Intelligent product risk identification and control method and system based on big data

Abstract

The invention discloses an intelligent product risk identification and control method and system based on big data, and relates to the technical field of product risk identification and control. The intelligent product risk identification and control method based on big data comprises the following steps: obtaining a relevance coincidence index; obtaining a security evaluation value; and obtaining a control effect evaluation value. According to the invention, correlation analysis is carried out on the obtained intelligent security data to obtain the correlation coincidence index and determine whether risk identification is carried out, then the risk identification result is input into the constructed risk identification model to obtain the security risk assessment value, and whether closed-loop feedback control is carried out is determined. Finally, whether closed-loop feedback control is completed is judged according to the obtained control effect evaluation value, the effect of improving the security risk evaluation effectiveness of the intelligent product is achieved, and the problem that the security risk evaluation effectiveness of the intelligent product is low in the prior art is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of product risk identification and control, and in particular to a method and system for intelligent product risk identification and control based on big data. Background Art

[0002] With the rapid development of information technology and the widespread application of big data technology, smart products have become an important part of modern life, especially in the field of smart home. These smart home devices realize environmental perception, data collection and processing, and intelligent decision-making and control by integrating advanced sensors, processors and communication technologies. Among them, sensors can monitor various parameters in the home environment in real time, such as temperature, humidity, light, air quality, human activity, etc. The processor is the "brain" of the smart home device control system, responsible for processing the data collected by the sensors and making decisions according to preset algorithms and rules. Communication technology is the key to the interconnection of smart home device control systems. However, with the popularization and complexity of smart products, their potential risks are becoming increasingly prominent. In order to effectively deal with these risks, smart product risk identification and control methods and systems based on big data have emerged.

[0003] In the prior art, the collected raw data is processed to obtain the correlation between the data, and then the risk factors are identified and quantified through machine learning algorithms. At the same time, the results of risk quantification are statistically analyzed and risk scoring is performed. Finally, corresponding risk control strategies are formulated according to the results of risk scoring, thereby achieving accurate identification of product risks.

[0004] For example, the invention patent with announcement number: CN116633682B announces an intelligent identification method and system based on security product risk threats, including: obtaining complex high-frequency sensitive information flows, and classifying the complex high-frequency sensitive information flows using the FCM clustering algorithm; randomly selecting c points from all complex high-frequency sensitive information flows as cluster centers, and selecting points to minimize the objective function; taking the mean of each cluster as the new center, iterating until the distribution of the complex high-frequency sensitive information flows in the cluster no longer changes, and obtaining the Euclidean distance; collecting and arranging complex high-frequency sensitive information flows with large similarity to obtain the target complex sensitive high-frequency information.

[0005] For example, a patent application with publication number: CN117896137A discloses a computer network security intelligent analysis system and method based on big data, including: collecting and processing network data, generating standard network data and performing time series analysis, and generating time series analysis results; further analyzing the time series analysis results, generating network abnormal behavior data and triggering an abnormal alarm level mechanism; collecting network threat intelligence information, using natural language processing technology to process the network threat intelligence information, and generating network threat intelligence data; using association analysis technology to mine the association rules between network threat intelligence data and time series analysis results, and generating threat behavior pattern data; constructing a network topology structure diagram, and using a graph computing analysis method to associate the network abnormal behavior data with the network topology structure diagram to generate threat behavior path data; integrating the association analysis technology, the graph computing analysis method and the security management tool, and triggering the security management tool to process the potential threat behavior of the network according to the threat behavior pattern data and the threat behavior path data.

[0006] However, in the process of implementing the technical solution of the invention in the embodiments of the present application, the present application found that the above technology has at least the following technical problems: In the existing technology, the collection of raw data is usually affected by sensor accuracy and environmental factors, resulting in low data quality, which in turn affects subsequent data processing and risk quantification. Secondly, risk scoring has subjective factors and cannot fully reflect the actual situation of the risk, resulting in low effectiveness of security risk assessment of smart products. Summary of the invention

[0007] The embodiments of the present application solve the problem of low effectiveness of security risk assessment of smart products in the prior art by providing a smart product risk identification and control method and system based on big data, thereby improving the effectiveness of security risk assessment of smart products.

[0008] The embodiment of the present application provides a method for risk identification and control of smart products based on big data, comprising the following steps: Step 1, obtaining the smart security data of the smart product at the current monitoring moment and performing a correlation analysis to obtain a correlation compliance index, and judging whether to perform risk identification based on the obtained correlation compliance index, wherein the correlation compliance index is used to quantify the degree of correlation between the smart security data; Step 2, if risk identification is performed, inputting the result of risk identification into the constructed risk identification model to obtain a security risk assessment value, and judging whether to perform closed-loop feedback control based on the obtained security risk assessment value, wherein the security risk assessment value is used to evaluate the probability of occurrence of the security risk at the current monitoring moment; Step 3, if closed-loop feedback control is performed, obtaining a control effect assessment value based on the security risk control effect in the closed-loop feedback control process, and judging whether the closed-loop feedback control is completed based on the obtained control effect assessment value, wherein the control effect assessment value is used to evaluate the effectiveness of the closed-loop feedback control.

[0009] Furthermore, the correlation compliance index is obtained by the following method: when the ambient temperature is greater than the reference ambient temperature, the ambient humidity is greater than the reference ambient humidity, and the smoke concentration is greater than the reference smoke concentration, the ambient temperature score, ambient humidity score, and smoke concentration score of the smart product at the current monitoring time are obtained and added to obtain a first correlation compliance index; the device working power score and the device response time score of the smart product at the current monitoring time are obtained and added to obtain a second correlation compliance index, and the usage frequency score and the number of incorrect operation scores of the smart product at the current monitoring time are obtained and added to obtain a third correlation compliance index; the correlation compliance index is obtained according to the obtained first correlation compliance index, the second correlation compliance index, and the third correlation compliance index in combination with the corresponding correlation compliance index weight factor; the ambient temperature score The number is the ratio of ambient temperature to the maximum allowable ambient temperature; the ambient humidity score is the ratio of ambient humidity to the maximum allowable ambient humidity; the smoke concentration score is the ratio of smoke concentration to the maximum allowable smoke concentration; the device working power score is the ratio of the absolute value of the difference between the device working power and the reference device working power to the reference device working power; the device response time score is the ratio of the absolute value of the difference between the device response time and the reference device response time to the reference device response time; the usage frequency score is the ratio of the smart product usage frequency to the reference smart product usage frequency; the number of misoperations score is the ratio of the number of misoperations of the smart product to the maximum allowable number of misoperations of the smart product; the correlation compliance index weight factor includes a first correlation compliance index weight factor, a second correlation compliance index weight factor and a third correlation compliance index weight factor.

