments and enterprises. In order to promote urban interconnection, the Ministry of Industry and Information Technology, together with ten other departments, has issued relevant policy documents, which clearly require the acceleration of the coordinated development of new information infrastructure and the promotion of the deep integration of information infrastructure and traditional infrastructure. This includes the intensive deployment of urban sensing terminals and the unified construction of city-level IoT sensing terminal management and data analysis platforms.

[0003] With the gradual implementation of IoT platforms for smart cities and smart parks, many problems that were not fully anticipated during the design phase have been exposed during the operation of the platforms, which has seriously affected the efficient operation of the business. At present, the IoT platforms for smart cities and smart parks usually consist of two parts: one is the front-end visualization system used to display the management area and the spatial layout of the building; the other is the back-end management platform used to manage the connection and protocol adaptation of IoT devices. The front-end visualization system usually uses two-dimensional plane maps or three-dimensional CIM technology to mark the location and status of the sensing device through longitude and latitude data, thereby achieving the digital twin effect and displaying the operating status of the device in real time. This system is usually used for management inspections by the leadership and daily inspections and troubleshooting by operators. The back-end management platform focuses on the connection and protocol adaptation of the device, is responsible for receiving device data, integrating with third-party platforms, and providing real-time and detailed sensing data support for the front-end.

[0004] However, with the increasing number of IoT sensing devices in smart cities and smart parks, the workload of manual equipment inspections on large visual screens has also increased significantly. There are currently two main ways to handle abnormal devices: one is to dispatch an order after discovering that the device is offline; the other is to rely on the platform's rule engine to trigger dispatch when the preset abnormal data rules are met. The above method has a lag and is difficult to achieve pre-emptive prevention. In addition, subsequent risk judgments mostly rely on the experience of operators, resulting in low accuracy in risk identification and difficulty in ensuring response time. Therefore, there are still technical problems that need to be solved in related technologies. Summary of the invention

[0005] The purpose of this application is to solve one of the technical problems existing in the prior art to at least a certain extent.

[0006] To this end, one purpose of an embodiment of the present application is to provide a method, system, device and storage medium for determining the risk intensity of a smart city area. This solution can reduce the transmission pressure of the server, improve the business carrying capacity of the server and improve the stability of the server.

[0007] In order to achieve the above-mentioned technical objectives, the technical solution adopted by the embodiments of the present application includes: a method for determining the risk intensity of a smart city area, comprising: obtaining the risk factor parameters of each device in the smart city at the current sampling moment; determining the device risk value of each device at the current sampling moment according to the risk factor parameters; when the device risk value is greater than a preset threshold, generating work order alarm data for each device; generating the risk category of each device at the current sampling moment according to the work order alarm data; determining the risk coefficient of a target area according to the device risk value and the risk category; wherein the target area includes a number of devices, and the risk coefficient is used to characterize the risk intensity of the smart city area.

[0008] The embodiment of the present application can obtain the risk factor parameters of each device at the current sampling moment of the smart city; determine the device risk value of each device at the current sampling moment according to the risk factor parameters; generate work order alarm data for each device when the device risk value is greater than a preset threshold; generate the risk category of each device at the current sampling moment according to the work order alarm data; determine the risk coefficient of the target area according to the device risk value and the risk category; wherein the target area includes several devices, and the risk coefficient is used to characterize the risk intensity of the smart city area. The embodiment of the present application calculates the device risk value through the risk factor parameters of each device at the current sampling moment of the smart city, and can generate the corresponding work order alarm data when the risk value is greater than the preset threshold, and then can generate the risk category of each device at the current sampling moment according to the work order alarm data, and finally determine the risk coefficient of the target area where several devices are located according to the device risk value and the risk category, which can characterize the risk intensity of the smart city area. The embodiment of the present application can quickly and timely calculate the risk coefficient of the area, improve the timeliness of the risk prediction of the smart city area, and improve the accuracy of the risk prediction.

[0009] In addition, the method for determining the risk intensity of a smart city area according to the above embodiment of the present invention may also have the following additional technical features:

[0010] Further, in the embodiment of the present application, determining the device risk value of each device at the current sampling moment according to the risk factor parameter specifically includes:

[0011] Determining the forget gate state at the current sampling moment according to the risk factor parameter;

[0012] Determine the cell state at the current sampling moment according to the forget gate state;

[0013] Determine the output gate state at the current sampling moment according to the risk factor parameter and the device risk value at the previous sampling moment;

[0014] The device risk value of each device at the current sampling moment is determined according to the unit state and the output gate state.

[0015] Further, in an embodiment of the present application, determining the output gate state at the current sampling moment according to the risk factor parameter and the device risk value at the previous sampling moment specifically includes:

[0016] The risk factor parameter and the device risk value at the previous sampling time are input into a third formula to determine the output gate state at the current sampling time; wherein the third formula is:

[0017] o t =σ(W o ·[h t-1 ,x(t)]+b o )

[0018] Among them, W o The weight matrix of the output gate, h t-1 is the risk value of the device at the last sampling moment, x(t) is the risk factor parameter, b o Represents the bias term of the output gate.

