A method and system for managing permissions based on user behavior

Abstract

The application discloses a kind of based on user behavior's right management method and system, belong to artificial intelligence field, including following steps: S1, the behavior data of electric user is obtained, and data is preliminarily cleaned;S2, the multi-dimensional feature extraction of behavior data after cleaning is carried out, and user behavior portrait is generated in combination with user identity label;S3, user behavior portrait is input into behavior scoring model, and behavior score and risk coefficient are output;S4, according to the risk mapping rule of scoring, and the right level is dynamically configured by right control engine.The application can monitor and analyze the behavior mode of user in real time by introducing the dynamic right adjustment mechanism based on user behavior score and risk assessment, and dynamically adjust the right level of user according to the real-time demand and risk level of user, can judge the right demand of user according to actual operation behavior, avoid excessive or insufficient right authorization, improve the intelligence and dynamic of right management.

Description

Technical Field

[0001] This invention relates to the field of artificial intelligence, and more specifically, to a method and system for permission management based on user behavior. Background Technology

[0002] Traditional role-based access control (RBAC) systems, while widely used in organizations, have significant limitations. Because roles and permissions are configured based on static rules, these systems fail to account for the diversity and real-time changes in user behavior, often leading to permission abuse, misoperation, and security vulnerabilities. Furthermore, traditional access control methods ignore changes in the system environment, contextual information about user behavior, and fail to effectively assess user risk behaviors, making access control too rigid and unable to adapt to complex security needs.

[0003] In recent years, with the development of intelligent technologies, more and more research has begun to explore how to optimize access control through dynamic and personalized user behavior data. For example, by analyzing users' specific behaviors in the system, such as login frequency, operation patterns, and abnormal behaviors, it is possible to assess users' real needs and operational risks, thereby achieving behavior-based dynamic access control. This type of method not only improves system security but also significantly enhances user experience, ensuring the rationality and accuracy of permissions. However, in practical applications of these methods in power grids (such as energy management platforms and power dispatching systems), due to the diversity of electricity user behavior, they still suffer from insufficient intelligence and a lack of dynamism. For instance, different users in the same role may have completely different usage habits, response efficiencies, and propensities for misoperation. RBAC models lack the ability to perceive and adjust for these differences. Therefore, problems such as permission redundancy, operational loss of control, and abuse risks may occur in the system, especially in security-sensitive business scenarios where the potential risks are particularly prominent. Furthermore, RBAC models typically fail to incorporate contextual information (such as current network status, device health status, and external environment), making it difficult to achieve effective dynamic access control in complex systems with high real-time requirements. Summary of the Invention

[0004] The purpose of this invention is to provide a user behavior-based permission management method and system to solve the problems mentioned in the background section.

[0005] A user behavior-based access control method includes the following steps:

[0006] S1. Obtain behavioral data of electricity users and perform preliminary data cleaning;

[0007] S2. Extract multi-dimensional features from the cleaned behavioral data and generate user behavior profiles by combining user identity tags.

[0008] S3. Input the user behavior profile into the behavior scoring model and output the behavior score and risk coefficient.

[0009] S4. The permission control engine dynamically configures the permission level according to the scoring risk mapping rules;

[0010] S5. The permission levels are used to control the scope of operation of users in the energy management platform and power dispatch system, and can realize hierarchical authorization, auditing, and traceability.

[0011] Preferably, the user behavior profile in S2 includes: electricity consumption pattern characteristics, energy-saving execution score, response delay level, operation stability score, and scheduling coordination index.

[0012] Preferably, the electricity consumption pattern characteristics Used to assess the periodicity and volatility of user electricity consumption:

[0013]

[0014] in Indicates daily average power. Indicates the daily power standard deviation;

[0015] Energy-saving performance score Used to measure whether users participate in energy-saving plans or peak-shifting instructions as required:

[0016]

[0017] in This indicates the number of times energy-saving control was successfully executed during the scheduling response. This indicates the total number of energy-saving control commands issued by the system.

[0018] The response latency level Measure the average response time after a user receives a scheduling command:

[0019]

[0020] in Indicates the number of scheduling instructions. Indicates the first Response time (in seconds) for each scheduling command. Indicates the maximum allowed response time (in seconds) of the system;

[0021] The operational stability score Reflects the reliability and error rate of equipment control operations:

[0022]

[0023] in Indicates the number of incorrect operations. Indicates the total number of operations;

[0024] The scheduling coordination index Reflects the user's ability to coordinate response with other users (group load response):

[0025]

[0026] in, This indicates the number of times users effectively collaborated in group scheduling. This indicates the number of requests the user has made to participate in group scheduling.