[0010] Furthermore, the specific steps for obtaining the security risk assessment value are as follows: counting the number of abnormal access users in the risk identification process within a preset monitoring period, and determining whether the number of abnormal access users is within the allowed range of the abnormal access number. If so, no calculation is performed; otherwise, an abnormal access score is obtained by combining the reference access number in the database; counting the number of environmental risk points, the number of equipment failure risk points, and the number of abnormal access risk points in the risk identification process within the preset monitoring period, and obtaining the environmental risk score, the equipment failure risk score, and the abnormal access risk score by combining the reference environmental risk point number, the reference equipment failure risk point number, and the reference abnormal access risk point number in the database. number; obtain the security risk assessment value according to the acquired abnormal access score, environmental risk score, equipment failure risk score, abnormal access risk score, correlation compliance index and the influencing factor of the correlation compliance index; the abnormal access score is the ratio of the number of abnormal access users to the reference access number; the abnormal access user is an unauthorized access user; the environmental risk score is the ratio of the number of environmental risk points to the number of reference environmental risk points; the equipment failure risk score is the ratio of the number of equipment failure risk points to the number of reference equipment failure risk points; the abnormal access risk score is the ratio of the number of abnormal access risk points to the number of reference abnormal access risk points.

[0011] Furthermore, the specific limiting expression of the security risk assessment value is: ; ; ; ; ; Where t is the number of the current monitoring time in the preset monitoring period, , T is the total number of current monitoring moments within the preset monitoring period, e is a natural constant, It indicates the security risk assessment value of the smart product during the preset monitoring period. Indicates that the correlation meets the indicator impact factor, Indicates the environmental risk score of the smart product during the preset monitoring period. Indicates the number of risk points in the reference environment, Indicates the number of environmental risk points of smart products within the preset monitoring period. Indicates the equipment failure risk score of the smart product within the preset monitoring period. represents the number of reference equipment failure risk points, Indicates the number of equipment failure risk points of smart products within the preset monitoring period. Indicates the abnormal access risk score of smart products within the preset monitoring period. Indicates the number of reference abnormal access risk points, Indicates the number of abnormal access risk points of smart products within the preset monitoring period. Indicates the abnormal access score of smart products within the preset monitoring period. Indicates the number of reference visits Indicates the number of abnormal users accessing the smart product during the preset monitoring period. Indicates the allowed range of abnormal access quantity. It indicates that the correlation of smart products at the current monitoring time t meets the index, Indicates that the preset correlation meets the indicator.

[0012] Furthermore, the control effect evaluation value is obtained by the following method: M1, obtaining the risk control execution time in the closed-loop feedback control process and determining whether it is less than the preset risk control execution time. If so, executing M2, otherwise adjusting the risk control parameters and re-performing closed-loop feedback control. The risk control parameters include response speed and control intensity; M2, obtaining the risk control execution time score, and at the same time obtaining the risk monitoring time, risk warning time and control effect feedback time in the closed-loop feedback control process and performing ratio operations with the total closed-loop feedback control time to obtain the risk monitoring time score, risk warning time score and control effect feedback time score, respectively. The risk control execution time score is the ratio of the risk control execution time to the total closed-loop feedback control time; M3, processing the obtained risk control execution time score, risk monitoring time score, risk warning time score and control effect feedback time score respectively to obtain the control effect evaluation value.

[0013] The embodiment of the present application provides a smart product risk identification and control system based on big data, including: a correlation compliance index acquisition module, a security risk assessment value acquisition module and a control effect assessment value acquisition module; wherein the correlation compliance index acquisition module is used to acquire the smart security data of the smart product at the current monitoring moment and perform correlation analysis to acquire the correlation compliance index, and judge whether to perform risk identification based on the acquired correlation compliance index, and the correlation compliance index is used to quantify the degree of correlation between smart security data; the security risk assessment value acquisition module is used to input the result of risk identification into the constructed risk identification model to obtain the security risk assessment value if risk identification is performed, and at the same time judge whether to perform closed-loop feedback control based on the acquired security risk assessment value, and the security risk assessment value is used to evaluate the probability of security risk occurring at the current monitoring moment; the control effect assessment value acquisition module is used to acquire the control effect assessment value according to the security risk control effect in the closed-loop feedback control process if closed-loop feedback control is performed, and at the same time judge whether the closed-loop feedback control is completed based on the acquired control effect assessment value, and the control effect assessment value is used to evaluate the effectiveness of the closed-loop feedback control.

[0014] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages: 1. Through correlation analysis on the acquired intelligent security data to obtain correlation compliance indicators and determine whether to perform risk identification, the results of risk identification are then input into the established risk identification model to obtain a security risk assessment value and determine whether to perform closed-loop feedback control. Finally, according to the security risk control effect in the closed-loop feedback control process, a control effect assessment value is obtained and it is determined whether the closed-loop feedback control is completed, thereby improving the accuracy of obtaining security risk assessment values, thereby achieving an improvement in the effectiveness of security risk assessment of smart products, and effectively solving the problem of low effectiveness of security risk assessment of smart products in the prior art.

[0015] 2. When the ambient temperature is greater than the reference ambient temperature, the ambient humidity is greater than the reference ambient humidity, and the smoke concentration is greater than the reference smoke concentration, the correlation compliance index is obtained according to the first correlation compliance index, the second correlation compliance index, and the third correlation compliance index, and combined with the corresponding correlation compliance index weight factor, thereby achieving more accurate acquisition of the correlation compliance index, and then achieving more accurate analysis of the degree of correlation between intelligent security data.

[0016] 3. By counting the number of abnormal access users within the allowed range of abnormal access numbers and combining it with the reference access number in the database, the abnormal access score is obtained. At the same time, the security risk assessment value is obtained based on the obtained abnormal access score, environmental risk score, equipment failure risk score, abnormal access risk score, correlation compliance index and the correlation compliance index influencing factor, thereby achieving more accurate acquisition of security risk assessment values, and further improving the accuracy and reliability of security risk identification and assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A flowchart of a method for identifying and controlling smart product risks based on big data provided in an embodiment of the present application; Figure 2 A two-dimensional coordinate diagram of security risk assessment value-correlation compliance index provided in an embodiment of the present application; Figure 3 A two-dimensional coordinate diagram of the security risk assessment value-security risk assessment coefficient provided in an embodiment of the present application; Figure 4 A schematic diagram of the structure of the intelligent product risk identification and control system based on big data provided in the embodiment of the present application; Figure 5 A specific flow chart of risk identification and control provided in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The embodiments of the present application solve the problem of low effectiveness of security risk assessment of smart products in the prior art by providing a smart product risk identification and control method and system based on big data. The smart security data of the smart product at the current monitoring moment is obtained, and correlation analysis is performed on the obtained smart security data according to a machine learning algorithm to obtain correlation compliance indicators. Whether to perform risk identification is determined based on the obtained correlation compliance indicators. If risk identification is performed, the result of risk identification is input into the constructed risk identification model to obtain a security risk assessment value. At the same time, based on the obtained security risk assessment value, it is determined whether to perform closed-loop feedback control. If closed-loop feedback control is performed, a control effect assessment value is obtained according to the security risk control effect in the closed-loop feedback control process. At the same time, based on the obtained control effect assessment value, it is determined whether the closed-loop feedback control is completed, thereby improving the effectiveness of security risk assessment of smart products.