[0019] Further, in the embodiment of the present application, determining the forget gate state at the current sampling moment according to the risk factor parameter specifically includes:

[0020] The risk factor parameters are input into the first formula group to determine the forget gate state at the current sampling moment; wherein the first formula group is:

[0021] f t =σ(W f ·[h t-1 ,x(t)]+b f )

[0022]

[0023] f t Indicates the state of the forget gate at the current sampling moment. Where x(t) represents the input of the risk factor parameter at the current sampling moment, h t-1 represents the output of the unit at the previous sampling moment, b f Represents the bias term of the forget gate, W f represents the weight matrix of the forget gate, and σ represents the sigmoid neuron function.

[0024] Further, in the embodiment of the present application, determining the unit state at the current sampling moment according to the forget gate state specifically includes:

[0025] The forget gate state is input into the second formula group to determine the unit state at the current sampling time; wherein the second formula group is:

[0026]

[0027]

[0028] Among them, i t Indicates the input gate state, Indicates the update status of the backup information at the current sampling time, C t Indicates the unit state at the current sampling moment; W i represents the weight matrix of the input gate, W c The weight matrix representing the cell state, C t-1 represents the cell state at the last sampling moment, h t-1 represents the output of the unit at the previous sampling moment, f t Indicates the state of the forget gate at the current sampling moment, b i represents the bias term of the input gate, b c represents the bias term of the unit state, and x(t) represents the input of the risk factor parameter at the current sampling moment.

[0029] Further, in the embodiment of the present application, generating the risk category of each device at the current sampling time according to the work order alarm data specifically includes:

[0030] Based on the GBDT classification algorithm, the probability corresponding to each preset risk category is determined according to the work order alarm data;

[0031] The preset risk category with the highest probability is used as the risk category of each device at the current sampling moment.

[0032] Further, in an embodiment of the present application, determining the risk coefficient of the target area based on the device risk value and the risk category includes: determining the risk coefficient of the target area by learning the relationship between the device risk score and the overall risk of the region through a random forest model based on the device risk value and the risk category; or, determining the risk coefficient of the target area by learning the relationship between the device risk score and the overall risk of the region through a logistic regression model based on the device risk value and the risk category; or, determining the risk coefficient of the target area by learning the relationship between the device risk score and the overall risk of the region through an XGBoost model based on the device risk value and the risk category.

[0033] On the other hand, the embodiment of the present application also provides a smart city regional risk intensity determination system, including:

[0034] An acquisition unit, used to obtain the risk factor parameters of each device in the smart city at the current sampling moment;

[0035] A first processing unit, configured to determine a device risk value of each device at a current sampling moment according to the risk factor parameter;

[0036] A second processing unit, configured to generate work order alarm data for each device when the device risk value is greater than a preset threshold;

[0037] A third processing unit is used to generate a risk category of each device at a current sampling moment according to the work order alarm data;

[0038] The fourth processing unit is used to determine the risk coefficient of the target area according to the device risk value and the risk category; wherein the target area includes a plurality of devices, and the risk coefficient is used to characterize the risk intensity of the smart city area.

[0039] On the other hand, the present application also provides a device for determining regional risk intensity in a smart city, comprising:

[0040] at least one processor;

[0041] at least one memory for storing at least one program;

[0042] When the at least one program is executed by the at least one processor, the at least one processor implements a method for determining the risk intensity of a smart city area as described in any one of the invention contents.

[0043] In addition, the present application also provides a computer-readable storage medium, which stores processor-executable instructions, and the processor-executable instructions are used to execute a method for determining the risk intensity of a smart city area as described in any of the above items when executed by the processor.

[0044] The advantages and benefits of the present application will be partially given in the following description, and partially become apparent from the following description, or be understood through the practice of the present application:

[0045] This application can obtain the risk factor parameters of each device at the current sampling moment of the smart city; determine the device risk value of each device at the current sampling moment according to the risk factor parameters; generate work order alarm data for each device when the device risk value is greater than a preset threshold; generate the risk category of each device at the current sampling moment according to the work order alarm data; determine the risk coefficient of the target area according to the device risk value and the risk category; wherein the target area includes several devices, and the risk coefficient is used to characterize the risk intensity of the smart city area. This application calculates the device risk value through the risk factor parameters of each device at the current sampling moment of the smart city, and can generate the corresponding work order alarm data when the risk value is greater than the preset threshold, and then can generate the risk category of each device at the current sampling moment according to the work order alarm data, and finally determine the risk coefficient of the target area where several devices are located that can characterize the risk intensity of the smart city area according to the device risk value and the risk category. This application can quickly and timely calculate the risk coefficient of the area, improve the timeliness of risk prediction in the smart city area, and improve the accuracy of risk prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A schematic diagram of the steps of a method for determining regional risk intensity in a smart city in a specific embodiment of the present invention;

[0047] Figure 2 A schematic diagram of the steps of determining the device risk value of each device at the current sampling moment according to the risk factor parameters in a specific embodiment of the present invention;

[0048] Figure 3 A schematic diagram of the steps of generating the risk category of each device at the current sampling moment according to the work order alarm data in a specific embodiment of the present invention;

[0049] Figure 4 It is a module schematic diagram of a method for determining regional risk intensity of a smart city in another specific embodiment of the present invention;

[0050] Figure 5 It is a structural schematic diagram of a smart city regional risk intensity determination system in a specific embodiment of the present invention;

[0051] Figure 6 It is a structural schematic diagram of a device for determining regional risk intensity in a smart city in a specific embodiment of the present invention. DETAILED DESCRIPTION

[0052] The following describes in detail the embodiments of the present invention in conjunction with the accompanying drawings to illustrate the principles and processes of the smart city area risk intensity determination method, system, device and storage medium in the embodiments of the present invention.

