A Cloud Platform and Method for Closed-Loop Management and Control of IoT Devices Based on Dynamic Security Index

Abstract

This invention discloses a cloud platform and method for closed-loop management and control of IoT devices based on a dynamic security index, relating to the field of IoT technology. It includes a database server, an application server, a monitoring module for real-time acquisition and synchronization of device operating status data, an alarm module for determining sensor operating status, an early warning module for providing early warnings for future time points, a processing module for calculating the dynamic security index of the cloud platform, and a feedback module for bidirectional transmission. The monitoring, alarm, early warning, processing, and feedback modules are all deployed in the application server, and the operating data, alarm information, early warning information, and processing results generated by each module are stored in the database server. By introducing a dynamic security index, real-time adjustments can be made to the dynamic changes in potential device faults, avoiding response lag. Through a complete closed-loop management and control mechanism, processing efficiency is improved, forming an effective device management and control system.

Description

Technical Field

[0001] This invention relates to the field of Internet of Things (IoT) technology, and in particular to a cloud platform and method for closed-loop management and control of IoT devices based on a dynamic security index. Background Technology

[0002] The Internet of Things (IoT) technology seamlessly connects sensors, controllers, and production equipment, forming a digital network that enables information-based production processes and intelligent remote control. Currently, the security protection of many IoT devices relies primarily on static security policies. These policies are set during device deployment and are difficult to adjust in real time according to dynamic changes in the device's operating environment and potential threats, resulting in limited control effectiveness. IoT devices are typically manufactured by different companies, lacking a unified security management platform for centralized control. This makes it difficult for managers to fully grasp the security status of the devices and promptly detect and respond to security incidents. Existing IoT security assessment methods are mostly based on static indicators, which fail to accurately reflect the actual security status of the devices during operation. Furthermore, the lack of effective security early warning mechanisms prevents the timely detection of potential threats before anomalies occur, leading to delayed security incident response. Current security control solutions often focus on single aspects, lacking a complete closed-loop control mechanism, resulting in low efficiency in handling security incidents and hindering the formation of an effective security control system.

[0003] Therefore, there is an urgent need for a cloud platform and method for closed-loop management of IoT devices based on dynamic security index, in order to overcome the limitations of existing technologies. Summary of the Invention

[0004] The purpose of this invention is to provide a cloud platform and method for closed-loop management and control of IoT devices based on dynamic security index, so as to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0006] The cloud platform for closed-loop management and control of IoT devices based on dynamic security index includes a database server, an application server, and a monitoring module, an alarm module, an early warning module, a processing module, and a feedback module deployed in the application server. The monitoring module, alarm module, early warning module, processing module, and feedback module are respectively used to generate operating data, alarm records, early warning information, and processing results, and all of the above information is stored in the database server.

[0007] The monitoring module is used to collect and synchronize equipment operating status data in real time;

[0008] The alarm module is used to determine the working status of the sensor and to issue multi-channel alarms according to the preset push mechanism.

[0009] The early warning module is used to provide early warnings for future time points and generate early warning information;

[0010] The processing module is used to calculate the dynamic security index of the cloud platform;

[0011] The feedback module is used to transmit information between the user and the system in both directions, and supports multimodal feedback input and command response.

[0012] As a preferred technical solution, the monitoring module includes a data acquisition unit and a transmission unit:

[0013] The acquisition unit collects real-time device operation status data of IoT devices through a multi-source sensor network deployed on IoT terminals.

[0014] The transmission unit uses a secure encrypted transmission protocol to synchronously transmit the collected device operation data and monitoring data from multiple sensors to the database server for storage, and sends them to the alarm module in real time for working status analysis.

[0015] It should be explained that IoT devices, such as headquarters special gas cabinets, contain multiple components. These components are composed of multi-source sensors, including leak sensors, high-pressure transmitters, temperature sensors, flame detectors, mass flow meters, and so on. The device operation data of IoT devices includes information from the multi-source sensors, information on vulnerable parts that need to be replaced regularly, information on the gas used by the device, information on the device's panel diagram, information on the device's operating instructions, and the device's working status.

[0016] As a preferred technical solution, the alarm module includes a judgment unit, a level classification unit, and an alarm unit;

[0017] The judgment unit determines the working state of the sensor based on a preset dynamic threshold judgment mechanism.

