Multi-role grading access control method, system and device for segment sliding electric door

By cleaning, normalizing, and analyzing user access control data in different time periods, and combining this with a risk assessment mechanism to dynamically adjust security responses, the problem of insufficient identification of abnormal behavior in existing access control systems has been solved, enabling precise monitoring and management of user behavior and risks.

CN121075018APending Publication Date: 2025-12-05WUXI JIANGUO INTELLIGENT TECH CO LTD

Patent Information

Application Number
CN202510950576.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-07-10
Publication Date
2025-12-05

AI Technical Summary

Technical Problem

Existing access control systems lack dynamic perception and intelligent analysis of abnormal passage behavior, resulting in security blind spots in high-risk areas.

Method used

By collecting user access control data, performing data cleaning and normalization, segmenting user behavior data by time period, extracting statistical features and identifying abnormal patterns, assessing the magnitude and trend of risk fluctuations in real time, dynamically adjusting security response measures, and generating strategy optimization suggestions.

Benefits of technology

It enables refined quantitative expression of user behavior and accurate identification of risk status, dynamically identifies the evolution trend of abnormal behavior, and improves the security and management efficiency of the access control system.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121075018A_ABST
    Figure CN121075018A_ABST
Patent Text Reader

Abstract

The invention discloses a multi-role grading access control method, system and device for a segment sliding electric door, and relates to the technical field of access control. The multi-role hierarchical access control method, system and device for the segment sliding electric door comprises the following steps: S1, collecting user access control data in the user identity verification and passing operation process, and carrying out data cleaning and normalization processing on the user access control data; s2, slicing the user behavior data in different time periods, extracting statistical characteristics, identifying abnormal modes, and carrying out quantitative analysis on the behavior state of the user; s3, collecting access control data in real time to construct a behavior sequence, evaluating the risk fluctuation amplitude and trend, judging the risk level and carrying out risk response; and S4, carrying out quantitative evaluation on risk response effects in different operation periods, dynamically adjusting safety response measures and regularly generating strategy optimization suggestions. The problem that in the prior art, traffic abnormal behaviors cannot be effectively recognized, and consequently a safety blind area exists is solved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of access control technology, specifically to a multi-role hierarchical access control method, system, and device for a sliding electric door. Background Technology

[0002] With the rapid development of artificial intelligence technology in the production field, especially its mature application in behavior recognition, risk assessment and strategy optimization, the evolution of access control systems towards artificial intelligence systems in the production field has become a key trend.

[0003] For example, an access control controller with announcement number CN105069878A is connected to an external server, an RFID card reader, and a sensor. The access control controller includes a microcontroller circuit, a switching circuit, a first level conversion circuit, a second level conversion circuit, a serial data communication circuit, a sound-generating circuit, a connection circuit, a drive unit, an indicator circuit, and a relay circuit. The microcontroller circuit is electrically connected to the switching circuit, the sound-generating circuit, the connection circuit, the RFID card reader, and the drive unit. The drive unit is electrically connected to the indicator circuit and the relay circuit. The microcontroller circuit includes a working mode and a sleep mode. When the sensor detects an object approaching, it wakes the microcontroller circuit from the sleep mode, and the sound-generating circuit emits a first alert sound.

[0004] For example, an access control controller with announcement number CN102222379A includes an electric door lock, an access control controller, and a control terminal. The electric door lock is equipped with a limit switch, which is connected to the control terminal. The access control controller is an integrated IC card identification controller using a contactless read / write chip. The control terminal is a remote control terminal using an anti-interference communication chip, and it communicates with the central control center and various systems of the access control controller using encrypted protocols. The access control controller uses a 32-bit embedded processor with a Cortex-M3 core. The access control controller uses a moisture-proof PCB board and has an anti-UV coating on its outer shell.

[0005] However, most current access control systems still rely on static identity authentication and preset permission management, lacking dynamic perception and intelligent analysis of the entire passage process. This makes it difficult to identify abnormal behaviors such as frequent failed verifications, unauthorized attempts, abnormal lingering, and abnormal door switching in a timely manner, resulting in security blind spots in high-risk areas during actual operation.

[0006] Therefore, in order to address the above problems, there is an urgent need for a multi-role hierarchical access control method, system, and device for sliding electric doors. Summary of the Invention

[0007] Technical problems to be solved

[0008] To address the shortcomings of existing technologies, this invention provides a multi-role hierarchical access control method, system, and device for sectional electric sliding doors, which solves the problem of existing technologies being unable to effectively identify abnormal passage behavior, resulting in security blind spots.

[0009] Technical solution

[0010] To achieve the above objectives, the present invention provides the following technical solution: a multi-role hierarchical access control method, system, and device for a sliding electric door, comprising the following steps: S1, collecting user access control data during user authentication and access operations, and performing data cleaning and normalization on the user access control data; S2, segmenting user behavior data by time period, extracting statistical features and identifying abnormal patterns, and quantitatively analyzing the user's behavioral state; S3, collecting access control data in real time to construct a behavior sequence, assessing the magnitude and trend of risk fluctuations, determining the risk level, and implementing risk response; S4, quantitatively evaluating the risk response effect within different operation cycles, dynamically adjusting security response measures, and periodically generating strategy optimization suggestions.

[0011] Furthermore, the specific steps for collecting user access control data during user authentication and access operations are as follows: The access control system collects user access control data during user authentication and access operations. This data includes user ID, role level, door segment number, verification time, historical maximum verification time, number of attempts for the current door segment, number of failures for the current door segment, average operation interval, dwell time, historical maximum dwell time, and door segment switching frequency. Specifically, the user ID is obtained by acquiring identity information through the access card reader terminal and parsing the record by the identity authentication module. After identity authentication, the system calls the permission database interface to obtain the role level. Finally, the current access target is recorded through communication with the sliding electric door controller to obtain the user ID. The system generates various data points, including: gate segment number; verification time and historical maximum verification time by recording the difference between the authentication request time and the gate access response time; average operation interval by comparing the timestamps of consecutive operations by the same user and calculating the average time difference of multiple operations; number of attempts for the current gate segment by counting the number of all attempts made by the user within a fixed time window; number of failures for the current gate segment by filtering the number of records of verification failures among the above attempts; dwell time and historical maximum dwell time by identifying and matching the user's passage time at the entrance and exit gate segments; and gate segment switching frequency by analyzing the changes in the gate segment number of consecutive attempts by the user and counting the number of switching per unit time.

[0012] Furthermore, the specific steps for cleaning and normalizing user access control data are as follows: A multi-stage processing flow is used to achieve data cleaning and normalization: a sliding window averaging method combined with median absolute deviation detection is used to identify and remove outliers and abrupt changes in verification time and dwell time; for missing access records caused by communication delays and local packet loss, time-series forward filling and local regression interpolation methods are used to fill in the number of failures and door segment switching frequency; through a unified association mapping of timestamps with user IDs and door segment numbers, time alignment and behavior synchronization of user identity, access behavior, and role permissions within a unified behavior cycle are achieved; an exponential sliding filter combined with a bilateral filter is used to smooth and reduce noise in operation frequency and number of attempts, preserving behavioral trend characteristics; for different dimensions of data characteristics, a joint normalization process is constructed using standard deviation standardization and Box-Cox transform to standardize user access control data and unify the scale range of user access control data.