[0053] With the rapid development and deep integration of cutting-edge technologies such as the Internet of Things (IoT), cloud computing, and artificial intelligence (AI), the interconnection of all things, intelligent management, and digital operations have gradually become key elements of the core strategies of governments and enterprises. In order to promote urban interconnection, the Ministry of Industry and Information Technology, together with ten other departments, has issued relevant policy documents, which clearly require the acceleration of the coordinated development of new information infrastructure and the promotion of the deep integration of information infrastructure and traditional infrastructure. This includes the intensive deployment of urban sensing terminals and the unified construction of city-level IoT sensing terminal management and data analysis platforms.

[0054] With the gradual implementation of IoT platforms for smart cities and smart parks, many problems that were not fully anticipated during the design phase have been exposed during the operation of the platforms, seriously affecting the efficient operation of the business. At present, the IoT platforms for smart cities and smart parks usually consist of two parts: one is the front-end visualization system used to display the management area and the spatial layout of the building; the other is the back-end management platform used to manage the connection and protocol adaptation of IoT devices. The front-end visualization system usually uses two-dimensional plane maps or three-dimensional CIM technology to mark the location and status of the sensing device through longitude and latitude data, thereby achieving the digital twin effect and displaying the operating status of the device in real time. This system is usually used for management inspections by the leadership and daily inspections and troubleshooting by operators. The back-end management platform focuses on the connection and protocol adaptation of the device, is responsible for receiving device data, integrating with third-party platforms, and providing real-time and detailed sensing data support for the front-end.

[0055] However, with the increasing number of IoT sensing devices in smart cities and smart parks, the workload of manual equipment inspections on large visual screens has also increased significantly. There are currently two main ways to handle abnormal devices: one is to dispatch an order after discovering that the device is offline; the other is to rely on the platform's rule engine to trigger dispatch when the preset abnormal data rules are met. The above method is less efficient when facing large-scale equipment, and there is a lag, making it difficult to achieve pre-emptive prevention. In addition, subsequent risk judgments rely more on the experience of operators, resulting in low accuracy in risk identification and difficulty in ensuring response time. Therefore, there are still problems in related technologies that need to be solved.

[0056] In view of the above-mentioned defects of the prior art, Figure 1 , Figure 1 Schematic diagram of the steps of a method for determining the risk intensity of a smart city area provided in an embodiment of the present application. Figure 1 In the method for determining the risk intensity of a smart city area, the method may include but is not limited to steps S101 to S105.

[0057] S101. Obtain risk factor parameters of each device in the smart city at the current sampling moment.

[0058] S102: Determine the device risk value of each device at the current sampling moment according to the risk factor parameters.

[0059] S103: When the equipment risk value is greater than a preset threshold, work order alarm data for each equipment is generated.

[0060] S104. Generate a risk category of each device at the current sampling time according to the work order alarm data.

[0061] S105. Determine the risk factor of the target area according to the equipment risk value and risk category.

[0062] It is understandable that the target area can be a specific area defined by the user. The target area can include several devices. The risk factor can be used to characterize the risk intensity of the smart city area. The risk factor parameters can be dimensional data such as device location data (latitude and longitude, jurisdiction...), perception data (numerical data such as temperature, humidity, air quality, etc.), status data (online time, offline time...), basic data (device type, communication protocol, manufacturer, installation time...). The above data can be updated regularly, and the above data collected at each sampling moment can form time series data respectively. Specifically, taking the temperature measured by the device as an example, the risk factor parameters can include the temperature collected at time t and the temperature collected at time t1, where t and t1 are two adjacent sampling moments. The preset threshold can be an arbitrary value, and the specific value can be adjusted according to user needs. The work order alarm data can include information such as the location and name of the device.

[0063] In some feasible embodiments of the present application, the processor can establish a wired or wireless connection with the database, and the processor can obtain the risk factor parameters of each device in the smart city at the current sampling moment from the database. After obtaining the risk factor parameters, the processor can input the risk factor parameters into the artificial intelligence model to determine the device risk value of each device at the current sampling moment. After obtaining the device risk value, when the device risk value is greater than the preset threshold, the processor can generate work order alarm data for each device. Based on the work order alarm data, the processor can generate a risk category for each device at the current sampling moment. Based on the device risk value and risk category, the processor can determine the risk coefficient of the target area where several devices exist.

[0064] It should be noted that the above-mentioned wired connection method may include a connection between a mobile device and a processing module, and may also include a connection between a processing module and a hardware device, as well as a wired connection between other devices currently known or to be developed in the future and the processing module; and the above-mentioned wireless connection method may include but is not limited to 3G / 4G / 5G connection, WiFi connection, Bluetooth connection, WiMAX connection, Zigbee connection, UWB (Ultra Wide Band) connection, and other wireless connection methods currently known or to be developed in the future.