[0019] The technical solution in the embodiment of the present application is to solve the problem of low effectiveness of security risk assessment of smart products mentioned above, and the overall idea is as follows: By performing correlation analysis on the acquired intelligent security data to obtain correlation compliance indicators and determine whether to perform risk identification, the results of risk identification are then input into the constructed risk identification model to obtain a security risk assessment value and determine whether to perform closed-loop feedback control. Finally, based on the security risk control effect in the closed-loop feedback control process, the control effect assessment value is obtained and it is determined whether the closed-loop feedback control is completed, thereby achieving the effect of improving the effectiveness of security risk assessment of smart products.

[0020] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0021] like Figure 1 As shown, it is a flow chart of the smart product risk identification and control method based on big data provided by the embodiment of the present application. The smart product risk identification and control method based on big data provided by the embodiment of the present application includes the following steps: Step 1, obtaining the smart security data of the smart product at the current monitoring moment, and performing correlation analysis on the obtained smart security data according to the machine learning algorithm to obtain the correlation compliance index, and judging whether to perform risk identification based on the obtained correlation compliance index, and the correlation compliance index is used to quantify the degree of correlation between the smart security data; Step 2, if risk identification is performed, the result of risk identification is input into the constructed risk The security risk assessment value is obtained from the identification model, and at the same time, based on the obtained security risk assessment value, it is determined whether to perform closed-loop feedback control. The risk identification model is used to obtain the security risk assessment value, and the security risk assessment value is used to evaluate the probability of security risk occurring at the current monitoring moment. The closed-loop feedback control includes risk monitoring, risk early warning, risk control execution, and control effect feedback. In step three, if closed-loop feedback control is performed, a control effect evaluation value is obtained based on the security risk control effect during the closed-loop feedback control process, and at the same time, based on the obtained control effect evaluation value, it is determined whether the closed-loop feedback control is completed, and the control effect evaluation value is used to evaluate the effectiveness of the closed-loop feedback control.

[0022] It should be added that smart security data includes environmental monitoring data, equipment operation status data and user behavior data; environmental monitoring data includes ambient temperature, ambient humidity and smoke concentration; equipment operation status data includes equipment working power and equipment response time; user behavior data includes frequency of use of smart products and number of misoperations of smart products; correlation analysis includes first correlation analysis, second correlation analysis and third correlation analysis; the first correlation analysis is used to reflect the correlation between environmental monitoring data and reference environmental monitoring data in data control; reference environmental monitoring data includes reference ambient temperature, reference ambient humidity and reference smoke concentration; the second correlation analysis is used to reflect the correlation between equipment operation status data and reference equipment operation status data in the database; reference equipment operation status data includes reference equipment working power and reference equipment response time; the third correlation analysis is used to reflect the correlation between user behavior data and reference user behavior data in the database; reference user behavior data includes reference frequency of use of smart products and the maximum allowable number of misoperations of smart products.

[0023] In this embodiment, the reference ambient temperature is represented by the sum and average of the historical ambient temperatures of the area where the smart product is located in the preset database, the reference ambient humidity is represented by the sum and average of the historical ambient humidity of the area where the smart product is located in the preset database, the reference smoke concentration is represented by the sum and average of the historical smoke concentration of the area where the smart product is located in the preset database, the reference device working power is represented by the sum and average of the historical device working powers of the smart products in the preset database, the reference device response time is represented by the sum and average of the historical device response time of the smart products in the preset database, the reference smart product usage frequency is represented by the sum and average of the historical smart product usage frequencies of the smart products in the preset database, and the maximum allowable smart product misoperation times represents the maximum value of the historical smart product misoperation times of the smart product in the preset database.

[0024] In a specific security application scenario (such as a smart home control system), users can control various smart products in their home through mobile phone apps or voice assistants, and monitor the temperature of the kitchen or bedroom (i.e., ambient temperature) in real time through temperature sensors deployed on the walls of the kitchen or bedroom; monitor the humidity of the bedroom (i.e., ambient humidity) in real time through humidity sensors deployed on the ceiling of the bedroom; monitor the smoke concentration generated during cooking in real time through smoke sensors deployed at a fixed position above the kitchen stove; and monitor the power generated by smart home devices during operation (i.e., device working power) in real time through power sensors deployed on smart home devices. For example, the power generated by smart light bulbs in a day (24 hours) can be monitored in real time through power sensors deployed on smart light bulbs.

[0025] It should be added that a response time monitoring module (used to record the time required for the device to complete execution after receiving a user command) is integrated in the control system of the smart home device to record the device response time in real time. For example, a response time monitoring module is integrated in the smart door lock to record the time required for the smart door lock to complete execution after receiving an unlocking command; a usage frequency statistics module (used to record the number of times and time users use various smart products) is integrated in the control system of the smart home device to record the number of times and usage time of the corresponding smart product (such as a smart speaker) in a day (24 hours) in real time to obtain the usage frequency of the smart product (the ratio of the number of times used to 24 hours)

[0026] Specifically, an error operation detection module (for identifying the types of the number of user's error operations) is integrated into the control system of the smart home device to detect the number of user's error operations when using smart products (i.e., the number of smart product error operations) in real time. For example, the error operation detection module is integrated into the remote control of the smart TV to identify the number and type of the user's mistaken presses of the remote control buttons (such as mistakenly pressing the power off button).

[0027] Compared with the existing technology, this example can comprehensively reflect the operating status and security environment of smart products by acquiring smart security data including environmental monitoring data, equipment operating status data and user behavior data. At the same time, it uses machine learning algorithms to perform correlation analysis on smart security data, which can dig out the connections between data and improve the accuracy and efficiency of risk identification. Secondly, by building a risk identification model, it can automatically process and analyze security data and quickly identify potential risk points. In addition, the closed-loop feedback control mechanism can automatically monitor, warn, execute control and feedback control effects, realize intelligent and automated risk management, help reduce the probability and impact of risks, ensure that risks are effectively controlled, and thus achieve the improvement of the effectiveness of security risk assessment of smart products.

[0028] Furthermore, the correlation compliance index is obtained by the following method: when the ambient temperature is greater than the reference ambient temperature, the ambient humidity is greater than the reference ambient humidity, and the smoke concentration is greater than the reference smoke concentration, the ambient temperature score, ambient humidity score, and smoke concentration score of the smart product at the current monitoring time are obtained and added to obtain the first correlation compliance index (i.e., the correlation compliance index in the restriction expression). ); obtain the device working power score and device response time score of the smart product at the current monitoring time and perform addition operation to obtain the second correlation compliance index (that is, ), and at the same time obtain the usage frequency score and the number of wrong operation scores of the smart product at the current monitoring time and perform addition operations to obtain the third correlation compliance index (that is, the correlation compliance index in the restriction expression ); obtaining a correlation compliance index based on the obtained first correlation compliance index, the second correlation compliance index and the third correlation compliance index and combining the corresponding correlation compliance index weight factor.