[0027] Preferably, the final behavioral profile is represented as a feature vector:

[0028]

[0029] Among them, all eigenvalues .

[0030] Preferably, the comprehensive behavioral score S in S3 is calculated as follows:

[0031]

[0032] Among them, weight Based on business importance and strategy configuration, these settings are configured by administrators themselves.

[0033] Risk coefficient R:

[0034]

[0035] in This indicates the number of abnormal behaviors within the recent observation period. This represents the total number of behavioral records within the observation period. Represents behavioral profile vectors The standard deviation reflects the degree of behavioral fluctuation. Indicates the weighting factor, recommended value: .

[0036] Preferably, a behavior context awareness mechanism is introduced before step S4 to determine whether the behavior constitutes a misoperation or an occasional behavior based on the input variables. If the behavior is identified as a misoperation or an occasional behavior, it enters an observation period, records a fault tolerance log, and temporarily suspends permission adjustments. If the abnormality persists during the observation period, a permission downgrade strategy is executed. If the behavior is not determined to be a misoperation, or is verified as a high-risk behavior during the observation period, the permission control engine dynamically configures the permission level according to the scoring risk mapping rules.

[0037] Preferably, the input variables for the behavior context awareness mechanism include: meteorological information, equipment operating status, historical scheduling frequency, operation period, terminal status, and user behavior frequency.

[0038] A user behavior-based access control system includes:

[0039] The behavior acquisition module is used to acquire the behavior data of electricity users and perform preliminary data cleaning.

[0040] The user profile modeling module is used to extract multi-dimensional features from the cleaned behavioral data and generate user behavior profiles by combining user identity tags.

[0041] The scoring engine is used to input user behavior profiles into the behavior scoring model and output behavior scores and risk coefficients.

[0042] The access control engine is used to dynamically configure permission levels based on the scoring risk mapping rules.

[0043] Preferably, the system also includes a context awareness module, which interfaces with a meteorological platform, a load monitoring system, and a power equipment status monitoring system to obtain real-time context data for behavior analysis. The context awareness module is used to determine whether a behavior constitutes a misoperation or an occasional behavior based on the current system load, equipment status, scheduling plan, weather information, etc. If the behavior is identified as a misoperation or an occasional behavior, it enters an observation period, records a fault tolerance log, and temporarily suspends permission adjustment. If the abnormality persists during the observation period, a permission downgrade strategy is executed. If the behavior is not determined to be a misoperation, or is verified as a high-risk behavior during the observation period, the permission control engine dynamically configures the permission level according to the scoring risk mapping rules.

[0044] Preferably, the system also includes a permission execution module, which supports remote blocking, dynamic permission issuance, behavior auditing, and user application retrospective interfaces to ensure the prudence and rationality of permission changes.

[0045] Compared with the prior art, the advantages of this invention are:

[0046] (1) By introducing a dynamic permission adjustment mechanism based on user behavior scoring and risk assessment, this invention can monitor and analyze user behavior patterns in real time, and dynamically adjust the user's permission level according to the user's real-time needs and risk level. It can judge the user's permission needs based on the user's actual operation behavior, avoid excessive or insufficient permission authorization, and improve the intelligence and dynamism of permission management.

[0047] (2) By combining user behavior scores and risk coefficients, the present invention can monitor and manage users in real time and dynamically, which can promptly detect and restrict potential risky behaviors and significantly reduce the risk of abuse of privileges and system attacks.

[0048] (3) This invention, through personalized permission control based on user behavior, can intelligently adjust permissions according to the actual needs and behavioral characteristics of users. This not only avoids the limitations caused by fixed roles or permission scope in traditional permission management methods, but also optimizes permission configuration according to changes in user behavior and environment, meets the permission needs of different users in different situations, and improves user experience.

[0049] (4) This invention further introduces a context-aware mechanism, which can combine real-time context data, such as user behavior history, device status, and external environment, to accurately identify the true intent and risk level behind user behavior. This precise access control can effectively reduce the occurrence of misoperations and provide the system with detailed auditing and backtracking functions, ensuring that every permission change can be clearly recorded, audited, and backtracked, thereby enhancing the transparency and traceability of the system. Attached Figure Description

[0050] Figure 1 This is a flowchart of the user behavior-based permission management method of the present invention. Detailed Implementation

[0051] For examples, see Figure 1 A user behavior-based access control method includes the following steps:

[0052] S1. Obtain behavioral data of electricity users and perform preliminary data cleaning;

[0053] S2. Extract multi-dimensional features from the cleaned behavioral data and generate user behavior profiles by combining user identity tags.