[0065] In summary, the present application can obtain the risk factor parameters of each device at the current sampling moment of the smart city; determine the device risk value of each device at the current sampling moment according to the risk factor parameters; generate work order alarm data for each device when the device risk value is greater than a preset threshold; generate the risk category of each device at the current sampling moment according to the work order alarm data; determine the risk coefficient of the target area according to the device risk value and the risk category; wherein the target area includes several devices, and the risk coefficient is used to characterize the risk intensity of the smart city area. This embodiment calculates the device risk value through the risk factor parameters of each device at the current sampling moment of the smart city, and can generate the corresponding work order alarm data when the risk value is greater than the preset threshold, and then can generate the risk category of each device at the current sampling moment according to the work order alarm data, and finally determine the risk coefficient of the target area where several devices are located that can characterize the risk intensity of the smart city area according to the device risk value and the risk category. This embodiment can quickly and timely calculate the risk coefficient of the area, improve the timeliness of risk prediction in the smart city area, and improve the accuracy of risk prediction.

[0066] Further, refer to Figure 2 , Figure 2 Schematic diagram of the steps of determining the device risk value of each device at the current sampling time according to the risk factor parameters in the embodiment of the present application. Figure 2 In the embodiment, the step may specifically include steps S201 to S204.

[0067] S201. Determine the forget gate state at the current sampling moment according to the risk factor parameters.

[0068] S202: Determine the unit state at the current sampling moment according to the forget gate state.

[0069] S203: Determine the output gate state at the current sampling time according to the risk factor parameter and the device risk value at the previous sampling time.

[0070] S204: Determine the device risk value of each device at the current sampling moment according to the unit state and the output gate state.

[0071] In some feasible embodiments of the present application, the processor can determine the forget gate state at the current sampling moment according to the risk factor parameter. According to the forget gate state, the processor can determine the unit state at the current sampling moment. According to the risk factor parameter and the device risk value at the previous sampling moment, the processor can determine the output gate state at the current sampling moment. According to the unit state and the output gate state, the processor can determine the device risk value of each device at the current sampling moment.

[0072] Further, in the embodiment of the present application, the output gate state at the current sampling time is determined according to the risk factor parameter and the device risk value at the previous sampling time, specifically including:

[0073] The risk factor parameter and the device risk value at the previous sampling time are input into the third formula to determine the output gate state at the current sampling time; wherein the third formula is:

[0074] o t =σ(W o ·[h t-1 ,x(t)]+b o )

[0075] Among them, W o The weight matrix of the output gate, h t-1 is the equipment risk value at the last sampling moment, x(t) is the risk factor parameter, b o Represents the bias term of the output gate.

[0076] Furthermore, in the embodiment of the present application, the forget gate state at the current sampling moment is determined according to the risk factor parameter, specifically including:

[0077] The risk factor parameters are input into the first formula group to determine the forget gate state at the current sampling moment; wherein the first formula group is:

[0078] f t =σ(W f ·[h t-1 ,x(t)]+b f )

[0079]

[0080] f t Indicates the state of the forget gate at the current sampling moment. Where x(t) represents the input of the risk factor parameter at the current sampling moment, h t-1 represents the output of the unit at the previous sampling moment, b f Represents the bias term of the forget gate, W f represents the weight matrix of the forget gate, and σ represents the sigmoid neuron function.

[0081] Further, in the embodiment of the present application, the unit state at the current sampling moment is determined according to the forget gate state, specifically including:

[0082] The forget gate state is input into the second formula group to determine the unit state at the current sampling moment; wherein the second formula group is:

[0083]

[0084] Among them, i t Indicates the input gate state, Indicates the update status of the backup information at the current sampling time, C t Indicates the unit state at the current sampling moment; W i represents the weight matrix of the input gate, W c The weight matrix representing the cell state, C t-1 represents the cell state at the last sampling moment, h t-1 represents the output of the unit at the previous sampling moment, f t Indicates the state of the forget gate at the current sampling moment, b i represents the bias term of the input gate, b c represents the bias term of the unit state, and x(t) represents the input of the risk factor parameter at the current sampling moment.

[0085] Further, refer to Figure 3 , Figure 3 FIG. 1 is a schematic diagram of the steps for generating the risk category of each device at the current sampling time according to the work order alarm data in an embodiment of the present application. Figure 3 In the embodiment, this step may specifically include step S301-step S302.

[0086] S301. Based on the GBDT classification algorithm and work order alarm data, determine the probability corresponding to each preset risk category.

[0087] S302: Taking the preset risk category with the highest probability as the risk category of each device at the current sampling moment.

[0088] Furthermore, in the embodiment of the present application, the step of determining the risk coefficient of the target area according to the equipment risk value and the risk category may include step S401, step S402 or step S403.

[0089] S401. According to the equipment risk value and risk category, the relationship between the equipment risk score and the overall regional risk is learned through a random forest model to determine the risk coefficient of the target area.

[0090] S402: According to the equipment risk value and risk category, the relationship between the equipment risk score and the overall regional risk is learned through a logistic regression model to determine the risk coefficient of the target area.

[0091] S403: According to the equipment risk value and risk category, the relationship between the equipment risk score and the overall regional risk is learned through the XGBoost model to determine the risk coefficient of the target area.

[0092] The specific implementation principle of this application is described below with reference to the accompanying drawings:

[0093] In some embodiments of the present application, the method for determining the risk intensity of a smart city area is implemented through steps S1 to S6.

[0094] S1. Perception data construction: Based on the data reported by the IoT device connection management platform in the smart city and smart park IoT platforms, the object model constructed by the device is processed to form a standardized and complete perception database. The database covers the real-time status, historical records and related operating parameters of various IoT devices.