[0018] When the sensor's monitoring data is within the alarm threshold interval, the sensor is determined to be in normal working condition. The alarm threshold interval includes an upper limit value and a lower limit value.

[0019] When the sensor's monitoring data is outside the alarm threshold interval, the sensor is determined to be in an abnormal working state.

[0020] The level classification unit extracts monitoring data from sensors in abnormal working states and classifies alarm levels according to a preset sensor level classification mechanism.

[0021] The alarm unit generates alarm information according to a preset push mechanism and performs multi-channel alarms.

[0022] It should be explained that the alarm threshold interval is set by the administrator for the sensors that require alarms. The setting process includes: querying safety index information, selecting the sensors that require alarms and modifying the alarm threshold. During operation, when the operating data of the sensor requiring alarms is higher than the upper limit or lower than the lower limit, an alarm will be triggered, indicating that the sensor is in an abnormal working state. The preset sensor level classification mechanism is set by the administrator, with three alarm levels, from the lowest level to the highest level three, each corresponding to a different alarm contact person. The preset push mechanism is set by the administrator, including alarm contact information, push method, and notification method. The alarm contact information includes the alarm operator's ID, alarm operator's name, alarm level, mobile phone number, app, and email address. The push method includes a first-time push and multiple pushes, where multiple pushes include push intervals. The notification method includes SMS notification (sending an SMS to the contact's mobile phone when an alarm occurs), email notification (sending an email to the contact's email address when an alarm occurs), APP notification, and voice notification.

[0023] As a preferred technical solution, the early warning module includes an extraction unit and a prediction unit:

[0024] The extraction unit is used to extract the working logs and real-time monitoring data of the multi-source sensors from the database server. The working logs include historical monitoring data and early warning types of the multi-source sensors. The early warning types include normal, high threshold, and fault warnings.

[0025] The prediction unit establishes a feature dataset based on historical monitoring data, constructs a prediction model based on an LSTM neural network, and outputs the warning probability of different warning types within future time nodes after inputting time-series feature data. Based on the warning probability, it generates a descending sequence of warning types and takes the first warning type in the sequence as the warning type for the future time node.

[0026] When the warning type for the future time node is not normal, a warning message is generated based on the warning type.

[0027] As a preferred technical solution, the processing module includes a statistical unit and a security index unit:

[0028] The statistical unit acquires and processes the device operating status data of all IoT devices through the monitoring module. The operating status includes normal operation and abnormal operation. The statistical unit is used to count the number of IoT devices that are currently operating normally and abnormally.

[0029] The statistics unit is also used to extract alarm information from the alarm module and count the number of alarms currently pending and already processed.

[0030] The statistical unit further extracts the early warning information at the current time point and counts the number of early warnings that are currently pending and have been processed.

[0031] The security index unit, based on the aforementioned statistical data, is used to calculate the dynamic security index of the cloud platform in real time. The calculation formula is as follows:

[0032]

[0033] Where S represents the dynamic security index of the closed-loop management cloud platform, M represents the total number of sensors, and w m V represents the dynamic weight of the m-th sensor. m Let N represent the standardized risk value of the m-th sensor, w represent the dynamic weight of the IoT device, and N represent the dynamic risk value of the m-th sensor. Normal N Unusual These represent the number of IoT devices currently operating normally and those operating abnormally, respectively. This represents the cumulative risk value;

[0034] When the dynamic safety index is less than the safety threshold, the system will automatically send a maintenance signal to the management personnel.

[0035] As a preferred technical solution, the risk accumulation value is used to reflect the accumulated risk of not processing alarm and warning information in a timely manner;

[0036] Based on the number of alarms pending and the number of alarms already processed, the processing index of the current alarm information is calculated using the following formula:

[0037]

[0038] Where H represents the processing index of the current alarm information, and I represents the alarm level. N represents the risk weight of an alarm message with alarm level i. i M represents the number of alarms of level i that are currently pending processing. i This indicates the number of alarms of level i that have been processed so far;

[0039] Based on the number of alerts pending and the number of alerts already processed, calculate the risk index of the current alert information:

[0040]

[0041] Where P represents the risk index of the current warning information, γ1 and γ2 represent the risk weights of high-threshold warnings and fault warnings, respectively, and N pre1 N pre2 N represents the number of high-threshold warnings and the number of fault warnings among the processed warnings, respectively. finish Indicates the number of warnings processed, and γ3 represents the error;

[0042] Based on the processing index of the current alarm information and the risk index of the current warning information, the cumulative risk value is calculated using the following formula:

[0043]

[0044] Wherein, ρ1 and ρ2 represent the risk coefficients of alarm information and early warning information, respectively.