[0013] Furthermore, the specific steps for segmenting user behavior data by time period, extracting statistical features, and identifying abnormal patterns are as follows: After sorting the input data by user ID and timestamp, the data is divided into continuous time segments using a sliding time window of fixed duration; statistical features are calculated for access control records within each behavior segment to form corresponding behavior feature vectors; conditional judgments are applied to the behavior feature vectors: behaviors with high door segment switching frequency and more than 3 failures per unit time are marked as trial access; repeated failed verifications within the same door segment in a short period of time exceeding the set repetition threshold are marked as brute-force attempts; behaviors with a dwell time exceeding twice the user's historical average are marked as dwell anomalies; behaviors attempting to enter door segments with a higher permission level than the user's own role and failing verification are marked as unauthorized behavior; a sliding window scrolling statistical method is used to assist in detecting abnormal behaviors, and all anomalies are recorded in a structured form.

[0014] Further, the specific steps for quantitatively analyzing user behavior are as follows: Record the number of unauthorized access attempts, obtain user access control data, and quantitatively analyze user behavior: Divide the current gate segment failure count by the current gate segment attempt count plus a minimum term, divide the verification time by the historical maximum verification time, take the reciprocal of the average operation interval, divide the dwell time by the historical maximum dwell time, add the above four results, and then add the gate segment switching frequency and the number of unauthorized access attempts; the sum is the first part; subtract the user level from the maximum role level and divide by the maximum role level, add 1 to this ratio as the second part; multiply the behavior weight, the first part, and the second part to obtain the user behavior status evaluation value; after calculating the obtained user behavior status evaluation value... The system compares user behavior status assessment values ​​with behavior status thresholds in real time to determine the user's behavior status. These thresholds include a primary behavior status threshold and a secondary behavior status threshold: when the user behavior status assessment value is less than or equal to the primary threshold, the user behavior is considered normal and requires no further intervention; when the user behavior status assessment value is greater than the primary threshold but less than the secondary threshold, the user behavior is considered suspicious, generating a warning notification to security management personnel, encrypting the verification process, and extending the verification time interval; when the user behavior status assessment value is greater than the secondary threshold, the user behavior is considered high-risk, automatically triggering a blocking policy, suspending the user's access control permissions, recording an early warning log, and triggering an alarm in conjunction with the security response module.

[0015] Furthermore, the specific steps for constructing a behavior sequence from real-time access control data, assessing risk fluctuations and trends, determining risk levels, and implementing risk responses are as follows: Real-time collection of user access control data; calculation of the latest user behavior status assessment value; construction of a continuous behavior sequence indexed by user ID and timestamp; division of operation cycles into minutes, extraction of key behavioral indicator changes within adjacent cycles, including changes in the number of current gate failures, dwell time, and gate switching frequency; extraction of the historical maximum user behavior status assessment value; acquisition of user access control data, user behavior status assessment value, changes in the number of current gate failures, dwell time, and gate switching frequency; assessment of the user risk status change trend: taking the absolute value of the change in the number of current gate failures, dividing the absolute value of the change in dwell time by the historical maximum dwell time plus 1, and taking the logarithm of the absolute value of the change in gate switching frequency plus 1; and combining the above three results. The process is as follows: First, the user behavior status assessment value is summed; second, the user behavior status assessment value is divided by the historical maximum user behavior status assessment value plus 1, and the ratio is added by 1; third, the first part is multiplied by the second part to obtain the risk fluctuation assessment value; after obtaining the risk fluctuation assessment value, it is compared with the preset risk trend threshold in real time to determine the user's risk trend level; the risk trend threshold includes a low fluctuation threshold and a high fluctuation threshold: when the risk fluctuation assessment value is less than or equal to the low fluctuation threshold, it is determined to be a stable state, and only behavior records are saved without intervention; when the risk fluctuation assessment value is greater than the low fluctuation threshold but less than the high fluctuation threshold, it is determined to be a medium fluctuation state, and risk change reminders are automatically pushed, while verification redundancy is increased to strengthen security verification; when the risk fluctuation assessment value is greater than or equal to the high fluctuation threshold, it is determined to be a high fluctuation state, and an early warning mechanism is immediately triggered, multi-factor authentication and permission freezing measures are initiated, and the security response module is linked to record alarms and early warning logs.

[0016] Furthermore, the specific steps for quantitatively evaluating the effectiveness of risk response within different operational cycles are as follows: After the risk response is implemented, continuously collect user behavior status assessment values ​​and risk fluctuation assessment values, track the changing trends of user behavior risk and fluctuation in subsequent operational cycles, and form a continuous data sequence; obtain the user behavior status assessment values ​​for the t-th operational cycle, the user behavior status assessment values ​​for the (t+1)-th operational cycle, the risk fluctuation assessment values ​​for the t-th operational cycle, and the risk fluctuation assessment values ​​for the (t+1)-th operational cycle, and quantitatively evaluate the effectiveness of the risk response measures: subtract the user behavior status assessment value for the (t+1)-th operational cycle from the user behavior status assessment value for the t-th operational cycle, multiply by the response weight, divide this product by 1, and add the user behavior status assessment value for the t-th operational cycle; the ratio is used as the first part; subtract the risk fluctuation assessment value for the (t+1)-th operational cycle from the risk fluctuation assessment value for the t-th operational cycle, multiply by 1, subtract the response weight, divide this product by 1, and add the user behavior status assessment value for the t-th operational cycle; the ratio is used as the second part; add the first part to the second part to obtain the risk response performance value.

[0017] Furthermore, the specific steps for dynamically adjusting security response measures and periodically generating strategy optimization suggestions are as follows: After calculating the risk response performance value, compare the risk response performance value with the risk response threshold in real time: If the risk response performance value is greater than the risk response threshold, the risk is effectively suppressed, the security response intensity is automatically reduced, the frequency of multi-factor authentication is reduced, and the verification time interval is extended; if the risk response performance value is less than or equal to the risk response threshold, the risk control is insufficient, intervention measures are strengthened, a cross-segment access verification mechanism is enabled and repeated verification is required between different segments, the user's access permission to high-privilege segments is temporarily frozen, and manual review is required; the risk response performance value obtained each time and its corresponding user action are recorded. The system provides a unified, structured storage for status assessment values, risk fluctuation assessment values, strategy adjustments, and execution results, forming a historical database of risk response effectiveness. It periodically extracts labeled samples from this database to construct a training dataset for continuous optimization of user behavior discrimination logic, risk status assessment mechanisms, and response strategy parameters. A priority matrix is ​​generated based on role level, gate number, and strategy combination effects to recommend optimal strategies. Regular strategy optimization suggestions are output, listing adjustment items, affected user groups, associated risk types, and expected intervention effects. Simultaneously, a response strategy change log is generated, recording parameter changes, change time, basis, and approval status to ensure that strategy versions are controllable, auditable, and rollbackable.

[0018] The second aspect of this invention provides a multi-role hierarchical access control system for a sliding electric door, comprising: a user access control data acquisition and preprocessing module, a user behavior quantitative analysis and behavior rating module, an access control risk fluctuation identification and early warning module, and an access control risk response optimization and strategy iteration module. The user access control data acquisition and preprocessing module is used to collect user access control data during user authentication and access operations, and to perform data cleaning and normalization on the user access control data. The user behavior quantitative analysis and behavior rating module is used to slice user behavior data by time period, extract statistical features, identify abnormal patterns, and perform quantitative analysis of user behavior status. The access control risk fluctuation identification and early warning module is used to collect access control data in real time to construct behavior sequences, assess the magnitude and trend of risk fluctuations, determine the risk level, and perform risk response. The access control risk response optimization and strategy iteration module is used to quantitatively evaluate the risk response effect within different operating cycles, dynamically adjust security response measures, and periodically generate strategy optimization suggestions.