[0095] S2. Automated dispatch and rule optimization: An automated interaction module is designed for various operational scenarios that exist during the long-term operation of equipment, especially a new dispatch work order function. By optimizing the rule engine, the device offline duration detection function is set. The rule engine matches rules based on newly added data fields such as offline duration. After a successful match, the system will automatically generate a dispatch work order. Operation personnel can use the dispatch work order module to conduct key verification, and after secondary confirmation, transfer the work order to the management unit of the corresponding equipment for processing.

[0096] S3. Equipment risk prediction: Based on the perception database in the platform, the long short-term memory network (LSTM) algorithm is used to predict the risk of the equipment. The algorithm can dynamically analyze the historical operation data and current status of the equipment to determine whether the equipment has risks at the current moment and its corresponding risk level.

[0097] S4. Risk category identification: If a potential risk is detected in the device, the system will start the device risk type identification algorithm. Through the gradient boosting decision tree (GBDT) algorithm, the risk category of the device is analyzed based on the perception data, so as to accurately determine the specific type of device failure or abnormality.

[0098] S5. Regional risk optimization and inspection scheduling: The front-end system dynamically updates the risk factor and category of the equipment according to the set cycle. On the large visualization screen, the system will give priority to displaying equipment with higher risk factors. In order to optimize the inspection efficiency, when the number of equipment increases significantly, through algorithms such as random forest, logistic regression or XGBoos t, the relationship between the risk score of the equipment and the overall risk of the region is learned, the risk factor of the region is automatically generated, and the regions are sorted. The inspection frequency of high-risk areas will be increased, and the number of inspections in low-risk areas will be appropriately reduced to ensure the rational allocation of inspection resources, thereby effectively reducing the inspection workload of operators.

[0099] S6. Return order processing and quality evaluation: After the work order is issued, the system will transfer the work order to the account of the person responsible for the maintenance of the equipment. After the maintenance personnel complete the on-site inspection and repair work, the equipment should return to normal and start uploading data again. At this time, the system will immediately input the new equipment data into the equipment risk prediction module for real-time risk assessment. After secondary confirmation by the risk prediction model and the operator, if the equipment no longer has abnormalities, the system will release the alarm status and mark the work order as completed, thus realizing closed-loop management of the work order process. In addition, the system will automatically evaluate the quality of work order processing, and conduct a comprehensive evaluation of the return order quality based on factors such as the speed at which the equipment returns to normal, the accuracy of risk elimination, and the subsequent operating status, to provide data support for subsequent operation management.

[0100] This application can effectively address the challenges of increasing equipment quantity and management complexity, improve operational efficiency, reduce the lag in risk identification, and significantly reduce inspection workload.

[0101] In some embodiments of the present application, reference Figure 4 ,The smart city regional risk intensity determination system mainly includes data ,processing module, automated operation module, equipment risk database module, equipment risk ,dynamic prediction module, equipment risk category identification module, and ,operation inspection module. The relationship between the modules and the ,detailed technical scheme of the application scenario includes the first step to the

[0102] Step 1: Build a data processing module. The background device connection management platform of the traditional Internet of Things platform is equipped with a physical model of the device. Through the physical model and data cleaning technology, a perception database can be formed. The database stores detailed data of each device in a structured manner. These data are used to support the subsequent work of the front-end visualization screen and the operation management center.

[0103] Step 2: Build an equipment risk database. First, based on the previous data of the IoT platform, a large amount of work order data of equipment alarms is left in the database. This part of the data can be used as a training set for the subsequent two models. At the same time, as the operation process progresses, new work order data will be used as a new training set as a supplement to optimize the model effect. First, construct several feature factor sets for equipment risk prediction, including location feature factors, perception feature factors, state feature factors, and basic feature factors. Then, build an equipment risk database based on the feature factor set, covering equipment location data (latitude and longitude, jurisdiction...), perception data (numerical data such as temperature, humidity, air quality, etc.), status data (online time, offline time...), basic data (equipment type, communication protocol, manufacturer, installation time...) and other dimensional data, and establish time series data based on the time dimension information of the data regularly updated.

[0104] Step 3: Build a risk dynamic prediction model based on the equipment risk database and LSTM algorithm:

[0105] Model training data: The LSTM model is trained based on the completed equipment work order alarm data in the equipment risk database. The sequence of equipment data input at time t is:

[0106]

[0107] in It represents the risk factor parameter of device p at time t, i = 1, 2, ..., 9; p = 1, 2, ..., P. Here, we take the temperature sensor device data as an example. The risk factor parameters include device location data (latitude and longitude, jurisdiction), perception data (temperature numerical data), status data (online time, offline time), and basic data (device type, communication protocol, manufacturer, installation time). Therefore, i = 9, the number of devices is P, and the work order alarm data is updated in real time with the alarm dynamics generated by the platform operation.

[0108] Apply LSTM algorithm to X(t) for model training. The working principle of LSTM algorithm mainly consists of the following update process:

[0109] Forget gate: f t Represents the state of the forget gate at time t, controls the degree of forgetting of the memory at the previous moment, and determines which information should be abandoned or retained.

[0110] f t =σ(W f ·[h t-1 ,x(t)]+b f )

[0111] Where x(t) represents the input of the unit at time t, h t-1represents the output of the unit at time t-1, b f Represents the bias term of the forget gate, W f represents the weight matrix of the forget gate, σ represents the sigmoid neuron function, and the function sigmoid is defined as

[0112]

[0113] Input gate: i t Represents the input gate, which determines the information update status required by the unit at the current moment. Indicates the update status of backup information, C t Indicates the unit status at the current moment, controls the updating of important information and the discarding of useless information.