[0029] Among them, the ambient temperature score is the ratio of the ambient temperature to the maximum allowable ambient temperature; the ambient humidity score is the ratio of the ambient humidity to the maximum allowable ambient humidity; the smoke concentration score is the ratio of the smoke concentration to the maximum allowable smoke concentration; the equipment working power score is the ratio of the absolute value of the difference between the equipment working power and the reference equipment working power to the reference equipment working power; the equipment response time score is the ratio of the absolute value of the difference between the equipment response time and the reference equipment response time to the reference equipment response time; the usage frequency score is the ratio of the smart product usage frequency to the reference smart product usage frequency; the number of erroneous operation score is the ratio of the number of erroneous operations of the smart product to the maximum allowable number of erroneous operations of the smart product; the correlation compliance indicator weight factor includes the first correlation compliance indicator weight factor, the second correlation compliance indicator weight factor and the third correlation compliance indicator weight factor.

[0030] In this embodiment, the maximum allowable ambient temperature represents the maximum value of the historical ambient temperature of the area where the smart product is located in the preset database, the maximum allowable ambient humidity represents the maximum value of the historical ambient humidity of the area where the smart product is located in the preset database, and the maximum allowable smoke concentration represents the maximum value of the historical smoke concentration of the area where the smart product is located in the preset database.

[0031] Specifically, the first correlation compliance indicator weight factor is the weight factor corresponding to the correlation compliance indicator acquisition process in the preset database, which represents the numerical value corresponding to the degree of influence of the first correlation compliance indicator on the correlation compliance indicator. When used, the weight factor corresponding to the first correlation compliance indicator can be directly obtained from the preset database, and the corresponding relationship can be a pre-set mapping relationship. For example, the first correlation compliance indicator corresponding to the first correlation compliance indicator in the correlation compliance indicator acquisition process and the weight factor corresponding to the first correlation compliance indicator in the preset database form a mapping set, and the real-time first correlation compliance indicator is input into the mapping set to obtain the corresponding first correlation compliance indicator weight factor, wherein the mapping relationship can be one-to-one or many-to-one. The value range of the first correlation compliance indicator weight factor in this example is .

[0032] The second relevance compliance indicator weight factor is the corresponding weight factor in the process of obtaining the relevance compliance indicator in the preset database, which represents the numerical value corresponding to the degree of influence of the second relevance compliance indicator on the relevance compliance indicator. When used, the weight factor corresponding to the second relevance compliance indicator can be directly obtained from the preset database, and the corresponding relationship can be a pre-set mapping relationship. For example, the second relevance compliance indicator corresponding to the second relevance compliance indicator in the process of obtaining the relevance compliance indicator and the weight factor corresponding to the second relevance compliance indicator in the preset database form a mapping set, and the real-time second relevance compliance indicator is input into the mapping set to obtain the corresponding second relevance compliance indicator weight factor, wherein the mapping relationship can be one-to-one or many-to-one. The value range of the second relevance compliance indicator weight factor in this example is .

[0033] It should be noted that the sum of the first relevance compliance indicator weight factor, the second relevance compliance indicator weight factor and the third relevance compliance indicator weight factor is 1.

[0034] Specifically, the specific restriction expression of the correlation compliance index is: ; ; ; ; Where t is the number of the current monitoring time in the preset monitoring period, , T is the total number of current monitoring moments within the preset monitoring period, e is a natural constant, It indicates that the correlation of smart products at the current monitoring time t meets the index, Indicates that the first correlation meets the indicator weight factor, Indicates the first correlation compliance index of the smart product at the current monitoring time t, Indicates the ambient temperature of the smart product at the current monitoring time t. Indicates the maximum allowable ambient temperature, Indicates the reference ambient temperature, It indicates the ambient humidity of the smart product at the current monitoring time t. Indicates the maximum allowable ambient humidity, Indicates the reference ambient humidity, It indicates the smoke concentration of the smart product at the current monitoring time t. Indicates the maximum allowable smoke concentration, Indicates the reference smoke density, Indicates that the second correlation meets the indicator weight factor, Indicates the second correlation compliance index of the smart product at the current monitoring time t. represents the usage frequency of smart products at the current monitoring time t, Indicates the frequency of use of reference smart products, It indicates the number of smart product misoperations at the current monitoring time t. Indicates the maximum number of times the intelligent product can be operated incorrectly. Indicates that the third correlation meets the indicator weight factor, It indicates that the third correlation compliance index of the smart product at the current monitoring time t is met. It indicates the working power of the smart product at the current monitoring time t. Indicates the reference equipment operating power, It indicates the device response time of the smart product at the current monitoring time t. Indicates the response time of the reference device.

[0035] It should be added that the statistical table of changes in correlation compliance indicators is shown in Table 1: Table 1 Statistics of changes in correlation compliance indicators

[0036] It should be understood that, as shown in Table 1, the correlation compliance index increases with the increase of the first correlation compliance index, the second correlation compliance index and the third correlation compliance index, wherein the first correlation compliance index increases with the increase of the ambient temperature, ambient humidity and smoke concentration, the second correlation compliance index increases with the increase of the frequency of use of the smart product and the number of misoperations of the smart product, and the third correlation compliance index increases with the absolute value of the difference between the working power of the device and the working power of the reference device (i.e., the absolute value of the difference between the working power of the device and the working power of the reference device in the calculation formula of the third correlation compliance index). ) and the absolute value of the difference between the device response time and the reference device response time (i.e., the value in the calculation formula of the third correlation compliance index ) increases with the increase of .

[0037] It should be noted that the first correlation compliance index also indirectly affects the value of the third correlation compliance index. Assuming that the working performance of the smoke alarm in the intelligent security system is affected by the ambient temperature and humidity, when the ambient temperature and humidity increase, the working power of the device may change, resulting in an increase in the difference between its actual working power and the reference working power. At the same time, due to changes in environmental conditions, the response time of the device may also be affected, resulting in an increase in the difference between the response time and the reference response time.

[0038] Therefore, the increase of the first correlation compliance index (due to the increase of ambient temperature and humidity) will indirectly lead to the increase of the third correlation compliance index (due to the change of equipment working power and response time). It helps the intelligent security system to more accurately identify and control security risks, thereby improving the effectiveness of security risk assessment for smart products and effectively solving the problem of low effectiveness of security risk assessment for smart products in the existing technology.

[0039] Furthermore, the specific process of determining whether to perform risk identification based on the obtained correlation compliance index is as follows: determining whether the obtained correlation compliance index is greater than a preset correlation compliance index: if the obtained correlation compliance index is greater than the preset correlation compliance index, then it indicates that the smart product at the current monitoring moment has security risks and security risk points are obtained. The security risk points include environmental risk points, equipment failure risk points and abnormal access risk points; if the obtained correlation compliance index is not greater than the preset correlation compliance index, then it indicates that the smart product at the current monitoring moment is in a normal working state. At this time, risk identification is not performed and the working state of the smart product continues to be monitored.