[0045] As a preferred technical solution, the feedback module includes an interaction unit and a visualization unit;

[0046] The interaction unit is used to transmit information between the user and the system in both directions, and supports multimodal feedback input and command response;

[0047] The visualization unit automatically generates visualization charts from the data generated by the monitoring module, alarm module, early warning module, processing module, and interaction unit based on the visualization template.

[0048] It should be explained that the visualization charts include system safety status, gas usage, and equipment operation and alarm status. System safety status includes a safety index, reflecting the overall safety status of the system, derived from equipment status, alarm information, and early warning information. Equipment status displays the real-time operating status of all equipment in the system. Alarm information displays the processing status of all alarm information in the system. Early warning information displays the processing status of all early warning information in the system. Gas usage includes liquid gas usage and gaseous gas usage, showing the monthly usage of each liquid gas and gaseous gas. Equipment operation and alarm status includes equipment operation status, equipment alarm status, and laboratory alarm records, enabling the classification and statistical analysis of equipment quantity and operating status, the classification and statistical analysis of alarm counts for each equipment, and the display of alarm counts at three levels for each laboratory. The horizontal axis represents the laboratory address, and the vertical axis represents the number of alarms. Through the visualization unit settings, spatiotemporal data is integrated to help users quickly locate the root cause of risks. Through the interactive unit settings, a closed loop of "machine decision-making - manual confirmation - model optimization" is achieved, reducing the risk of misoperation.

[0049] As a preferred technical solution, the minimum environmental standard for the database server includes a quad-core processor, 8GB of I / O-optimized memory, a 50GB hard disk, and a standard network connection with 5Mbps bandwidth; the minimum environmental standard for the application server environment includes an eight-core processor, 32GB of I / O-optimized memory, a 100GB hard disk, and a standard network connection with 10Mbps bandwidth.

[0050] A closed-loop management and control method for IoT devices based on a dynamic security index, the method comprising the following steps:

[0051] Step S100: Collect the device operation status data of IoT devices in real time through the multi-source sensor network of IoT terminals, and record the monitoring data of multi-source sensors and the device operation status data of IoT devices in the database simultaneously;

[0052] Step S200: Based on the monitoring data from the multi-source sensors and the preset dynamic threshold determination mechanism, determine the working status of each sensor, including normal working status and abnormal working status.

[0053] When the sensor's monitoring data is within the alarm threshold interval, the sensor is determined to be in normal working condition. The alarm threshold interval includes an upper limit value and a lower limit value. When the sensor's monitoring data is not within the alarm threshold interval, the sensor is determined to be in abnormal working condition.

[0054] Extract monitoring data from sensors in abnormal working state, classify alarm levels according to a preset sensor level classification mechanism, generate alarm information according to a preset push mechanism, and perform multi-channel alarms, including SMS notification, email notification, APP notification and voice notification.

[0055] Step S300: Extract the working logs and real-time monitoring data of the multi-source sensors from the database server. The working logs include historical monitoring data and early warning types of the multi-source sensors. The early warning types include normal, high threshold early warning and fault early warning.

[0056] A feature dataset is established based on historical monitoring data. A prediction model is constructed based on an LSTM neural network. After inputting time-series feature data, the warning probability of different warning types within future time nodes is output. The warning types are generated in descending order according to the warning probability. The first warning type in the sequence is taken as the warning type for the future time node.

[0057] When the warning type at the future time node is not normal, a warning message is generated based on the warning type. The warning message includes the warning time node, warning type, sensor name, and sensor data.

[0058] Step S400: Based on the device operation status data of all IoT devices at the current time node, the operation status includes normal operation and abnormal operation, count the number of IoT devices currently operating normally and abnormally;

[0059] Extract alarm information at the current time point, and count the number of alarms currently pending and the number of alarms already processed;

[0060] Extract the early warning information at the current time point, and count the number of early warnings pending and the number of early warnings already processed;

[0061] Calculate the processing index of the current alarm information based on the number of alarms to be processed and the number of alarms already processed;

[0062] Calculate the risk index of the current warning information based on the number of warnings pending and the number of warnings already processed;

[0063] Calculate the cumulative risk value based on the processing index of the current alarm information and the risk index of the current early warning information;

[0064] Real-time monitoring of the dynamic security index of the cloud platform;

[0065] When the dynamic safety index is lower than the safety threshold, a maintenance signal is automatically sent to the administrator.