[0019] The third aspect of this invention provides a multi-role hierarchical access control device for a sliding electric door, comprising: an identity recognizer, a behavior evaluator, a risk responder, and a strategy optimizer, wherein: the identity recognizer is used to collect and parse user identity, role level, and access behavior information to achieve real-time perception and structured processing of access control operation data; the behavior evaluator is used to perform time-sharing analysis of user access control behavior data, extract key features and identify abnormal patterns, and quantitatively assess the behavior risk level by combining role level and behavior weight; the risk responder is used to continuously analyze user behavior status and risk fluctuation changes, dynamically evaluate the security control effect, and execute corresponding intervention measures to adjust access permissions and verification strength; the strategy optimizer is used to perform strategy adaptation analysis and weight fine-tuning based on historical response effects and role scenario data, generate the optimal response strategy, and maintain auditable rule iteration records.

[0020] Beneficial effects

[0021] The present invention has the following beneficial effects:

[0022] (1) The multi-role hierarchical access control method, system, and device for this section of electric sliding doors, by constructing a multi-dimensional behavioral parameter system, comprehensively analyzes the number of failures, verification time, dwell time, and role level of users during the passage process, forming quantifiable behavioral status indicators to reflect the passage behavior characteristics of users with different roles. Combining the number of failures in the current door segment, the number of attempts in the current door segment, the verification time, the historical maximum verification time, the average operation interval, and the dwell time, a unified user behavior status evaluation value is constructed to achieve a refined quantitative expression of user behavior status.

[0023] (2) The multi-role hierarchical access control method, system, and device for this section of electric sliding doors, by analyzing the changes in key behavioral variables within a continuous operation cycle and combining them with user behavior status assessment values, establishes a risk fluctuation trend assessment mechanism, which can dynamically identify the phased changes in user risk status. Based on the changes in the number of current door failures, the changes in dwell time, the changes in door switching frequency, and the historical maximum dwell time, combined with the current and historical maximum user behavior status assessment values, a risk fluctuation assessment value is quantitatively generated, realizing accurate perception and process monitoring of the evolution trend of abnormal behavior.

[0024] (3) The multi-role hierarchical access control method, system, and device for this section of electric sliding doors establishes a feedback analysis mechanism for risk response effectiveness by correlating user behavior status evaluation values ​​and risk fluctuation evaluation values ​​in the operation cycle before and after the response. This mechanism can quantitatively evaluate behavioral changes before and after the implementation of safety response measures, achieving closed-loop optimization of risk control strategies. In subsequent cycles, by comparing evaluation indicators within consecutive cycles and combining them with response weights, risk response performance values ​​are generated, providing data support for the adjustment and evolution of response strategies and enhancing the system's dynamic adaptability and control accuracy.

[0025] (4) The multi-role hierarchical access control method, system and device of the sliding electric door, by combining historical behavior data and response execution records, dynamically adjusts the user behavior evaluation logic and security response parameters, improves the adaptability of the strategy to different role levels, door segment numbers and behavior patterns, enhances the flexibility and pertinence of the strategy response, provides data support for the optimization selection of security response strategies, and promotes the adaptive evolution of the access control mechanism.

[0026] Of course, any product implementing this invention does not necessarily need to achieve all of the advantages described above at the same time. Attached Figure Description

[0027] Figure 1 Flowchart of a multi-role hierarchical access control method for a sliding electric door;

[0028] Figure 2 This is a structural diagram of a multi-role hierarchical access control system for a sliding electric door;

[0029] Figure 3 This is a distribution chart of risk volatility intensity under the operating cycle;

[0030] Figure 4 A line graph showing the dynamic evolution of risk response performance value in relation to risk volatility assessment value. Detailed Implementation

[0031] 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] Please see Figures 1-4 This invention provides a technical solution: a multi-role hierarchical access control method, system, and device for a sliding electric door, comprising the following steps: S1, collecting user access control data during user authentication and access operations, and performing data cleaning and normalization on the user access control data; S2, segmenting user behavior data by time period, extracting statistical features and identifying abnormal patterns, and quantitatively analyzing the user's behavioral state; S3, collecting access control data in real time to construct a behavior sequence, assessing the magnitude and trend of risk fluctuations, determining the risk level, and implementing risk response; S4, quantitatively evaluating the risk response effect within different operation cycles, dynamically adjusting security response measures, and periodically generating strategy optimization suggestions.

[0033] Specifically, the steps for collecting user access control data during user authentication and access operations are as follows: The access control system collects user access control data during authentication and access operations. This data includes user ID, role level, door segment number, verification time, historical maximum verification time, number of attempts for the current door segment, number of failures for the current door segment, average operation interval, dwell time, historical maximum dwell time, and door segment switching frequency. Specifically, the user ID is obtained by reading the access card at the access control card terminal and parsing the record by the identity authentication module. After authentication, the system calls the permission database interface to obtain the role level. Finally, the current access target is recorded by communicating with the sliding electric door controller to obtain the door access control data. The system uses the following methods: segment number; verification time and historical maximum verification time are obtained by recording the difference between the authentication request time and the gate access response time; average operation interval is obtained by comparing the timestamps of consecutive operations by the same user and then calculating the average of the time differences of multiple operations; number of attempts for the current gate segment is obtained by counting the number of all passage attempts by the user within a fixed time window; number of failures for the current gate segment is obtained by filtering the number of verification failure records among the above attempts; dwell time and historical maximum dwell time are obtained by identifying and matching the passage time of the user at the entrance and exit gate segments; and gate segment switching frequency is obtained by analyzing the changes in the gate segment number of consecutive attempts by the user and counting the number of switching per unit time.

[0034] This implementation plan comprehensively covers the user authentication and access control process through the access control system, collecting user ID, role level, door segment number, authentication time, historical maximum authentication time, number of attempts for the current door segment, number of failures for the current door segment, average operation interval, dwell time, historical maximum dwell time, and door segment switching frequency, reflecting user access characteristics from multiple dimensions. From identity recognition to access efficiency and operational behavior, it covers all aspects, providing accurate and detailed data support for access control management. This helps analyze user access patterns, identify abnormal behavior, optimize access control strategies, and improve the security and management efficiency of the access control system.

[0035] Specifically, the steps for cleaning and normalizing user access control data are as follows: To ensure data accuracy and usability, a rigorous and scientific multi-stage processing procedure is adopted to achieve data cleaning and normalization. First, regarding verification time and dwell time, since they are susceptible to outliers caused by sudden factors, a sliding window averaging method combined with median absolute deviation detection is used. This method can accurately identify and remove outliers and abrupt changes, effectively improving data reliability. Second, considering that communication delays and partial packet loss may cause missing access records, short-term missing values ​​for the current gate segment failure count and gate segment switching frequency are filled in using time-series forward filling and local regression interpolation methods to ensure data integrity. Third, given that user access control data covers user identity, access behavior, and role permissions, to achieve effective correlation and analysis between data, timestamps are compared with user IDs and gate segment numbers. A unified association mapping is used to align and synchronize different types of data within a unified behavioral cycle, constructing a complete data chain. Then, to avoid data noise interfering with the analysis results, a combination of exponential sliding filter and bilateral filter is used for smoothing and noise reduction, removing noise while fully preserving behavioral trend characteristics. Finally, due to the differences in the characteristics of various dimensions of user access control data, a joint normalization process is constructed using standard deviation standardization and Box-Cox transform to facilitate subsequent data analysis and model building. This process comprehensively standardizes the user access control data, unifying the scale range of the data and making it more suitable for analysis needs.