[0040] In this embodiment, the preset correlation compliance index (set by the preset personnel) is represented by the sum and average of the correlation compliance indicators of the smart products in the preset database during the historical monitoring period; when the correlation compliance index is greater than the preset correlation compliance index, the control system in the smart product will automatically trigger the risk identification algorithm (specifically referring to the feedforward deep learning network in the deep learning network in this example) to identify security risk points.

[0041] Specifically, the environmental risk point refers to the location corresponding to the security risk caused by external environmental factors. For example, if the ambient temperature of the smart product is higher than the maximum allowable ambient temperature set for the smart product (such as 40°C), it may cause the device performance to decline or be damaged (the corresponding location is the location of the ambient temperature risk point); if the ambient humidity is higher than the maximum allowable ambient humidity (such as 85%RH, relative humidity), it may cause a short circuit (the corresponding location is the location of the ambient humidity risk point); if the smoke concentration is higher than the maximum allowable smoke concentration, it may cause a fire (the corresponding location is the location of the smoke concentration risk point); the equipment failure risk point refers to the location corresponding to the damage of the circuit board of the smart product itself and the failure of the sensor; the abnormal access risk point refers to the location corresponding to the unauthorized access attempt. The accuracy and efficiency of security risk identification have been improved.

[0042] Furthermore, the specific steps for obtaining the security risk assessment value are as follows: when the obtained correlation compliance index is greater than the preset correlation compliance index, the number of abnormal access users in the risk identification process within the preset monitoring period is counted, and it is determined whether the number of abnormal access users is within the allowable range of the abnormal access number (including being equal to the maximum and minimum values ​​of the historical abnormal access number). If so, no calculation is performed, otherwise, the abnormal access score is obtained by combining the reference access number in the database; the number of environmental risk points, the number of equipment failure risk points, and the number of abnormal access risk points in the risk identification process within the preset monitoring period are counted, and the environmental risk score, the equipment failure risk score, and the abnormal access risk score are obtained by combining the reference environmental risk point number, the reference equipment failure risk point number, and the reference abnormal access risk point number in the database; according to the obtained abnormal access score (i.e., the security risk assessment value in the restricted expression), the security risk assessment value is obtained. ), environmental risk score (i.e., the security risk assessment value in the restricted expression ), equipment failure risk score (i.e., the security risk assessment value in the restricted expression ), abnormal access risk score (i.e., the security risk assessment value in the restricted expression ), correlation compliance index (i.e., the security risk assessment value in the restricted expression ) and combined with the correlation compliance index influencing factor (i.e. the restriction expression of the security risk assessment value ) to obtain the security risk assessment value.

[0043] Among them, the abnormal access score is the ratio of the number of abnormal access users to the reference access number; abnormal access users are unauthorized access users; the environmental risk score is the ratio of the number of environmental risk points to the number of reference environmental risk points; the equipment failure risk score is the ratio of the number of equipment failure risk points to the number of reference equipment failure risk points; the abnormal access risk score is the ratio of the number of abnormal access risk points to the number of reference abnormal access risk points.

[0044] The specific limiting expression of the security risk assessment value is: ; ; ; ; ; Where t is the number of the current monitoring time in the preset monitoring period, , T is the total number of current monitoring moments within the preset monitoring period, e is a natural constant, It indicates the security risk assessment value of the smart product during the preset monitoring period. Indicates that the correlation meets the indicator impact factor, Indicates the environmental risk score of the smart product during the preset monitoring period. Indicates the number of risk points in the reference environment, Indicates the number of environmental risk points of smart products within the preset monitoring period. Indicates the equipment failure risk score of the smart product within the preset monitoring period. represents the number of reference equipment failure risk points, Indicates the number of equipment failure risk points of smart products within the preset monitoring period. Indicates the abnormal access risk score of smart products within the preset monitoring period. Indicates the number of reference abnormal access risk points, Indicates the number of abnormal access risk points of smart products within the preset monitoring period. Indicates the abnormal access score of smart products within the preset monitoring period. Indicates the number of reference visits Indicates the number of abnormal users accessing the smart product during the preset monitoring period. Indicates the allowed range of abnormal access quantity. It indicates that the correlation of smart products at the current monitoring time t meets the index, Indicates that the preset correlation meets the indicator.

[0045] In this embodiment, the allowable range of the number of abnormal accesses represents the range corresponding to the maximum and minimum values ​​of the number of historical abnormal accesses in the preset database, and the reference number of accesses, the reference number of environmental risk points, the reference number of equipment failure risk points and the reference number of abnormal access risk points are respectively represented by the sum and average of the number of historical accesses, the number of historical environmental risk points, the number of historical equipment failure risk points and the number of historical abnormal access risk points in the preset database, wherein the number of historical accesses, the number of historical environmental risk points, the number of historical equipment failure risk points and the number of historical abnormal access risk points in the preset database are not 0.

[0046] Specifically, when , the correlation compliance indicator impact factor is the impact factor corresponding to the process of obtaining the security risk assessment value in the preset database, which represents the numerical value corresponding to the degree of influence of the security risk assessment impact value (representing the sum of the abnormal access score, environmental risk score, equipment failure risk score and abnormal access risk score) on the security risk assessment value. When used, the impact factor corresponding to the correlation compliance indicator can be directly obtained from the preset database, and the corresponding relationship can be a pre-set mapping relationship. For example, the correlation compliance indicator corresponding to the correlation compliance indicator in the process of obtaining the security risk assessment value and the impact factor corresponding to the correlation compliance indicator in the preset database form a mapping set, and the real-time correlation compliance indicator is input into the mapping set to obtain the corresponding correlation compliance indicator impact factor, where the mapping relationship can be one-to-one or many-to-one. In this example, the value range of the correlation compliance indicator impact factor is .

[0047] Compared with the existing technology, this example helps to more comprehensively evaluate the security risks of smart products by considering the impact of correlation compliance indicators on the security risk impact value, thereby improving the accuracy and reliability of obtaining security risk assessment values, and further achieving efficiency and accuracy in security risk assessment.

[0048] It should be added that when To simplify the analysis, we define , where It represents the security risk assessment coefficient of the smart product within the preset monitoring period. The specific restriction expression of the simplified security risk assessment value is: ,like Figure 2 As shown, it is a two-dimensional coordinate diagram of the security risk assessment value-correlation compliance index provided in the embodiment of the present application (the security risk assessment coefficient is a fixed value of 0.9 at this time), which is composed of Figure 2 It can be seen that the security risk assessment value (the smaller the better) increases with the increase of the correlation compliance index, such as Figure 3 As shown, it is a two-dimensional coordinate diagram of the security risk assessment value-security risk assessment coefficient provided in the embodiment of the present application (at this time, the correlation compliance index is a fixed value of 1.3), by Figure 3 It can be seen that the security risk assessment value decreases as the security risk assessment coefficient increases.

[0049] Among them, the environmental risk score increases with the increase in the number of environmental risk points, the equipment failure risk score increases with the increase in the number of equipment failure risk points, the abnormal access risk score increases with the increase in the number of abnormal access risk points, and the abnormal access score increases with the increase in the number of abnormal access users.