[0066] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: The IoT device closed-loop management cloud platform and method based on dynamic security index provided by this invention includes a database server, an application server, a monitoring module for real-time collection and synchronization of device operating status data, an alarm module for judging the working status of sensors, an early warning module for providing early warnings for future time nodes, a processing module for calculating the dynamic security index of the cloud platform, and a feedback module for bidirectional transmission; the monitoring module, alarm module, early warning module, processing module, and feedback module are all deployed in the application server, and the operating data, alarm information, early warning information, and processing results generated by each module are all stored in the database server; by introducing the dynamic security index, real-time adjustments can be made to the dynamic changes of potential device faults, avoiding response lag; through a complete closed-loop management mechanism, processing efficiency is improved, forming an effective device management system. Attached Figure Description

[0067] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0068] in:

[0069] Figure 1 This is a schematic diagram of the structure in an embodiment of the present invention. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0071] like Figure 1 As shown, this is an embodiment of the present invention. This embodiment provides a cloud platform for closed-loop management and control of IoT devices based on dynamic security index, including a database server, an application server, and a monitoring module, an alarm module, an early warning module, a processing module, and a feedback module deployed in the application server.

[0072] The monitoring module, alarm module, early warning module, processing module, and feedback module are respectively used to generate operating data, alarm information, early warning information, and processing results, and all of the above information is stored in the database server;

[0073] The monitoring module is used to collect and synchronize equipment operating status data in real time;

[0074] The alarm module is used to determine the working status of the sensor and to issue multi-channel alarms according to the preset push mechanism.

[0075] The early warning module is used to provide early warnings for future time points and generate early warning information;

[0076] The processing module is used to calculate the dynamic security index of the cloud platform;

[0077] The feedback module is used to transmit information between the user and the system in both directions, and supports multimodal feedback input and command response.

[0078] Specifically, the monitoring module includes a data acquisition unit and a transmission unit:

[0079] The acquisition unit collects real-time device operation status data of IoT devices through a multi-source sensor network deployed on IoT terminals.

[0080] The transmission unit uses a secure encrypted transmission protocol to synchronously transmit the collected device operation data and monitoring data from multiple sensors to the database server for storage, and sends them to the alarm module in real time for working status analysis.

[0081] Specifically, the alarm module includes a judgment unit, a level classification unit, and an alarm unit;

[0082] The judgment unit determines the working state of the sensor based on a preset dynamic threshold judgment mechanism.

[0083] When the sensor's monitoring data is within the alarm threshold interval, the sensor is determined to be in normal working condition. The alarm threshold interval includes an upper limit value and a lower limit value.

[0084] When the sensor's monitoring data is outside the alarm threshold interval, the sensor is determined to be in an abnormal working state.

[0085] The level classification unit extracts monitoring data from sensors in abnormal working states and classifies alarm levels according to a preset sensor level classification mechanism.

[0086] The alarm unit generates alarm information according to a preset push mechanism and performs multi-channel alarms.

[0087] Specifically, the early warning module includes an extraction unit and a prediction unit:

[0088] The extraction unit is used to extract the working logs and real-time monitoring data of the multi-source sensors from the database server. The working logs include historical monitoring data and early warning types of the multi-source sensors. The early warning types include normal, high threshold, and fault warnings.

[0089] The prediction unit establishes a feature dataset based on historical monitoring data, constructs a prediction model based on an LSTM neural network, and outputs the warning probability of different warning types within future time nodes after inputting time-series feature data. Based on the warning probability, it generates a descending sequence of warning types and takes the first warning type in the sequence as the warning type for the future time node.

[0090] When the warning type for the future time node is not normal, a warning message is generated based on the warning type.

[0091] Specifically, the processing module includes a statistics unit and a security index unit:

[0092] The statistical unit acquires and processes the device operating status data of all IoT devices through the monitoring module. The operating status includes normal operation and abnormal operation. The statistical unit is used to count the number of IoT devices that are currently operating normally and abnormally.