[0114]

[0115] The tanh function is defined as

[0116]

[0117] Output gate: o t Represents the state of the output gate, which is used to determine the value of the next hidden state and controls the influence of long-term memory on the current output.

[0118] h t =o t ×tanh(C t )

[0119] o t =σ(W o ·[h t-1 ,x(t)]+b o )

[0120] The risk prediction of a device at time t can be input into the sequence data of device p at time t. The sequence data is as follows:

[0121]

[0122] X p (t)′ is input into the LSTM algorithm for risk prediction, and the device risk value h′ is output in real time t .

[0123] Whether the equipment currently has risks: Based on the completed equipment work order alarm data in the equipment risk database, determine the average risk value of the equipment with alarm work orders at time t As the risk threshold, , it is considered that the device p is at risk at the current time point.

[0124] The fourth step is the automatic identification module of equipment risk categories. In this embodiment, the gradient boosting decision tree (GBDT) algorithm is used to train the model based on the completed equipment work order alarm data in the equipment risk database. The algorithm can flexibly process various types of data, with high prediction accuracy and good generalization performance.

[0125] (1) Initialize the first weak learner F0(x):

[0126]

[0127] Among them, P(Y=1|x) is the proportion of y=1 in the training samples, and the prior information is used to initialize the learner.

[0128] (2) Establish M classification and regression trees m = 1, 2, ..., M:

[0129] For i=1,2,…,N, calculate the response value corresponding to the mth tree (the negative gradient of the loss function, i.e., the pseudo residual):

[0130]

[0131] For i = 1, 2, ..., N, use CART regression tree to fit the data (x i ,r m,i ), we get the mth regression tree, whose corresponding leaf node area is R m,j , where j = 1, 2, ..., J m , and J m is the number of child nodes of the th regression tree.

[0132] For J m leaf node regions j=1,2,…,J m , calculate the best fit value:

[0133]

[0134] Update the strong learner F m (x):

[0135]

[0136] The final strong learner F M (x) can be expressed as:

[0137]

[0138] The GBDT classification algorithm uses a series of gradient boosting trees to fit the logarithmic probability, and finally outputs the probability of comparing categories. If the probability is large, it is predicted to be that category. Its model can be expressed as:

[0139]

[0140] When it is predicted that there is a risk in the in-transit project p, the data of the in-transit project p is input into the trained gradient boosting decision tree algorithm for automatic identification of risk categories. The corresponding risk categories are location risk, perception indicator risk, equipment status risk, and equipment attribute risk.

[0141] The fifth step is the operation inspection module. After obtaining the risk value and risk category of each device at time t in the previous four steps, machine learning models such as random forest, logistic regression, or XGBoost can be used to learn the relationship between the device risk score and the overall risk of the region, and finally automatically generate a risk coefficient for the region.

[0142] First, we need to build a dataset suitable for machine learning models, which includes the device information characteristics of each region, such as the risk score of the device (such as average, maximum, minimum, variance, etc.), the number of devices, categories, importance, historical risk events, time-related characteristics (such as the change trend of the device risk score, the operating time of the device), and regional characteristics such as the overall number of devices in the region, device density, historical risk records of the region (whether there have been serious accidents), and geographical location characteristics of the region (such as network environment, security strategy, etc.). The predicted value is the risk level of the region. 0 represents a low-risk area, 1 represents a medium-risk area, and 2 represents a high-risk area. The training dataset is shown in Table 1.

[0143]

[0144] Table 1

[0145] Then, you can get the actual risk value based on the regional equipment, and use random forest, logistic regression, XGBoost / LightGBM and other algorithms to calculate the regional risk K. If you use a classification model (such as a random forest classifier), the output will be the risk level of the region (for example: 0, 1, 2), and you can judge the risk of the region based on the classification result. If you use a regression model (such as XGBoost regression), the output will be a continuous risk score, which can be sorted by score to get the risk ranking of the region.

[0146] The risk K of each area is ranked, and the number of inspections is increased for the top 25% of areas, and the number of inspections can be appropriately reduced for the bottom 25% of areas. The inspection strategy is dynamically adjusted by the regional risk coefficient, which can effectively reduce the inspection workload and make it easier to detect equipment alarms and other situations in high-risk areas. At the same time, equipment with higher risk values ​​will be highlighted on the visualization screen, and the equipment predicted risk information will be displayed in the equipment details, which is convenient for platform operators to conduct inspections on the visualization screen.

[0147] Step 6. Processing of return orders and quality evaluation: After the work order is issued, the system will transfer the work order to the account of the person responsible for the maintenance of the equipment. After the maintenance personnel complete the on-site inspection and repair work, the equipment should return to normal and start uploading data again. At this time, the system will immediately input the new equipment data into the equipment risk prediction module for real-time risk assessment. After secondary confirmation by the risk prediction model and the operator, if the equipment no longer has abnormalities, the system will remove the alarm status and mark the work order as completed, thus realizing closed-loop management of the work order process. In addition, the system will automatically evaluate the quality of work order processing, and conduct a comprehensive evaluation of the return order quality based on factors such as the speed at which the equipment returns to normal, the accuracy of risk elimination, and the subsequent operating status, to provide data support for subsequent operation management.