[0050] It should be noted that the environmental risk score also indirectly affects the value of the equipment failure risk score. Suppose an intelligent monitoring system is installed in an outdoor environment. When the ambient temperature rises, the heat dissipation performance of the camera may be affected, resulting in an increase in temperature. If the temperature exceeds the normal operating range of the camera (set by the manufacturer), it may cause failures such as blurred images and color distortion. In this case, the increase in the environmental risk score (due to the increase in ambient temperature) will indirectly lead to an increase in the equipment failure risk score (camera failure). By considering the above-mentioned indirect impact mechanism, it helps to ensure the stability and safety of the operation of smart products, thereby improving the effectiveness of security risk assessment of smart products, and effectively solving the problem of low effectiveness of security risk assessment of smart products in the existing technology.

[0051] Furthermore, the specific process of determining whether to perform closed-loop feedback control based on the obtained security risk assessment value is as follows: determining whether the obtained security risk assessment value is within the allowable range of the security risk assessment: if the obtained security risk assessment value is within the allowable range of the security risk assessment (excluding the value equal to the historical maximum security risk assessment value and the historical minimum security risk assessment value), then closed-loop feedback control is not performed; if the obtained security risk assessment value is not within the allowable range of the security risk assessment, then closed-loop feedback control is performed and probability level division is performed.

[0052] Among them, the allowable range of security risk assessment represents the range corresponding to the historical maximum security risk assessment value and the historical minimum security risk assessment value of the smart product in the preset database; the probability level includes the first probability, the second probability and the third probability; the first probability indicates that the obtained security risk assessment value is greater than the historical maximum security risk assessment value; the second probability indicates that the obtained security risk assessment value is less than the historical minimum security risk assessment value; the third probability indicates that the obtained security risk assessment value is equal to the historical maximum security risk assessment value or the historical minimum security risk assessment value.

[0053] In this embodiment, the maximum historical security risk assessment value of the smart product in the preset database during the historical monitoring period is input into the calculation formula of the security risk assessment value to obtain the historical maximum security risk assessment value, and the minimum historical security risk assessment value of the smart product in the preset database during the historical monitoring period is input into the calculation formula of the security risk assessment value to obtain the historical minimum security risk assessment value. This example can achieve accurate management and effective control of smart security products by classifying security risk probabilities, thereby improving the safety and reliability of the system.

[0054] Furthermore, the control effect evaluation value is obtained by the following method: M1, when the obtained security risk assessment value is within the allowable range of the security risk assessment, the risk control execution time in the closed-loop feedback control process is obtained and it is determined whether it is less than the preset risk control execution time. If so, M2 is executed, otherwise the risk control parameters are adjusted and the closed-loop feedback control is performed again. The risk control parameters include response speed and control intensity; M2, the risk control execution time score is obtained (i.e., A in the limiting expression of the control effect evaluation value), and the risk monitoring time, risk warning time and control effect response time in the closed-loop feedback control process are obtained at the same time. The risk control execution time score is the ratio of the risk control execution time to the total closed-loop feedback control time. M3 processes the obtained risk control execution time score, risk monitoring time score, risk warning time score and control effect feedback time score respectively to obtain the control effect evaluation value.

[0055] Specifically, the specific limiting expression of the control effect evaluation value is: ; ; ; ; ; In the formula, e is a natural constant, It indicates the control effect evaluation value of the smart product in the closed-loop feedback control process within the preset monitoring period. It indicates the risk control execution time score of the smart product in the closed-loop feedback control process within the preset monitoring period. It indicates the execution time of risk control in the closed-loop feedback control process of the smart product within the preset monitoring period. It indicates the total duration of closed-loop feedback control during the closed-loop feedback control process of the smart product within the preset monitoring period. Indicates the preset risk control execution time. It indicates the risk monitoring duration score of the smart product in the closed-loop feedback control process within the preset monitoring period. It indicates the risk monitoring duration of the smart product in the closed-loop feedback control process within the preset monitoring period. It indicates the risk warning duration score of the smart product in the closed-loop feedback control process within the preset monitoring period. It indicates the risk warning duration of the smart product in the closed-loop feedback control process within the preset monitoring period. It indicates the control effect feedback duration score of the smart product during the closed-loop feedback control process within the preset monitoring period. Indicates the duration of control effect feedback. It indicates the security risk assessment value of the smart product during the preset monitoring period. Indicates the allowable range of security risk assessment.

[0056] In this embodiment, the preset risk control execution duration is represented by the sum and average of the historical risk control execution times corresponding to the control systems of the smart products in the preset database, and the risk monitoring duration, risk warning duration, control effect feedback duration and the total duration of closed-loop feedback control are obtained through a time counter.

[0057] Specifically, when the risk control execution time is not less than the preset risk control execution time, the response speed is accelerated by increasing the response speed parameters of the actuator in the smart product. At the same time, the control intensity is strengthened by increasing the monitoring frequency and the number of access applications for the smart product during the preset monitoring period. The adjustment of the risk control parameters is stopped until the risk control execution time is not less than the preset risk control execution time, thereby achieving more precise management and control of the security risks of smart products, and further achieving improved accuracy and reliability in obtaining control effect evaluation values.

[0058] It needs to be understood that the control effect evaluation value (the larger the better) decreases with the increase of the risk control execution time score, the risk monitoring time score, the risk warning time score and the control effect feedback time score. Among them, the risk control execution time score increases with the increase of the risk control execution time, the risk monitoring time score increases with the increase of the risk monitoring time, the risk warning time score increases with the increase of the risk warning time, and the control effect feedback time score increases with the increase of the control effect feedback time. By considering the proportion of each link in the closed-loop feedback control process (including risk control execution, risk monitoring, risk warning and control effect feedback) relative to the total time of closed-loop feedback control, a more accurate evaluation of the closed-loop feedback control efficiency is achieved.

[0059] It should be noted that the risk control execution time score also indirectly affects the value of the control effect feedback time score. Assuming that the total time of closed-loop feedback control remains unchanged, when the risk control execution time increases, in order to maintain the balance of other links (including risk monitoring, risk warning and control effect feedback) within the total time of closed-loop feedback control, the corresponding control link after the risk control execution link will be delayed accordingly. At this time, the control effect feedback time will be reduced. By considering the above-mentioned indirect influence mechanism, it is helpful to improve the effectiveness and execution efficiency of the security risk control process, thereby realizing the improvement of the effectiveness of security risk assessment of smart products, and effectively solving the problem of low effectiveness of security risk assessment of smart products in the existing technology.

[0060] Furthermore, the specific process of judging whether the closed-loop feedback control is completed based on the obtained control effect evaluation value is as follows: judging whether the obtained control effect evaluation value is within the control effect evaluation threshold range: if the obtained control effect evaluation value is within the control effect evaluation threshold range, the closed-loop feedback control is completed; if the obtained control effect evaluation value is not within the control effect evaluation threshold range (including the maximum and minimum values ​​equal to the historical control effect evaluation values), it indicates that the effect of the closed-loop feedback control does not meet the expected risk control effect requirements, and at this time, the closed-loop feedback control gain is adjusted and the closed-loop feedback control is performed again; the closed-loop feedback control gain includes the control gain and the feedback gain.