[0036] This implementation plan comprehensively ensures data quality through a multi-stage process. Sliding windows and median absolute deviation detection are used to remove outliers in verification time and dwell time, ensuring data authenticity and reliability. Time series filling and regression interpolation are used to fill in missing values ​​for current gate segment failure counts and gate segment switching frequencies, maintaining data integrity. A mapping between timestamps, user IDs, and gate segment numbers achieves time alignment and behavioral synchronization of heterogeneous data, facilitating comprehensive analysis. Exponential sliding filters and bilateral filters are combined to smooth operation frequency and attempt counts, preserving behavioral trend characteristics. Standard deviation standardization and Box-Cox transform are used to normalize all data, unifying the scale range. Ultimately, this provides a high-quality, standardized data foundation for access control management analysis, abnormal behavior identification, and strategy optimization, effectively improving the data application value and management decision accuracy of the access control system.

[0037] Specifically, the steps for segmenting user behavior data by time period, extracting statistical features, and identifying abnormal patterns are as follows: To accurately analyze user behavior patterns, the input data is first strictly sorted according to user ID and timestamp to ensure the accuracy of the data's temporal sequence. Based on this, a fixed-duration sliding time window is used to systematically divide the data. This division method is like cutting a continuous stream of user behavior data into closely connected, time-series-continuous behavior segments, laying the foundation for subsequent refined analysis. For each generated behavior segment, the access control records are deeply mined. By calculating various statistical features, the originally fragmented access control data is transformed into representative behavioral feature vectors, allowing the behavioral features to be quantified. After obtaining the behavioral feature vector, rigorous conditional judgment logic is further executed: when an abnormally high frequency of door segment switching is detected within a unit of time, and the number of failures exceeds 3, combined with the actual risk scenarios of access control management, this behavior is highly likely that the user is testing the passage path, and therefore it is marked as a trial passage; if within the same door segment, the number of times a user repeatedly fails verification within a short period of time exceeds a pre-set repetition threshold, this frequent and abnormal verification failure behavior is highly likely to be an attempt to brute-force the access control, and is therefore marked as a brute-force attempt; for behavior with a dwell time exceeding twice the user's historical average duration, considering the normal passage time pattern, this behavior significantly deviates from the norm and is judged as an abnormal dwell time; when it is detected that a user attempts to enter a door segment with a permission level higher than their own role, and the verification result fails, according to the access control permission management rules, this behavior is an unauthorized behavior and needs to be marked and warned in a timely manner. In addition, to improve the accuracy and real-time performance of abnormal behavior detection, a sliding window scrolling statistical method is used to dynamically monitor behavioral data to ensure that no abnormal behavior is missed. All identified abnormal behaviors are recorded in detail in a structured format, facilitating subsequent tracing and analysis, thereby providing reliable data support for access control security management.

[0038] This implementation plan achieves precise behavioral analysis and risk warning through an orderly data processing workflow. First, data is sorted by user ID and timestamp, and then segmented using a sliding window to cut continuous behaviors into easily analyzable segments. Next, statistical features are calculated to form a quantified behavioral feature vector, providing a data foundation for anomaly detection. By setting multi-dimensional conditional judgment rules such as door segment switching frequency, number of failures, dwell time, and permission level, various abnormal patterns such as trial access, forced entry attempts, abnormal dwelling, and unauthorized behavior can be accurately identified. Simultaneously, the sliding window scrolling statistics assist in detection, effectively improving the timeliness and accuracy of abnormal behavior discovery. Finally, anomalies are recorded in a structured form for easy traceability and analysis, providing reliable and efficient data support for the security management, risk prevention and control, and strategy optimization of the access control system.

[0039] Specifically, the steps for quantitatively analyzing user behavior are as follows: Record the number of unauthorized access attempts, obtain user access control data, and quantitatively analyze user behavior: Divide the number of failed access attempts at the current access segment by the number of attempts at the current access segment plus a minimum term; divide the verification time by the historical maximum verification time; take the reciprocal of the average operation interval; divide the dwell time by the historical maximum dwell time; add the above four results together, then add the access segment switching frequency and the number of unauthorized access attempts, and use the sum as the first part; subtract the user level from the maximum role level and divide by the maximum role level, then add 1 to this ratio as the second part; multiply the behavior weight, the first part, and the second part to obtain the user behavior status evaluation value; The algorithm uses logistic regression to train and... Parameter optimization yields behavior weights, with values ​​ranging from [0, 1]. After calculating the user behavior status assessment value, it is compared in real-time with the behavior status threshold to determine the user behavior status. The behavior status threshold includes a primary behavior status threshold and a secondary behavior status threshold: when the user behavior status assessment value is less than or equal to the primary behavior status threshold, the user behavior is normal and requires no additional intervention; when the user behavior status assessment value is greater than the primary behavior status threshold but less than the secondary behavior status threshold, the user behavior is suspicious, a warning is generated to notify security management personnel, the verification process is encrypted, and the verification time interval is extended; when the user behavior status assessment value is greater than the secondary behavior status threshold, the user behavior is high-risk, an automatic blocking policy is triggered, the user's access control permissions are suspended, a warning log is recorded, and the security response module is activated to issue an alarm.

[0040] The specific formula for calculating the user behavior status evaluation value is as follows:

[0041]

[0042] In the formula, R represents the user behavior state evaluation value, N represents the number of attempts for the current gate segment, F represents the number of failures for the current gate segment, ∈ represents the minima, and T represents the verification time. max This indicates the longest verification time in history. L represents the average operation interval, and L represents the dwell time. max Indicates the maximum historical dwell time, C represents the gate switching frequency, U represents the number of unauthorized access attempts, and D represents the maximum historical dwell time. g Indicates user level, D max α represents the maximum role level, and α represents the behavior weight.

[0043] This implementation plan achieves accurate assessment and efficient control of user behavior status through multi-dimensional data fusion and a hierarchical judgment mechanism. First, by comprehensively considering the number of failed access attempts, the number of attempts, verification time, average operation interval, dwell time, access switching frequency, and the number of unauthorized access attempts, combined with user role level information, a comprehensive quantitative assessment model is constructed, transforming user behavior into measurable numerical indicators. Second, by setting primary and secondary behavior status thresholds, a tiered risk judgment system is established, enabling rapid differentiation between normal, suspicious, and high-risk user behavior. Finally, differentiated control strategies are automatically matched to different risk levels, including no intervention in normal states, warning notifications and enhanced verification processes in suspicious states, and permission blocking, logging, and security alarms in high-risk states, effectively improving the access control system's risk prevention and control capabilities and ensuring access control security and management efficiency.

[0044] Specifically, the steps for real-time acquisition of access control data to construct behavioral sequences, assess risk fluctuations and trends, determine risk levels, and implement risk responses are as follows: Real-time acquisition of user access control data; calculation of the latest user behavior status assessment value; construction of a continuous behavior sequence indexed by user ID and timestamp; division of operation cycles into minutes, extraction of key behavioral indicator changes within adjacent cycles, including changes in the number of current gate failures, dwell time, and gate switching frequency; extraction of the historical maximum user behavior status assessment value; acquisition of user access control data, user behavior status assessment value, changes in the number of current gate failures, dwell time, and gate switching frequency; assessment of the user risk status change trend: taking the absolute value of the change in the number of current gate failures, dividing the absolute value of the change in dwell time by the historical maximum dwell time plus 1, and taking the logarithm after adding 1 to the absolute value of the change in gate switching frequency; and combining the above three results. The process is as follows: First, the user behavior status assessment value is divided by the historical maximum user behavior status assessment value plus 1, and the ratio is added to 1 as the second part. The first part is multiplied by the second part to obtain the risk fluctuation assessment value. After calculating the risk fluctuation assessment value, it is compared in real-time with a preset risk trend threshold to determine the user's risk trend level. The risk trend threshold includes a low fluctuation threshold and a high fluctuation threshold: When the risk fluctuation assessment value is less than or equal to the low fluctuation threshold, it is considered a stable state, and only behavior records are saved without intervention. When the risk fluctuation assessment value is greater than the low fluctuation threshold but less than the high fluctuation threshold, it is considered a medium fluctuation state, and a risk change reminder is automatically pushed, while verification redundancy is increased to strengthen security verification. When the risk fluctuation assessment value is greater than or equal to the high fluctuation threshold, it is considered a high fluctuation state, and an early warning mechanism is immediately triggered, initiating multi-factor authentication and permission freezing measures, and linking the security response module to record alarms and early warning logs.