[0093] The statistics unit is also used to extract alarm information from the alarm module and to count the number of alarms currently pending and those already processed.

[0094] The statistical unit further extracts the early warning information at the current time point and counts the number of early warnings that are currently pending and have been processed.

[0095] The security index unit, based on the aforementioned statistical data, is used to calculate the dynamic security index of the cloud platform in real time. The calculation formula is as follows:

[0096]

[0097] Where S represents the dynamic security index of the closed-loop management cloud platform, M represents the total number of sensors, and w m V represents the dynamic weight of the m-th sensor. m Let N represent the standardized risk value of the m-th sensor, w represent the dynamic weight of the IoT device, and N represent the dynamic risk value of the m-th sensor. Normal N Unusual These represent the number of IoT devices currently operating normally and those operating abnormally, respectively. This represents the cumulative risk value;

[0098] When the dynamic safety index is less than the safety threshold, the system will automatically send a maintenance signal to the management personnel.

[0099] Specifically, the risk accumulation value is used to reflect the accumulated risk of not processing alarm and warning information in a timely manner;

[0100] Based on the number of alarms pending and the number of alarms already processed, the processing index of the current alarm information is calculated using the following formula:

[0101]

[0102] Where H represents the processing index of the current alarm information, and I represents the alarm level. N represents the risk weight of an alarm message with alarm level i. i M represents the number of alarms of level i that are currently pending processing. i This indicates the number of alarms of level i that have been processed so far;

[0103] Based on the number of alerts pending and the number of alerts already processed, calculate the risk index of the current alert information:

[0104]

[0105] Where P represents the risk index of the current warning information, γ1 and γ2 represent the risk weights of high-threshold warnings and fault warnings, respectively, and N pre1 N pre2 N represents the number of high-threshold warnings and the number of fault warnings among the processed warnings, respectively. finish Indicates the number of warnings processed, and γ3 represents the error;

[0106] Based on the processing index of the current alarm information and the risk index of the current early warning information, the cumulative risk value is calculated using the following formula:

[0107]

[0108] Wherein, ρ1 and ρ2 represent the risk coefficients of alarm information and early warning information, respectively.

[0109] Specifically, the feedback module includes an interaction unit and a visualization unit:

[0110] The interaction unit is used to transmit information between the user and the system in both directions, and supports multimodal feedback input and command response;

[0111] The visualization unit automatically generates corresponding charts from the data generated by the monitoring module, alarm module, early warning module, processing module, and interaction unit based on the visualization template.

[0112] Specifically, the minimum environmental standards for the database server include a quad-core processor, 8GB of I / O-optimized memory, a 50GB hard drive, and a standard network connection with 5Mbps bandwidth; the minimum environmental standards for the application server environment include an eight-core processor, 32GB of I / O-optimized memory, a 100GB hard drive, and a standard network connection with 10Mbps bandwidth.

[0113] A closed-loop management and control method for IoT devices based on a dynamic security index, the method comprising the following steps:

[0114] Step S100: Collect the device operation status data of IoT devices in real time through the multi-source sensor network of IoT terminals, and record the monitoring data of multi-source sensors and the device operation status data of IoT devices in the database simultaneously;

[0115] Step S200: Based on the monitoring data from the multi-source sensors and the preset dynamic threshold determination mechanism, determine the working status of each sensor, including normal working status and abnormal working status.

[0116] When the sensor's monitoring data is within the alarm threshold interval, the sensor is determined to be in normal working condition. The alarm threshold interval includes an upper limit value and a lower limit value. When the sensor's monitoring data is not within the alarm threshold interval, the sensor is determined to be in abnormal working condition.

[0117] Extract monitoring data from sensors in abnormal working state, classify alarm levels according to a preset sensor level classification mechanism, generate alarm information according to a preset push mechanism, and perform multi-channel alarms, including SMS notification, email notification, APP notification and voice notification.

[0118] Step S300: Extract the working logs and real-time monitoring data of the multi-source sensors from the database server. The working logs include historical monitoring data and early warning types of the multi-source sensors. The early warning types include normal, high threshold early warning and fault early warning.

[0119] A feature dataset is established based on historical monitoring data. A prediction model is constructed based on an LSTM neural network. After inputting time-series feature data, the warning probability of different warning types within future time nodes is output. The warning types are generated in descending order according to the warning probability. The first warning type in the sequence is taken as the warning type for the future time node.