[0148] In summary, the method for determining the regional risk intensity of a smart city in this application has the following advantages:

[0149] 1. This application can reduce human intervention and improve the overall operational efficiency of the IoT platform by optimizing automated dispatching, equipment risk prediction, and regional inspection scheduling. It can significantly reduce the workload of operators, especially in the context of increasing device numbers and management complexity.

[0150] 2. By introducing advanced algorithms such as LSTM and GBDT, this application can achieve accurate prediction of equipment risks and accurate identification of risk categories, reduce the lag in risk identification, and improve the timeliness and effectiveness of operations management.

[0151] 3. This application can avoid the explosion of inspection workload caused by the increase in the number of equipment and reduce labor costs by intelligently evaluating and ranking regional risk factors, dynamically adjusting inspection frequency, and rationally allocating inspection resources.

[0152] 4. This application builds a closed-loop management of the entire process from work order generation, processing to receipt and quality evaluation, and evaluates the quality of work order processing through feedback from equipment operation status, which can further improve the accuracy and efficiency of fault handling and provide data support for subsequent operation management.

[0153] 5. The visual operation system of the Internet of Things platform of this application can be fully optimized, which can effectively solve the management challenges brought about by the increase in the number and complexity of devices, and can realize an automated, efficient and intelligent inspection and management system.

[0154] In addition, refer to Figure 5 ,and Figure 1Corresponding to the method, a smart city regional risk intensity determination system is also provided in an embodiment of the present application. The system may include: an acquisition unit 1001, a first processing unit 1002, a second processing unit 1003, a third processing unit 1004 and a fourth processing unit 1005. The acquisition unit 1001 can be used to obtain the risk factor parameters of each device at the current sampling moment of the smart city. The first processing unit 1002 can be used to determine the device risk value of each device at the current sampling moment according to the risk factor parameters; the second processing unit 1003 can be used to generate work order alarm data for each device when the device risk value is greater than a preset threshold. The third processing unit 1004 can be used to generate the risk category of each device at the current sampling moment according to the work order alarm data. The fourth processing unit 1005 can be used to determine the risk coefficient of the target area according to the device risk value and the risk category; wherein the target area includes a number of devices, and the risk coefficient is used to characterize the risk intensity of the smart city area.

[0155] It should be noted that the acquisition unit may be any integrated circuit unit or microprocessor unit obtained by integrating a chip having a processing function and its peripheral circuits through existing integration technology. The first processing unit and the second processing unit may also be any integrated circuit module or microprocessor module obtained by integrating a chip having a processing function and its peripheral circuits through existing integration technology. The first processing unit and the second processing unit may also include one or more memories.

[0156] It should be noted that the contents of the above-mentioned smart city regional risk intensity determination method embodiment are all applicable to the smart city regional risk intensity determination system embodiment, and the functions specifically implemented by the smart city regional risk intensity determination system embodiment are the same as those of the above-mentioned smart city regional risk intensity determination method embodiment, and the beneficial effects achieved are also the same as the beneficial effects achieved by the above-mentioned smart city regional risk intensity determination method embodiment.

[0157] and Figure 1 Corresponding to the method, the embodiment of the present application also provides a smart city regional risk intensity determination device, the specific structure of which can be referred to Figure 6 ,include:

[0158] at least one processor 1011;

[0159] At least one memory 1012, used to store at least one program;

[0160] When the at least one program is executed by the at least one processor, the at least one processor implements the method for determining the risk intensity of the smart city area.

[0161] The contents of the above method embodiments are all applicable to the present device embodiments. The functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0162] and Figure 1 Corresponding to the method, an embodiment of the present application also provides a computer-readable storage medium, which stores processor-executable instructions, and the processor-executable instructions are used to execute the method for determining the risk intensity of a smart city area when executed by the processor.

[0163] The contents of the above-mentioned smart city regional risk intensity determination method embodiment are all applicable to the present storage medium embodiment. The functions specifically implemented by the present storage medium embodiment are the same as those of the above-mentioned smart city regional risk intensity determination method embodiment, and the beneficial effects achieved are also the same as those achieved by the above-mentioned smart city regional risk intensity determination method embodiment.

[0164] In some optional embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the application is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Optional embodiments are expected, wherein the order of various operations is changed and the sub-operation described as a part of a larger operation is performed independently.

[0165] In addition, although the present application is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features can be integrated into a single physical device and / or software module, or one or more functions and / or features can be implemented in a separate physical device or software module. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present application. More specifically, in view of the properties, functions, and internal relationships of the various functional modules in the device disclosed herein, the actual implementation of the module will be understood within the conventional techniques of the engineer. Therefore, those skilled in the art can implement the present application set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present application, which is determined by the full scope of the attached claims and their equivalents.

[0166] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several programs to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0167] The logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable programs for implementing the logical functions, and may be embodied in any computer-readable medium for use by a program execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch and execute a program from a program execution system, device or apparatus), or in conjunction with such program execution systems, devices or apparatuses. For purposes of this specification, a "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by a program execution system, device or apparatus, or in conjunction with such program execution systems, devices or apparatuses.

[0168] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0169] It should be understood that the various parts of the present application can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable program execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0170] In the above description of this specification, the description with reference to the terms "one embodiment / example", "another embodiment / example" or "certain embodiments / examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0171] Although the embodiments of the present application have been shown and described, those skilled in the art will appreciate that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present application, and that the scope of the present application is defined by the claims and their equivalents.