[0061] The control effect evaluation threshold range represents the range corresponding to the maximum and minimum values ​​of the historical control effect evaluation values ​​in the preset database.

[0062] In this embodiment, it is assumed that there is an intelligent security system, and the control effect evaluation threshold is set to 90% (indicating that the control effect is considered to meet expectations when it reaches 90% or above). It is assumed that the initial control gain is K1 and the feedback gain is K2. After closed-loop feedback control is performed, the measured control effect evaluation value is 85%, which is lower than the control effect evaluation threshold of 90%.

[0063] At this time, the control gain K1 needs to be increased, for example, adjusted to 1.2 times the original value. At the same time, in order to maintain the stability of the system, the feedback gain K2 can be reduced, for example, adjusted to 0.9 times the original value. The closed-loop feedback control is performed again using the adjusted gain value, and the new control effect evaluation value is measured to be 92%, which is higher than the control effect evaluation threshold of 90%, indicating that the adjustment is effective, and the closed-loop feedback control gain is effectively adjusted, thereby achieving the improvement of the accuracy and effectiveness of the closed-loop feedback control.

[0064] like Figure 4As shown, it is a structural schematic diagram of the intelligent product risk identification and control system based on big data provided by the embodiment of the present application. The intelligent product risk identification and control system based on big data provided by the embodiment of the present application includes: a correlation compliance index acquisition module, a security risk assessment value acquisition module and a control effect assessment value acquisition module; wherein the correlation compliance index acquisition module is used to obtain the intelligent security data of the intelligent product at the current monitoring moment, and at the same time, the obtained intelligent security data is subjected to correlation analysis according to the machine learning algorithm to obtain the correlation compliance index, and whether to perform risk identification is determined based on the obtained correlation compliance index, and the correlation compliance index is used to quantify the degree of correlation between the intelligent security data; the security risk assessment value acquisition module is used to determine whether to perform risk identification. For risk identification, the result of risk identification is input into the established risk identification model to obtain the security risk assessment value. At the same time, based on the obtained security risk assessment value, it is judged whether to perform closed-loop feedback control. The risk identification model is used to obtain the security risk assessment value. The security risk assessment value is used to evaluate the probability of security risk occurring at the current monitoring moment. Closed-loop feedback control includes risk monitoring, risk early warning, risk control execution and control effect feedback. The control effect evaluation value acquisition module is used to obtain the control effect evaluation value according to the security risk control effect in the closed-loop feedback control process if closed-loop feedback control is performed. At the same time, based on the obtained control effect evaluation value, it is judged whether the closed-loop feedback control is completed. The control effect evaluation value is used to evaluate the effectiveness of the closed-loop feedback control.

[0065] like Figure 5 As shown, it is a specific flow chart of risk identification and control provided in the embodiment of the present application. By integrating multiple modules and a closed-loop feedback control mechanism, comprehensive and efficient management of security risks of smart products is achieved. The system can automatically collect and analyze security data, accurately identify risks and take corresponding control measures, thereby effectively ensuring the safety and reliability of smart products.

[0066] To summarize, the embodiment of the present application performs correlation analysis on the acquired smart security data to obtain correlation compliance indicators and determine whether to perform risk identification, then inputs the results of risk identification into the constructed risk identification model to obtain a security risk assessment value and determine whether to perform closed-loop feedback control, and finally obtains a control effect assessment value based on the security risk control effect in the closed-loop feedback control process and determines whether the closed-loop feedback control is completed, thereby improving the accuracy of obtaining the security risk assessment value, and further achieving an improvement in the effectiveness of security risk assessment of smart products, effectively solving the problem of low effectiveness of security risk assessment of smart products in the prior art.

[0067] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0068] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0069] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0070] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0071] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0072] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for intelligent product risk identification and control based on big data, characterized in that: The following steps are involved: Step 1: Obtain the smart security data of the smart product at the current monitoring moment and perform correlation analysis to obtain a correlation compliance index, and determine whether to perform risk identification based on the obtained correlation compliance index. The correlation compliance index is used to quantify the degree of correlation between smart security data; Step 2: If risk identification is performed, the result of risk identification is input into the established risk identification model to obtain a security risk assessment value, and at the same time, based on the obtained security risk assessment value, it is determined whether to perform closed-loop feedback control. The security risk assessment value is used to assess the probability of the security risk occurring at the current monitoring moment; Step three, if closed-loop feedback control is performed, a control effect evaluation value is obtained based on the security risk control effect during the closed-loop feedback control process, and at the same time, based on the obtained control effect evaluation value, it is determined whether the closed-loop feedback control is completed. The control effect evaluation value is used to evaluate the effectiveness of the closed-loop feedback control.

2. The method for identifying and controlling smart product risks based on big data as claimed in claim 1, characterized in that: The smart security data includes environmental monitoring data, equipment operation status data and user behavior data; The environmental monitoring data include ambient temperature, ambient humidity and smoke concentration; The equipment operation status data includes equipment operating power and equipment response time; The user behavior data includes the frequency of use of smart products and the number of erroneous operations of smart products; The correlation analysis includes a first correlation analysis, a second correlation analysis and a third correlation analysis; The first correlation analysis is used to reflect the correlation between the environmental monitoring data and the reference environmental monitoring data in the data control; The reference environmental monitoring data includes reference environmental temperature, reference environmental humidity and reference smoke concentration; The second correlation analysis is used to reflect the correlation between the equipment operation status data and the reference equipment operation status data in the database; The reference device operating status data includes the reference device operating power and the reference device response time; The third correlation analysis is used to reflect the correlation between the user behavior data and the reference user behavior data in the database; The reference user behavior data includes reference smart product usage frequency and a maximum allowable number of smart product misoperations.

3. The method for identifying and controlling smart product risks based on big data as claimed in claim 1, characterized in that: The correlation compliance index is obtained by the following method: When the ambient temperature is greater than the reference ambient temperature, the ambient humidity is greater than the reference ambient humidity, and the smoke concentration is greater than the reference smoke concentration, the ambient temperature score, ambient humidity score, and smoke concentration score of the smart product at the current monitoring time are obtained and added to obtain a first correlation compliance index; The device working power score and the device response time score of the smart product at the current monitoring time are obtained and added to obtain a second correlation compliance index. At the same time, the usage frequency score and the number of erroneous operation scores of the smart product at the current monitoring time are obtained and added to obtain a third correlation compliance index. Obtaining a correlation compliance index according to the obtained first correlation compliance index, the second correlation compliance index and the third correlation compliance index and in combination with corresponding correlation compliance index weight factors; The ambient temperature fraction is the ratio of the ambient temperature to the maximum allowable ambient temperature; The ambient humidity fraction is the ratio of the ambient humidity to the maximum allowable ambient humidity; The smoke concentration fraction is the ratio of the smoke concentration to the maximum allowable smoke concentration; The device operating power score is the ratio of the absolute value of the difference between the device operating power and the reference device operating power to the reference device operating power; The device response time score is the ratio of the absolute value of the difference between the device response time and the reference device response time to the reference device response time; The usage frequency score is the ratio of the usage frequency of the smart product to the usage frequency of the reference smart product; The misoperation times score is the ratio of the number of misoperations of the smart product to the maximum allowable number of misoperations of the smart product; The relevance compliance index weight factor includes a first relevance compliance index weight factor, a second relevance compliance index weight factor and a third relevance compliance index weight factor.