[0120] When the warning type at the future time node is not normal, a warning message is generated based on the warning type. The warning message includes the warning time node, warning type, sensor name, and sensor data.

[0121] Step S400: Based on the device operation status data of all IoT devices at the current time node, the operation status includes normal operation and abnormal operation, count the number of IoT devices currently operating normally and abnormally;

[0122] Extract alarm information at the current time point, and count the number of alarms currently pending and the number of alarms already processed;

[0123] Extract the early warning information at the current time point, and count the number of early warnings pending and the number of early warnings already processed;

[0124] Calculate the processing index of the current alarm information based on the number of alarms pending and the number of alarms already processed:

[0125]

[0126] Where H represents the processing index of the current alarm information, and I represents the alarm level. N represents the risk weight of an alarm message with alarm level i. i M represents the number of alarms of level i that are currently pending processing. i This indicates the number of alarms of level i that have been processed so far;

[0127] Based on the number of alerts pending and the number of alerts already processed, calculate the risk index of the current alert information:

[0128]

[0129] Where P represents the risk index of the current warning information, γ1 and γ2 represent the risk weights of high-threshold warnings and fault warnings, respectively, and N pre1 N pre2 N represents the number of high-threshold warnings and the number of fault warnings among the processed warnings, respectively. finish Indicates the number of warnings processed, and γ3 represents the error;

[0130] Based on the processing index of the current alarm information and the risk index of the current warning information, the cumulative risk value is calculated using the following formula:

[0131]

[0132] Wherein, ρ1 and ρ2 represent the risk coefficients of alarm information and early warning information, respectively;

[0133] The dynamic security index of the real-time monitoring cloud platform is calculated using the following formula:

[0134]

[0135] Where S represents the dynamic security index of the closed-loop management cloud platform, M represents the total number of sensors, and w m V represents the dynamic weight of the m-th sensor. m Let N represent the standardized risk value of the m-th sensor, w represent the dynamic weight of the IoT device, and N represent the dynamic risk value of the m-th sensor. Normal N Unusual These represent the number of IoT devices currently operating normally and those operating abnormally, respectively. This represents the cumulative risk value;

[0136] When the dynamic safety index is lower than the safety threshold, a maintenance signal is automatically sent to the administrator.

[0137] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any other combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product, which includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another.

[0138] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. This storage medium can be a read-only memory, a disk, or an optical disk, etc.

[0139] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in this application, and these should all be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A cloud platform for closed-loop management and control of IoT devices based on dynamic security index, characterized in that: It includes a database server, an application server, and monitoring modules, alarm modules, early warning modules, processing modules, and feedback modules deployed in the application server; The monitoring module, alarm module, early warning module, processing module, and feedback module are respectively used to generate operating data, alarm information, early warning information, and processing results, and all of the above information is stored in the database server; The monitoring module is used to collect and synchronize equipment operating status data in real time; The alarm module is used to determine the working status of the sensor and to issue multi-channel alarms according to the preset push mechanism. The early warning module is used to provide early warnings for future time points and generate early warning information; The processing module is used to calculate the dynamic security index of the cloud platform, and includes a statistical unit and a security index unit: The statistics unit is used to count the number of IoT devices that are currently operating normally and those that are operating abnormally. Extract alarm information from the alarm module and count the number of alarms currently pending and those already processed. Extract the early warning information at the current time point, and count the number of early warnings that are currently pending and those that have been processed. The security index unit calculates the dynamic security index of the cloud platform in real time based on the above statistical data: Where S represents the dynamic security index of the closed-loop management cloud platform, M represents the total number of sensors, and w m V represents the dynamic weight of the m-th sensor. m Let N represent the standardized risk value of the m-th sensor, w represent the dynamic weight of the IoT device, and N represent the dynamic risk value of the m-th sensor. Normal N Unusual These represent the number of IoT devices currently operating normally and those operating abnormally, respectively. This represents the cumulative risk value; When the dynamic safety index is less than the safety threshold, the system will automatically send a maintenance signal to the management personnel. The risk accumulation value is used to reflect the accumulated risk of failing to process alarm and warning information in a timely manner; Based on the number of alarms pending and the number of alarms already processed, the processing index of the current alarm information is calculated using the following formula: Where H represents the processing index of the current alarm information, and I represents the alarm level. N represents the risk weight of an alarm message with alarm level i. i M represents the number of alarms of level i that are currently pending processing. i This indicates the number of alarms of level i that have been processed so far; Based on the number of alerts pending and the number of alerts already processed, calculate the risk index of the current alert information: Where P represents the risk index of the current warning information, γ1 and γ2 represent the risk weights of high-threshold warnings and fault warnings, respectively, and N pre1 N pre2 N represents the number of high-threshold warnings and the number of fault warnings among the processed warnings, respectively. finish Indicates the number of warnings processed, and γ3 represents the error; Based on the processing index of the current alarm information and the risk index of the current early warning information, the cumulative risk value is calculated using the following formula: Wherein, ρ1 and ρ2 represent the risk coefficients of alarm information and early warning information, respectively; The feedback module is used to transmit information between the user and the system in both directions, and supports multimodal feedback input and command response.