[0045] The specific formula for calculating the risk volatility assessment value is as follows:

[0046]

[0047] In the formula, S represents the risk volatility assessment value, ΔF represents the change in the number of current gate failures, ΔL represents the change in dwell time, and L max The maximum historical dwell time is represented by ΔC, the change in gate switching frequency is represented by R, and the user behavior status evaluation value is represented by R. max This represents the historical maximum user behavior status assessment value.

[0048] In this embodiment, Table 1 is a risk fluctuation assessment data table, recording the behavior and passage data for even-numbered cycles within 10 operation cycles. Specifically, in cycle 2: the change in the number of current gate segment failures is 2, the change in dwell time is 3, the change in gate segment switching frequency is 1, the user behavior status assessment value is 0.297, and the risk fluctuation assessment value is 3.137; in cycle 4: the change in the number of current gate segment failures is 2, the change in dwell time is 30, the change in gate segment switching frequency is -2, the user behavior status assessment value is 0.392, and the risk fluctuation assessment value is 3.876; in cycle 6: the change in the number of current gate segment failures is -1, the change in dwell time is -48, and the gate segment switching frequency... The change is -1, the user behavior status assessment value is 0.608, and the risk fluctuation assessment value is 2.461; Period 8: The change in the number of current gate segment failures is 0, the change in dwell time is -13, the change in gate segment switching frequency is 1, the user behavior status assessment value is 0.684, and the risk fluctuation assessment value is 1.009; Period 10: The change in the number of current gate segment failures is 2, the change in dwell time is -63, the change in gate segment switching frequency is -2, the user behavior status assessment value is 0.842, and the risk fluctuation assessment value is 4.813.

[0049] Table 1. Risk Volatility Assessment Data Table

[0050] cycle ΔF ΔL ΔC R S 2 2 3 1 0.30 3.14 4 2 30 -2 0.40 3.88 6 -1 -48 -1 0.61 2.46 8 0 -13 1 0.68 1.01 10 2 -63 -2 0.84 4.81

[0051] like Figure 3 As shown, this is a distribution chart of risk volatility intensity during the operating cycle, combined with Table 1 and... Figure 3 As can be seen, the changes in risk fluctuation assessment values ​​at even-numbered cycle points within a continuous operating cycle are highlighted by connecting even-numbered cycle nodes with red broken lines, emphasizing the fluctuation characteristics of user behavior risk at these key time points. Each marked blue diamond node in the figure represents an even-numbered cycle, and its corresponding risk fluctuation assessment value reflects the degree and trend change of abnormal user behavior during that period. Observing the fluctuations between even-numbered cycles in the figure can help identify the periodic fluctuation characteristics of behavioral risk and assist in determining whether there is a persistent abnormal trend, providing a more targeted time-series judgment basis for anomaly warning and response strategies.

[0052] This implementation plan achieves precise monitoring and efficient response to access control risks through a dynamic data processing and hierarchical control system. Within the framework of an AI system for the production field, firstly, based on real-time collected user access control data, a continuous behavior sequence is constructed by combining behavioral status assessment values, ensuring data timeliness and consistency, and laying a solid foundation for intelligent risk analysis. Secondly, the number of failures, dwell time, and door switching frequency for the current door segment are extracted on a minute-by-minute basis, and historical behavioral data is integrated to derive a risk fluctuation assessment value through a scientific quantitative calculation model, effectively depicting the risk change trend of user behavior, aligning with the stringent requirements for personnel safety in production scenarios. Thirdly, by setting low and high fluctuation thresholds, a tiered risk level determination mechanism is established, which can quickly distinguish risk states and provide support for automated decision-making. Finally, differentiated response strategies are implemented for different risk levels: records are archived under stable conditions, verification redundancy is enhanced during moderate fluctuations, and early warnings and linked security measures are immediately triggered during high fluctuations, thereby achieving closed-loop management of access control risks throughout the entire process. This significantly improves the security and emergency response capabilities of the access control system under the AI ​​system in the production field, ensuring the orderly conduct of production activities.

[0053] Specifically, the steps for quantitatively evaluating the effectiveness of risk response within different operational cycles are as follows: After the implementation of risk response measures, to comprehensively and accurately measure their actual effectiveness, the system will continuously collect user behavior status assessment values ​​and risk fluctuation assessment values, closely tracking the dynamic changes in user behavior risk and fluctuations over subsequent operational cycles. By continuously accumulating data, a continuous and complete data sequence is formed, laying a solid foundation for subsequent analysis. Based on this, the system obtains the user behavior status assessment values ​​for the t-th operational cycle, the user behavior status assessment values ​​for the (t+1)-th operational cycle, the risk fluctuation assessment values ​​for the t-th operational cycle, and the risk fluctuation assessment values ​​for the (t+1)-th operational cycle, thereby quantitatively evaluating the effectiveness of the risk response measures. Specifically, firstly, the user behavior status assessment value for the t-th operational cycle is subtracted from the user behavior status assessment value for the (t+1)-th operational cycle, obtaining the difference between the two. This difference can intuitively reflect the changes in user behavior status within adjacent cycles. Subsequently, this difference is multiplied by the response weight. The weight can be flexibly adjusted according to actual business needs to determine the importance of the user behavior status assessment value in the overall evaluation. Next, the product is divided by 1 and the user behavior status assessment value of the t-th operation period is added. This calculation method normalizes the result, and the resulting ratio serves as the first part of the quantitative assessment. Simultaneously, a similar but slightly different calculation logic is used for the risk fluctuation assessment value. First, the risk fluctuation assessment value of the t-th operation period is subtracted from the risk fluctuation assessment value of the (t+1)-th operation period to obtain the difference in risk fluctuation between adjacent periods. Then, this difference is multiplied by 1 and the response weight is subtracted. This setting reflects the relative importance of the risk fluctuation assessment value in the overall assessment, complementing the weight of the user behavior status assessment value. The response weight is obtained by training and optimizing using a logistic regression algorithm, combined with user historical access control behavior characteristics, abnormal event annotations, role levels, and verification results. The response weight's value range is [0, 1]. The product is then divided by 1 and the user behavior status assessment value of the t-th operation period is added, again undergoing normalization. The resulting ratio serves as the second part of the quantitative assessment. Finally, the first part and the second part are added together. Through this comprehensive calculation, a risk response performance value that can fully and accurately reflect the actual effect of risk response measures in different operating cycles is finally obtained.

[0054] The specific formula for calculating the risk response performance value is as follows:

[0055]

[0056] In the formula, E represents the risk response performance value, and R... t R represents the user behavior status evaluation value in the t-th operation period. t+k S represents the user behavior state evaluation value in the (t+k)th operation cycle.t S represents the risk volatility assessment value for the t-th operating period. t+k β represents the risk volatility assessment value for the (t+k)th operating period, and β represents the response weight.