4. The method for identifying and controlling smart product risks based on big data as claimed in claim 1, characterized in that: The specific process of determining whether to perform risk identification based on the obtained correlation compliance index is as follows: Determine whether the obtained correlation compliance index is greater than the preset correlation compliance index: If the obtained correlation compliance index is greater than the preset correlation compliance index, a security risk point is obtained, wherein the security risk point includes an environmental risk point, an equipment failure risk point, and an abnormal access risk point; If the obtained correlation compliance index is not greater than the preset correlation compliance index, risk identification will not be performed and the working status of the smart product will continue to be monitored.

5. The method for identifying and controlling smart product risks based on big data as claimed in claim 1, characterized in that: The specific steps for obtaining the security risk assessment value are as follows: Count the number of abnormal access users in the risk identification process during the preset monitoring period, and determine whether the number of abnormal access users is within the allowed range of abnormal access. If so, no calculation is performed. Otherwise, the abnormal access score is obtained by combining the reference access number in the database. Count the number of environmental risk points, equipment failure risk points and abnormal access risk points in the risk identification process during the preset monitoring period, and combine the reference number of environmental risk points, reference number of equipment failure risk points and reference number of abnormal access risk points in the database to obtain the environmental risk score, equipment failure risk score and abnormal access risk score; The security risk assessment value is obtained based on the acquired abnormal access score, environmental risk score, equipment failure risk score, abnormal access risk score, correlation compliance index and the correlation compliance index influencing factor; The abnormal access score is the ratio of the number of abnormal access users to the number of accesses; The abnormal access user is an unauthorized access user; The environmental risk score is the ratio of the number of environmental risk points to the number of reference environmental risk points; The equipment failure risk score is the ratio of the number of equipment failure risk points to the number of reference equipment failure risk points; The abnormal access risk score is the ratio of the number of abnormal access risk points to the number of reference abnormal access risk points.

6. The method for identifying and controlling smart product risks based on big data as claimed in claim 5, characterized in that: The specific limiting expression of the security risk assessment value is: ； ； ； ； ； Where t is the number of the current monitoring time in the preset monitoring period, , T is the total number of current monitoring moments within the preset monitoring period, e is a natural constant, It indicates the security risk assessment value of the smart product during the preset monitoring period. Indicates that the correlation meets the indicator impact factor, Indicates the environmental risk score of the smart product during the preset monitoring period. Indicates the number of risk points in the reference environment, Indicates the number of environmental risk points of smart products within the preset monitoring period. Indicates the equipment failure risk score of the smart product within the preset monitoring period. represents the number of reference equipment failure risk points, Indicates the number of equipment failure risk points of smart products within the preset monitoring period. Indicates the abnormal access risk score of smart products within the preset monitoring period. Indicates the number of reference abnormal access risk points, Indicates the number of abnormal access risk points of smart products within the preset monitoring period. Indicates the abnormal access score of smart products within the preset monitoring period. Indicates the number of reference visits Indicates the number of abnormal users accessing the smart product during the preset monitoring period. Indicates the allowed range of abnormal access quantity. It indicates that the correlation of smart products at the current monitoring time t meets the index, Indicates that the preset correlation meets the indicator.

7. The method for identifying and controlling smart product risks based on big data as claimed in claim 1, characterized in that: The specific process of judging whether to perform closed-loop feedback control based on the acquired security risk assessment value is as follows: Determine whether the obtained security risk assessment value is within the allowable range of security risk assessment: If the security risk assessment value obtained is within the permissible range of the security risk assessment, no closed-loop feedback control will be performed; If the security risk assessment value obtained is not within the permissible range of the security risk assessment, closed-loop feedback control is performed and probability level division is performed; The probability levels include a first probability, a second probability, and a third probability.

8. The method for identifying and controlling smart product risks based on big data as claimed in claim 1, characterized in that: The control effect evaluation value is obtained by the following method: M1, obtain the risk control execution time in the closed-loop feedback control process and determine whether it is less than the preset risk control execution time. If so, execute M2; otherwise, adjust the risk control parameters and re-perform closed-loop feedback control. The risk control parameters include response speed and control intensity; M2, obtain the risk control execution time score, and at the same time obtain the risk monitoring time, risk warning time and control effect feedback time in the closed-loop feedback control process and perform ratio calculations with the total closed-loop feedback control time to obtain the risk monitoring time score, risk warning time score and control effect feedback time score, the risk control execution time score being the ratio of the risk control execution time to the total closed-loop feedback control time; M3, processes the acquired risk control execution time score, risk monitoring time score, risk warning time score, and control effect feedback time score respectively to obtain the control effect evaluation value.

9. The method for identifying and controlling smart product risks based on big data as claimed in claim 1, characterized in that: The specific process of judging whether the closed-loop feedback control is completed based on the acquired control effect evaluation value is as follows: Determine whether the obtained control effect evaluation value is within the control effect evaluation threshold range: If the obtained control effect evaluation value is within the control effect evaluation threshold range, the closed-loop feedback control is completed; If the obtained control effect evaluation value is not within the control effect evaluation threshold range, the closed-loop feedback control gain is adjusted and the closed-loop feedback control is performed again; The closed-loop feedback control gain includes a control gain and a feedback gain.

10. Intelligent product risk identification and control system based on big data, characterized by: include: Relevance compliance index acquisition module, security risk assessment value acquisition module and control effect assessment value acquisition module; The correlation compliance index acquisition module is used to acquire the smart security data of the smart product at the current monitoring moment and perform correlation analysis to acquire the correlation compliance index, and determine whether to perform risk identification based on the acquired correlation compliance index. The correlation compliance index is used to quantify the degree of correlation between the smart security data. The security risk assessment value acquisition module is used to input the result of risk identification into the established risk identification model to obtain a security risk assessment value if risk identification is performed, and to determine whether to perform closed-loop feedback control based on the obtained security risk assessment value, wherein the security risk assessment value is used to evaluate the probability of occurrence of the security risk at the current monitoring moment; The control effect evaluation value acquisition module is used to obtain a control effect evaluation value according to the security risk control effect during the closed-loop feedback control process if closed-loop feedback control is performed, and at the same time, judge whether the closed-loop feedback control is completed based on the obtained control effect evaluation value. The control effect evaluation value is used to evaluate the effectiveness of the closed-loop feedback control.

Images