2. The cloud platform for closed-loop management and control of IoT devices based on dynamic security index as described in claim 1, characterized in that, The monitoring module includes a data acquisition unit and a transmission unit: The acquisition unit collects real-time device operation status data of IoT devices through a multi-source sensor network deployed on IoT terminals. The transmission unit uses a secure encrypted transmission protocol to synchronously transmit the collected device operation data and monitoring data from multiple sensors to the database server for storage, and sends them to the alarm module in real time for working status analysis.

3. The cloud platform for closed-loop management and control of IoT devices based on dynamic security index as described in claim 2, characterized in that, The alarm module includes a judgment unit, a level classification unit, and an alarm unit; The judgment unit determines the working state of the sensor based on a preset dynamic threshold judgment mechanism. When the sensor's monitoring data is within the alarm threshold interval, the sensor is determined to be in normal working condition. The alarm threshold interval includes an upper limit value and a lower limit value. When the sensor's monitoring data is outside the alarm threshold interval, the sensor is determined to be in an abnormal working state. The level classification unit extracts monitoring data from sensors in abnormal working states and classifies alarm levels according to a preset sensor level classification mechanism. The alarm unit generates alarm information according to a preset push mechanism and performs multi-channel alarms.

4. The cloud platform for closed-loop management and control of IoT devices based on dynamic security index as described in claim 3, characterized in that, The early warning module includes an extraction unit and a prediction unit: The extraction unit is used to extract the working logs and real-time monitoring data of the multi-source sensors from the database server. The working logs include historical monitoring data and early warning types of the multi-source sensors. The early warning types include normal, high threshold, and fault warnings. The prediction unit establishes a feature dataset based on historical monitoring data, constructs a prediction model based on an LSTM neural network, and outputs the warning probability of different warning types within future time nodes after inputting time-series feature data. Based on the warning probability, it generates a descending sequence of warning types and takes the first warning type in the sequence as the warning type for the future time node. When the warning type for the future time node is not normal, a warning message is generated based on the warning type.

5. The cloud platform for closed-loop management and control of IoT devices based on dynamic security index as described in claim 4, characterized in that, The processing module includes a statistics unit and a security index unit: The statistical unit acquires and processes the device operating status data of all IoT devices through the monitoring module. The operating status includes normal operation and abnormal operation. The statistical unit is used to count the number of IoT devices that are currently operating normally and abnormally. The statistics unit is also used to extract alarm information from the alarm module and to count the number of alarms currently pending and those already processed. The statistical unit further extracts the early warning information at the current time point and counts the number of early warnings that are currently pending and have been processed. The security index unit, based on the aforementioned statistical data, is used to calculate the dynamic security index of the cloud platform in real time. The calculation formula is as follows: Where S represents the dynamic security index of the closed-loop management cloud platform, M represents the total number of sensors, and w m V represents the dynamic weight of the m-th sensor. m Let N represent the standardized risk value of the m-th sensor, w represent the dynamic weight of the IoT device, and N represent the dynamic risk value of the m-th sensor. Normal N Unusual These represent the number of IoT devices currently operating normally and those operating abnormally, respectively. This represents the cumulative risk value; When the dynamic safety index is less than the safety threshold, the system will automatically send a maintenance signal to the management personnel.