[0057] In this embodiment, Table 2 is a risk response performance value data table, which records in detail the changes in user behavior status assessment value, risk fluctuation assessment value, and risk response performance value in the first five operation cycles. Specifically, in the first cycle, the user behavior status assessment value was 0.078, and the risk fluctuation assessment value was 1.768. In the next cycle, the user behavior status assessment value rose to 0.297, and the risk fluctuation assessment value rose to 3.137. The calculated risk response performance value was -0.319, indicating that the behavior status deteriorated and the risk increased during this cycle, resulting in a poor response effect. In the second cycle, the user behavior status assessment value was 0.297, and the risk fluctuation assessment value was 3.137. In the next cycle, the user behavior status assessment value became 0.321, and the risk fluctuation assessment value decreased to 1.488. The risk response performance value reached 0.148, indicating that the risk fluctuation significantly decreased during this cycle, leading to a positive improvement in response performance. In the third cycle, the user behavior status assessment value was 0.321, and the risk fluctuation assessment value was 1.488. In the next period, the user behavior status assessment value rose to 0.392, the risk fluctuation assessment value rose sharply to 3.876, and the risk response performance value dropped to -0.417, indicating that the decline in response performance was caused by the dual deterioration of behavior and risk. In the fourth period, the user behavior status assessment value was 0.392, and the risk fluctuation assessment value was 3.876. In the next period, the user behavior status assessment value rose slightly to 0.393, the risk fluctuation assessment value dropped to 3.463, and the risk response performance value was 0.034, meaning that the slight reduction in risk led to a slight improvement in response performance. In the fifth period, the user behavior status assessment value was 0.393, and the risk fluctuation assessment value was 3.463. In the next period, the user behavior status assessment value rose sharply to 0.608, the risk fluctuation assessment value dropped to 2.461, and the risk response performance value was -0.002.

[0058] Table 2 Risk Response Performance Value Data Table

[0059] cycle <![CDATA[R t ]]> <![CDATA[S t ]]> <![CDATA[R t+1 ]]> <![CDATA[S t+1 ]]> E 1 0.078 1.768 0.297 3.137 -0.319 2 0.297 3.137 0.321 1.488 0.148 3 0.321 1.488 0.392 3.876 -0.417 4 0.392 3.876 0.393 3.463 0.034 5 0.393 3.463 0.608 2.461 -0.002

[0060] like Figure 4 As shown, this is a line graph illustrating the dynamic evolution of risk response performance value in relation to risk volatility assessment value, combined with Table 2 and... Figure 4It can be seen that the risk response performance value fluctuates cyclically, with only periods 2 and 4 showing positive values. This is mainly driven by a decrease in the risk fluctuation assessment value; for example, the risk fluctuation assessment value in period 2 decreased from 3.137 to 1.488. The user behavior status assessment value continuously increased from period 1 to 5, rising from 0.078 to 0.393, indicating a deterioration in behavioral norms. The risk fluctuation assessment value fluctuates drastically and without a clear pattern, and its rise and fall are highly correlated with the positive and negative values ​​of the risk response performance value. Overall, this indicates that the risk response effect is unstable, and priority should be given to controlling risk fluctuations and curbing the upward trend of the behavior status assessment value.

[0061] This implementation plan achieves objective and accurate measurement of risk response effectiveness by constructing a scientific and rigorous evaluation system. Within the framework of an AI system in the production field, continuous collection of user behavior status assessment values ​​and risk fluctuation assessment values ​​forms a coherent data sequence, fully preserving dynamic information on risk changes. Based on the differences in key indicators between adjacent operation cycles, and combined with response weights for weighted calculation and normalization, behavioral status and risk fluctuation changes are quantified into comparable values, effectively distinguishing their relative importance in the evaluation. The resulting risk response performance value comprehensively integrates the dynamic changes in behavior and risk, providing a quantitative basis for judging the effectiveness of risk response measures. This helps to optimize response strategies in a timely manner, improve the risk control capabilities and management efficiency of access control systems under the AI ​​system in the production field, and thus better meet the stringent requirements for security and efficiency in production scenarios.

[0062] Specifically, the steps for dynamically adjusting security response measures and periodically generating strategy optimization suggestions are as follows: After calculating the risk response performance value, compare the risk response performance value with the risk response threshold in real time: If the risk response performance value is greater than the risk response threshold, the risk is effectively suppressed, the security response intensity is automatically reduced, the frequency of multi-factor authentication is reduced, and the verification time interval is extended; if the risk response performance value is less than or equal to the risk response threshold, the risk control is insufficient, intervention measures are strengthened, a cross-gateway access verification mechanism is enabled and repeated verification is required between different gateways, the user's access permission to high-privilege gateways is temporarily frozen, and manual review is required; the risk response performance value obtained each time and its corresponding user action are recorded. The system provides a unified, structured storage for status assessment values, risk fluctuation assessment values, strategy adjustments, and execution results, forming a historical database of risk response effectiveness. It periodically extracts labeled samples from this database to construct a training dataset for continuous optimization of user behavior discrimination logic, risk status assessment mechanisms, and response strategy parameters. A priority matrix is ​​generated based on role level, gate number, and strategy combination effects to recommend optimal strategies. Regular strategy optimization suggestions are output, listing adjustment items, affected user groups, associated risk types, and expected intervention effects. Simultaneously, a response strategy change log is generated, recording parameter changes, change time, basis, and approval status to ensure that strategy versions are controllable, auditable, and rollbackable.

[0063] This implementation plan achieves dynamic adjustment of security response intensity by comparing risk response performance values ​​with risk response thresholds in real time. When risks are under control, verification costs are reduced; when risks are not effectively controlled, intervention is strengthened, balancing security and access efficiency. Structured storage of risk response-related data forms a historical database, providing rich samples for model optimization and continuously iterating user behavior identification, risk assessment, and response strategies. A priority matrix is ​​generated based on role level, gate number, and strategy effectiveness to accurately recommend the optimal strategy, enhancing the scientific nature of decision-making. Strategy optimization suggestions are periodically output along with impact analysis and expected effects to assist managers in making decisions. Simultaneously, a complete strategy change log is recorded to ensure the transparency and traceability of the strategy adjustment process, achieving dynamic optimization and full lifecycle management of access control security strategies, effectively addressing complex and ever-changing security risks.

[0064] like Figure 2 As shown, the second aspect of this invention provides a multi-role hierarchical access control system for a sliding electric door, comprising: a user access control data acquisition and preprocessing module, a user behavior quantitative analysis and behavior rating module, an access control risk fluctuation identification and early warning module, and an access control risk response optimization and strategy iteration module. The user access control data acquisition and preprocessing module is used to collect user access control data during user authentication and access operations, and to perform data cleaning and normalization on the user access control data. The user behavior quantitative analysis and behavior rating module is used to slice user behavior data by time period, extract statistical features, identify abnormal patterns, and perform quantitative analysis of user behavior status. The access control risk fluctuation identification and early warning module is used to collect access control data in real time to construct behavior sequences, assess the magnitude and trend of risk fluctuations, determine the risk level, and perform risk response. The access control risk response optimization and strategy iteration module is used to quantitatively evaluate the risk response effect within different operating cycles, dynamically adjust security response measures, and periodically generate strategy optimization suggestions.

[0065] This implementation plan constructs a complete, intelligent, and dynamic access control security management system through the coordinated operation of four functional modules. The user access control data acquisition and preprocessing module ensures the accuracy and standardization of raw data, laying a solid data foundation for subsequent analysis; the user behavior quantitative analysis and behavior rating module enables in-depth analysis of user behavior, accurately identifying abnormal patterns and quantifying behavioral states, effectively preventing potential risks; the access control risk fluctuation identification and early warning module monitors risk trends based on real-time data, responding promptly and at different levels to control risks at their nascent stage; and the access control risk response optimization and strategy iteration module dynamically adjusts security measures and continuously optimizes strategies by quantitatively evaluating the effectiveness of risk responses, achieving a closed-loop management system. These modules complement each other, not only improving the security and management efficiency of the access control system but also allowing for flexible adjustments based on actual needs, adapting to complex and ever-changing application scenarios, and providing a scientific and efficient solution for access control management.