[0172] The above is a specific description of the preferred implementation of the present application, but the present application is not limited to the described embodiments. Technical personnel familiar with the field may make various equivalent modifications or substitutions without violating the spirit of the present application. These equivalent modifications or substitutions are all included in the scope defined by the claims of the present application.

Claims

1. A method for determining the regional risk intensity of a smart city, characterized in that: The following steps are involved: Obtain the risk factor parameters of each device in the smart city at the current sampling moment; Determine the device risk value of each device at the current sampling moment according to the risk factor parameters; When the equipment risk value is greater than a preset threshold, work order alarm data for each equipment is generated; Generate a risk category of each device at the current sampling time according to the work order alarm data; Determining a risk factor of a target area according to the equipment risk value and the risk category; The target area includes several devices, and the risk coefficient is used to characterize the risk intensity of the smart city area.

2. A method for determining regional risk intensity in a smart city according to claim 1, characterized in that: Determining the device risk value of each device at the current sampling moment according to the risk factor parameter specifically includes: Determining the forget gate state at the current sampling moment according to the risk factor parameter; Determine the cell state at the current sampling moment according to the forget gate state; Determine the output gate state at the current sampling moment according to the risk factor parameter and the device risk value at the previous sampling moment; The device risk value of each device at the current sampling moment is determined according to the unit state and the output gate state.

3. A method for determining regional risk intensity in a smart city according to claim 2, characterized in that: Determining the output gate state at the current sampling moment according to the risk factor parameter and the device risk value at the previous sampling moment specifically includes: The risk factor parameter and the device risk value at the previous sampling time are input into a third formula to determine the output gate state at the current sampling time; wherein the third formula is: o t =σ(W o ·[h t-1 ,x(t)]+b o ) Among them, W o The weight matrix of the output gate, h t-1 is the risk value of the device at the last sampling moment, x(t) is the risk factor parameter, b o Represents the bias term of the output gate.

4. According to claim 2, a method for determining the regional risk intensity of a smart city is characterized in that: Determining the forget gate state at the current sampling moment according to the risk factor parameter specifically includes: The risk factor parameters are input into the first formula group to determine the forget gate state at the current sampling moment; wherein the first formula group is: f t =σ(W f ·[h t-1 ,x(t)]+b f ) f t Indicates the state of the forget gate at the current sampling moment. Where x(t) represents the input of the risk factor parameter at the current sampling moment, h t-1 represents the output of the unit at the previous sampling moment, b f Represents the bias term of the forget gate, W f represents the weight matrix of the forget gate, and σ represents the sigmoid neuron function.

5. According to claim 2, a method for determining the risk intensity of a smart city region is characterized in that: Determining the unit state at the current sampling moment according to the forget gate state specifically includes: The forget gate state is input into the second formula group to determine the unit state at the current sampling time; wherein the second formula group is: Among them, i t Indicates the input gate state, Indicates the update status of the backup information at the current sampling time, C t Indicates the unit state at the current sampling moment; W i represents the weight matrix of the input gate, W c The weight matrix representing the cell state, C t-1 represents the cell state at the last sampling moment, h t-1 represents the output of the unit at the previous sampling moment, f t Indicates the state of the forget gate at the current sampling moment, b i represents the bias term of the input gate, b c represents the bias term of the unit state, and x(t) represents the input of the risk factor parameter at the current sampling moment.

6. A method for determining regional risk intensity in a smart city according to claim 1, characterized in that: Generating the risk category of each device at the current sampling time according to the work order alarm data specifically includes: Based on the GBDT classification algorithm, the probability corresponding to each preset risk category is determined according to the work order alarm data; The preset risk category with the highest probability is used as the risk category of each device at the current sampling moment.

7. A method for determining regional risk intensity in a smart city according to claim 1, characterized in that: The step of determining the risk coefficient of the target area according to the equipment risk value and the risk category includes: According to the equipment risk value and the risk category, the relationship between the equipment risk score and the overall risk of the region is learned through a random forest model to determine the risk coefficient of the target region; Alternatively, according to the equipment risk value and the risk category, a relationship between the equipment risk score and the overall risk of the region is learned through a logistic regression model to determine the risk coefficient of the target region; Alternatively, according to the equipment risk value and the risk category, the relationship between the equipment risk score and the overall risk of the region is learned through an XGBoost model to determine the risk coefficient of the target region.

8. A smart city regional risk intensity determination system, characterized in that: include: An acquisition unit, used to obtain the risk factor parameters of each device in the smart city at the current sampling moment; A first processing unit, configured to determine a device risk value of each device at a current sampling moment according to the risk factor parameter; A second processing unit, configured to generate work order alarm data for each device when the device risk value is greater than a preset threshold; A third processing unit is used to generate a risk category of each device at a current sampling moment according to the work order alarm data; A fourth processing unit, configured to determine a risk coefficient of a target area according to the equipment risk value and the risk category; The target area includes several devices, and the risk coefficient is used to characterize the risk intensity of the smart city area.

9. A device for determining regional risk intensity in a smart city, characterized in that include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements a method for determining the risk intensity of a smart city area as described in any one of claims 1-7.

10. A computer-readable storage medium storing instructions executable by a processor, characterized in that: The processor executable instructions are used to execute a method for determining the risk intensity of a smart city area as described in any one of claims 1-7 when executed by the processor.

Images