6. The cloud platform for closed-loop management and control of IoT devices based on dynamic security index as described in claim 5, characterized in that, The feedback module includes an interactive unit and a visualization unit: The interaction unit is used to transmit information between the user and the system in both directions, and supports multimodal feedback input and command response; The visualization unit automatically generates corresponding charts from the data generated by the monitoring module, alarm module, early warning module, processing module, and interaction unit based on the visualization template.

7. The cloud platform for closed-loop management and control of IoT devices based on dynamic security index as described in claim 6, characterized in that, The minimum environmental standards for the database server include a quad-core processor, 8GB of I / O-optimized memory, a 50GB hard drive, and a standard network connection with 5Mbps bandwidth; the minimum environmental standards for the application server environment include an eight-core processor, 32GB of I / O-optimized memory, a 100GB hard drive, and a standard network connection with 10Mbps bandwidth.

8. The closed-loop control method of the IoT device closed-loop control cloud platform based on dynamic security index according to any one of claims 1-7, characterized in that, The method includes the following steps: Step S100: Collect the device operation status data of IoT devices in real time through the multi-source sensor network of IoT terminals, and record the monitoring data of multi-source sensors and the device operation status data of IoT devices in the database simultaneously; Step S200: Based on the monitoring data from the multi-source sensors and the preset dynamic threshold determination mechanism, determine the working status of each sensor, including normal working status and abnormal working status. When the sensor's monitoring data is within the alarm threshold interval, the sensor is determined to be in normal working condition. The alarm threshold interval includes an upper limit value and a lower limit value. When the sensor's monitoring data is not within the alarm threshold interval, the sensor is determined to be in abnormal working condition. Extract monitoring data from sensors in abnormal working state, classify alarm levels according to a preset sensor level classification mechanism, generate alarm information according to a preset push mechanism, and perform multi-channel alarms, including SMS notification, email notification, APP notification and voice notification. Step S300: Extract the working logs and real-time monitoring data of the multi-source sensors from the database server. The working logs include historical monitoring data and early warning types of the multi-source sensors. The early warning types include normal, high threshold early warning and fault early warning. A feature dataset is established based on historical monitoring data. A prediction model is constructed based on an LSTM neural network. After inputting time-series feature data, the warning probability of different warning types within future time nodes is output. The warning types are generated in descending order according to the warning probability. The first warning type in the sequence is taken as the warning type for the future time node. When the warning type at the future time node is not normal, a warning message is generated based on the warning type. The warning message includes the warning time node, warning type, sensor name, and sensor data. Step S400: Based on the device operation status data of all IoT devices at the current time node, the operation status includes normal operation and abnormal operation, count the number of IoT devices currently operating normally and abnormally; Extract alarm information at the current time point, and count the number of alarms currently pending and the number of alarms already processed; Extract the early warning information at the current time point, and count the number of early warnings pending and the number of early warnings already processed; Calculate the processing index of the current alarm information based on the number of alarms pending and the number of alarms already processed: Where H represents the processing index of the current alarm information, and I represents the alarm level. N represents the risk weight of an alarm message with alarm level i. i M represents the number of alarms of level i that are currently pending processing. i This indicates the number of alarms of level i that have been processed so far; Based on the number of alerts pending and the number of alerts already processed, calculate the risk index of the current alert information: Where P represents the risk index of the current warning information, γ1 and γ2 represent the risk weights of high-threshold warnings and fault warnings, respectively, and N pre1 N pre2 N represents the number of high-threshold warnings and the number of fault warnings among the processed warnings, respectively. finish Indicates the number of warnings processed, and γ3 represents the error; Based on the processing index of the current alarm information and the risk index of the current early warning information, the cumulative risk value is calculated using the following formula: Wherein, ρ1 and ρ2 represent the risk coefficients of alarm information and early warning information, respectively; The dynamic security index of the real-time monitoring cloud platform is calculated using the following formula: Where S represents the dynamic security index of the closed-loop management cloud platform, M represents the total number of sensors, and w m V represents the dynamic weight of the m-th sensor. m Let N represent the standardized risk value of the m-th sensor, w represent the dynamic weight of the IoT device, and N represent the dynamic risk value of the m-th sensor. Normal N Unusual These represent the number of IoT devices currently operating normally and those operating abnormally, respectively. This represents the cumulative risk value; When the dynamic safety index is lower than the safety threshold, a maintenance signal is automatically sent to the administrator.

Images