[0066] The third aspect of this invention provides a multi-role hierarchical access control device for a sliding electric door, comprising: an identity recognizer, a behavior evaluator, a risk responder, and a strategy optimizer, wherein: the identity recognizer is used to collect and parse user identity, role level, and access behavior information to achieve real-time perception and structured processing of access control operation data; the behavior evaluator is used to perform time-sharing analysis of user access control behavior data, extract key features and identify abnormal patterns, and quantitatively assess the behavior risk level by combining role level and behavior weight; the risk responder is used to continuously analyze user behavior status and risk fluctuation changes, dynamically evaluate the security control effect, and execute corresponding intervention measures to adjust access permissions and verification strength; the strategy optimizer is used to perform strategy adaptation analysis and weight fine-tuning based on historical response effects and role scenario data, generate the optimal response strategy, and maintain auditable rule iteration records.

[0067] This implementation plan focuses on artificial intelligence systems in the production field. Through the organic collaboration of four components—identity recognizer, behavior evaluator, risk responder, and strategy optimizer—a highly efficient and intelligent access control management system is constructed. Within this framework, the identity recognizer enables real-time collection and structured processing of access control data, providing precise data support for subsequent analysis. The behavior evaluator deeply mines user behavior data, scientifically quantifies risk levels based on roles and weights, and accurately identifies abnormal behaviors, meeting the strict control requirements for personnel operation procedures and permissions in production scenarios. The risk responder dynamically monitors changes in behavior and risk, evaluates control effectiveness in real time, and flexibly adjusts access permissions and verification strength to effectively prevent risks and ensure the safe and stable operation of production areas. The strategy optimizer continuously optimizes response strategies based on historical data and role scenarios, and retains rule iteration records to ensure the scientific nature and traceability of strategies, adapting to the management challenges brought about by dynamic changes in production processes. Each component of this device performs its specific function while working closely together, significantly improving the security, intelligence level, and management efficiency of the access control system. It meets the multi-role hierarchical management needs of artificial intelligence systems in the production field, provides reliable protection for access control security, and helps achieve intelligent and refined management in production scenarios.

[0068] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0069] The preferred embodiments of the present invention disclosed above are merely illustrative of the invention. These preferred embodiments do not exhaustively describe all details, nor do they limit the invention to the specific implementations described. Clearly, many modifications and variations can be made based on the content of this specification. This specification selects and specifically describes these embodiments to better explain the principles and practical applications of the invention, thereby enabling those skilled in the art to better understand and utilize the invention. The invention is limited only by the claims and their full scope and equivalents.

Claims

1. A multi-role hierarchical access control method of a segment sliding electric door, characterized by, The method comprises the following steps: S1, collecting user access control data during user identity authentication and access operation, and performing data cleaning and normalization processing on the user access control data; S2, slicing user behavior data by time period, extracting statistical features and identifying abnormal patterns, and quantitatively analyzing the behavior state of the user; S3, collecting access control data in real time to construct a behavior sequence, evaluating the risk fluctuation range and trend, determining the risk level and performing risk response; S4, quantitatively evaluating the risk response effect in different operation periods, dynamically adjusting the security response measures, and generating strategy optimization suggestions periodically.

2. The multi-role hierarchical access control method of a segment sliding electric door according to claim 1, characterized in that: The specific steps of collecting user access control data during user identity authentication and access operation are as follows: The user access control data is collected by the access control system during user identity authentication and access operation, and the user access control data includes user ID, role level, door segment number, verification time consumption, historical maximum verification time consumption, current door segment attempt number, current door segment failure number, average operation interval, residence time, historical maximum residence time, and door segment switching frequency; wherein, the identity information is obtained by reading the terminal through the access card and recorded by the identity authentication module, and the user ID is obtained by the identity authentication module; the role level is obtained by calling the permission database interface after identity authentication; the current access target is obtained by communicating with the segment sliding electric door controller to record the current access target, and the door segment number is obtained; the verification time consumption and the historical maximum verification time consumption are obtained by recording the difference between the identity authentication request time and the door body release response time; the average operation interval is obtained by comparing and calculating the time stamps of the same user's continuous operation, and then calculating the average of the time difference of the user's multiple operations; the current door segment attempt number is obtained by counting all the access attempts of the user in the current door segment within a fixed time window; the current door segment failure number is obtained by screening the number of records of the above attempts that fail to verify; the residence time and the historical maximum residence time are obtained by identifying the access time pairing of the user in the entrance and exit door segments; and the door segment switching frequency is obtained by analyzing the change of the door segment number of the user's continuous attempts and counting the number of switching times per unit time.

3. The multi-role hierarchical access control method of a segment sliding electric door according to claim 1, characterized in that: The specific steps of performing data cleaning and normalization processing on the user access control data are as follows: A multi-stage processing procedure is adopted to realize data cleaning and normalization processing: a sliding window average combined with a median absolute deviation detection method is adopted to identify and eliminate outliers and mutation abnormalities in the verification time consumption and residence time; a time sequence forward filling and local regression interpolation method is adopted to fill in the missing access records caused by communication delay and local packet loss, and the failure number and door segment switching frequency are filled in; the time alignment and behavior synchronization of the user identity, access behavior and role permission in a unified behavior period are realized through the unified association mapping of the time stamp, user ID and door segment number; An exponential moving filter combined with a bilateral filter is adopted to smooth and denoise the operation frequency and attempt number, and the behavior trend characteristics are retained; a standard deviation standardization and Box-Cox transformation are adopted to construct a joint normalization procedure for different dimension data characteristics, and the user access control data is standardized to unify the scale range of the user access control data.

4. The multi-role hierarchical access control method of a segment sliding electric door according to claim 1, characterized in that: The specific steps of extracting statistical features and identifying abnormal patterns from the time-sliced user behavior data are as follows: After sorting the input data according to user ID and timestamp, a fixed-length sliding time window is used to divide the data and generate time-continuous behavior segments. Statistical features are calculated for each access record in the behavior segment to form a corresponding behavior feature vector. Conditional judgment is performed on the behavior feature vector: behaviors with high door segment switching frequency and more than 3 failed attempts within a unit time are marked as trial access; behaviors with repeated failed verification within the same door segment for more than a set repetition threshold are marked as brute force attempt; behaviors with a stay time more than twice the user's historical average are marked as stay anomaly; behaviors attempting to enter a door segment with a higher permission level than their own role and failing verification are marked as unauthorized behavior; a sliding window rolling statistics method is used to assist in detecting abnormal behavior, and all abnormalities are recorded in a structured form.

5. The multi-role hierarchical access control method of a segment sliding electric door according to claim 1, characterized in that: The specific steps of quantitatively analyzing the user's behavior state are as follows: Record the number of unauthorized attempts by the user, obtain the user's access data, and quantitatively analyze the user's behavior state: divide the current door segment failure count by the current door segment attempt count plus a small item, divide the verification time consumption by the historical maximum verification time consumption, take the inverse of the average operation interval, divide the stay time by the historical maximum stay time, add the above four results, then add the door segment switching frequency and unauthorized attempt count, and the sum is the first part; subtract the user's level from the maximum role level and divide by the maximum role level, then add 1 to this ratio as the second part; multiply the behavior weight, the first part, and the second part to obtain the user behavior state evaluation value; After calculating the user behavior state evaluation value, compare it with the behavior state threshold value in real time to determine the user behavior state; the behavior state threshold value includes a first-level behavior state threshold value and a second-level behavior state threshold value: when the user behavior state evaluation value is less than or equal to the first-level behavior state threshold value, the user behavior is normal and no additional intervention is needed; when the user behavior state evaluation value is greater than the first-level behavior state threshold value and less than the second-level behavior state threshold value, the user behavior is suspicious and a warning notification is generated to inform the security management personnel, the verification process is encrypted and the verification time interval is extended; when the user behavior state evaluation value is greater than the second-level behavior state threshold value, the user behavior is high-risk and the automatic blocking strategy is triggered, the user access permission is suspended, the warning log is recorded, and the security response module is linked for alarm.

6. The multi-role hierarchical access control method of a segment sliding electric door according to claim 1, characterized in that: The specific steps of real-time collection of access data, construction of behavior sequence, evaluation of risk fluctuation amplitude and trend, determination of risk level, and risk response are as follows: Real-time collection of user access data, calculation of the latest user behavior state evaluation value; construction of continuous behavior sequence with user ID and timestamp as index; division of operation period in units of minutes, extraction of key behavior indicator changes in adjacent two periods, including current door segment failure count change, stay time change, and door segment switching frequency change; The historical maximum user behavior state evaluation value is extracted, user access control data, user behavior state evaluation value, current door section failure number change, retention time change and door section switching frequency change are obtained, and user risk state change trend is evaluated: the absolute value of the current door section failure number change is taken, the absolute value of the retention time change is divided by the historical maximum retention time plus 1, the absolute value of the door section switching frequency change is added by 1 and the logarithm is taken, and the sum of the above three results is taken as the first part; the user behavior state evaluation value is divided by the historical maximum user behavior state evaluation value plus 1, and the ratio is added by 1 as the second part; The first part is multiplied by the second part to obtain the risk fluctuation evaluation value; After obtaining the risk fluctuation evaluation value, the risk fluctuation evaluation value is compared with the preset risk trend threshold in real time, and the user risk trend level is determined; The risk trend threshold includes a low fluctuation threshold and a high fluctuation threshold: when the risk fluctuation evaluation value is less than or equal to the low fluctuation threshold, it is determined to be a stable state, only the behavior record is saved, and no intervention is required; when the risk fluctuation evaluation value is greater than the low fluctuation threshold and less than the high fluctuation threshold, it is determined to be a medium fluctuation state, and a risk change reminder is automatically pushed, and verification redundancy is increased to strengthen security verification; When the risk fluctuation evaluation value is greater than or equal to the high fluctuation threshold, it is determined to be a high fluctuation state, and an early warning mechanism is triggered immediately, multi-factor authentication and permission freezing measures are started, and a security response module is linked to alarm and warning log recording.

7. The multi-role hierarchical access control method of a segment sliding electric door according to claim 1, characterized in that: The specific steps of quantitatively evaluating the risk response effect in different operation periods are as follows: After the risk response is executed, the user behavior state evaluation value and the risk fluctuation evaluation value are continuously collected, the change trend of the user behavior risk and fluctuation in the subsequent multiple operation periods is tracked, and a continuous data sequence is formed; The user behavior state evaluation value of the tth operation period, the user behavior state evaluation value of the t+1th operation period, the risk fluctuation evaluation value of the tth operation period and the risk fluctuation evaluation value of the t+1th operation period are obtained, and the effectiveness of the risk response measure is quantitatively evaluated: the user behavior state evaluation value of the tth operation period is subtracted from the user behavior state evaluation value of the t+1th operation period, multiplied by the response weight, and divided by 1 plus the user behavior state evaluation value of the tth operation period, and the ratio is taken as the first part; the risk fluctuation evaluation value of the tth operation period is subtracted from the risk fluctuation evaluation value of the t+1th operation period, multiplied by 1 minus the response weight, and divided by 1 plus the user behavior state evaluation value of the tth operation period, and the ratio is taken as the second part; The first part is added to the second part to obtain the risk response performance value.

8. The multi-role hierarchical access control method of a segment sliding electric door according to claim 1, characterized in that: The specific steps of dynamically adjusting the security response measure and periodically generating strategy optimization suggestions are as follows: After the risk response performance value is calculated, the risk response performance value and the risk response threshold value are compared in real time: if the risk response performance value is greater than the risk response threshold value, the risk is effectively suppressed, the security response strength is automatically reduced, the multi-factor authentication frequency is reduced, and the verification time interval is extended; if the risk response performance value is less than or equal to the risk response threshold value, the risk control is insufficient, the intervention measures are strengthened, the cross-door access verification mechanism is enabled and repeated verification between different door segments is required, the access permission of the user to the high-privilege door segment is temporarily frozen, and manual review is required; The risk response performance value obtained each time, the corresponding user behavior state evaluation value, the risk fluctuation evaluation value, and the policy adjustment and execution result are stored in a unified structure to form a risk response effect history database; labeled samples are extracted from the risk response effect history database at regular intervals to construct a training data set for continuously optimizing user behavior judgment logic, risk state evaluation mechanism and response strategy parameters; A priority matrix is generated based on the role level, the door segment number and the policy combination effect, which is used to recommend the optimal strategy; Periodically output policy optimization suggestions, list adjustment items, affected user groups, associated risk types and expected intervention effects; at the same time, generate response strategy change logs to record parameter changes, change time, basis and approval status, to ensure controllable, auditable and rollbackable policy versions.

9. A multi-role hierarchical access control system for a segment slide electric door, characterized in that: It comprises: a user access control data acquisition and preprocessing module, a user behavior quantitative analysis and behavior rating module, a door access risk fluctuation identification and early warning module, and a door access risk response optimization and strategy iteration module, wherein: The user access control data acquisition and preprocessing module is configured to collect user access control data during user identity verification and access operation, and perform data cleaning and normalization processing on the user access control data; The user behavior quantitative analysis and behavior rating module is configured to slice user behavior data by time period, extract statistical features and identify abnormal patterns, and quantitatively analyze the behavior state of the user; The door access risk fluctuation identification and early warning module is configured to collect door access data in real time to construct a behavior sequence, evaluate the risk fluctuation amplitude and trend, determine the risk level and perform risk response; The door access risk response optimization and strategy iteration module is configured to quantitatively evaluate the risk response effect in different operation periods, dynamically adjust the security response measures, and generate policy optimization suggestions at regular intervals.

10. A multi-role hierarchical access control device for a segment sliding electric door, characterized by: It comprises: an identity recognizer, a behavior evaluator, a risk responder, and a strategy optimizer, wherein: The identity recognizer is configured to collect and analyze user identity, role level and access behavior information, realize real-time sensing and structured processing of access operation data; The behavior evaluator is configured to analyze user access behavior data by time, extract key features and identify abnormal patterns, and quantitatively evaluate the behavior risk level in combination with the role level and behavior weight; The risk responder is configured to continuously analyze the user behavior state and risk fluctuation changes, dynamically evaluate the security control effect, and execute corresponding intervention measures to adjust the access permission and verification strength; The strategy optimizer is configured to perform strategy adaptation analysis and weight fine-tuning based on historical response effects and role-scene data, generate an optimal response strategy, and maintain an auditable rule iteration record.

Citation Information

Patent Citations

  • Access controller

    CN102222379A

  • Gate inhibition controller

    CN105069878A

  • Intelligent access control method and system

    CN112037410A

  • Intelligent access control management method and system based on multi-mode identification and Internet of Things technology

    CN118968665A

  • Intelligent access control recognition system

    CN120071486A

Cited By

  • Credit dynamic assessment risk control method and system based on tendency score matching

    CN121504598A

  • Credit dynamic assessment risk control method and system based on propensity score matching

    CN121504598B

  • Non-inductive access method and device based on Bluetooth key, medium and program product

    CN121617169A

  • A method, device, medium and program product for contactless entry based on a Bluetooth key

    CN121617169B