[0054] S3. Input the user behavior profile into the behavior scoring model and output the behavior score and risk coefficient.

[0055] S4. The permission control engine dynamically configures the permission level according to the scoring risk mapping rules;

[0056] S5. The permission levels are used to control the scope of operation of users in the energy management platform and power dispatch system, and can realize hierarchical authorization, auditing, and traceability.

[0057] The user behavior profile in S2 includes: electricity consumption pattern characteristics, energy-saving execution score, response latency level, operation stability score, and scheduling coordination index.

[0058] The characteristics of electricity consumption patterns Used to assess the periodicity and volatility of user electricity consumption:

[0059]

[0060] in Indicates daily average power. This represents the daily power standard deviation; the higher the stability (smaller the standard deviation), the closer the score is to 1.

[0061] Energy-saving performance score Used to measure whether users participate in energy-saving plans or peak-shifting instructions as required:

[0062]

[0063] in This indicates the number of times energy-saving control was successfully executed during the scheduling response. This indicates the total number of energy-saving control commands issued by the system. A higher response execution rate results in a higher score. ).

[0064] The response latency level Measure the average response time after a user receives a scheduling command:

[0065]

[0066] in Indicates the number of scheduling instructions. Indicates the first Response time (in seconds) for each scheduling command. This indicates the maximum allowable response time (in seconds). The faster the response, the slower the exponential decay, and the closer the score is to 1.

[0067] The operational stability score Reflects the reliability and error rate of equipment control operations:

[0068]

[0069] in Indicates the number of incorrect operations. This represents the total number of operations. The fewer the errors, the higher the score.

[0070] The scheduling coordination index Reflects the user's ability to coordinate response with other users (group load response):

[0071]

[0072] in, This indicates the number of times users effectively collaborated in group scheduling. This indicates the number of requests the user made to participate in group scheduling; the higher the cooperation level within the group, the higher the coordination score.

[0073] The final behavioral profile is represented as a feature vector:

[0074]

[0075] Among them, all eigenvalues .

[0076] The comprehensive behavioral score S in S3 is calculated as follows:

[0077]

[0078] Among them, weight Based on business importance and strategy configuration, these settings are configured by administrators themselves.

[0079] Risk coefficient R:

[0080]

[0081] in This indicates the number of abnormal behaviors within the recent observation period. This represents the total number of behavioral records within the observation period. Represents behavioral profile vectors The standard deviation reflects the degree of behavioral fluctuation. Indicates the weighting factor, recommended value: .

[0082] Ultimately, these rating results replaced the uniform role permissions in traditional RBAC, enabling differentiated management of user permissions for the same role.

[0083] Prior to S4, a behavior context awareness mechanism is introduced to determine whether a behavior constitutes a misoperation or an occasional behavior based on input variables. If the behavior is identified as a misoperation or an occasional behavior, it enters an observation period, records a fault tolerance log, and temporarily suspends permission adjustments. If the abnormality persists during the observation period, a permission downgrade strategy is implemented. If the behavior is not determined to be a misoperation, or is verified as a high-risk behavior during the observation period, the permission control engine dynamically configures the permission level according to the scoring risk mapping rules.

[0084] Step 1: Input user behavior metrics

[0085] Receive the following input data: behavioral scores Risk coefficient ;

[0086] Step 2: Define the scoring risk mapping matrix as shown in Table 1

[0087] Construct a two-dimensional rating-risk mapping table and define permission levels (P0~P4):

[0088]

[0089] Table 1

[0090] Permission Level Description: P0: Block (prohibits all active operations); P1: Read-only (can only view data); P2: Restricted (some operations require authorization); P3: Normal (normal operation permissions); P4: Full Control (allows all operations and automatic control).

[0091] Step 3: Determine if the current permission status matches the target permission.

[0092] Let the current permission level be Calculate the target level as ,like

[0093] If the time remains unchanged, then continue to step 4;

[0094] Step 4: Set permission adjustment policies (including fault tolerance and buffering)

[0095] 4.1 Determine if an observation period exists. If a user is marked as "observation status" under the error tolerance mechanism, then: if the target level is lower than the current level, only record the change request and do not immediately demote the user; start the observation timer (if no new anomalies are added within 24 hours, it can be restored).

[0096] 4.2 Implement a buffer and demotion mechanism to prevent drastic changes. If the system has set a throttling threshold for permission changes (e.g., maximum change of 1 level per day), then permissions will be decremented / incremented level by level.

[0097] Step 5: Execute the permission level change operation. Based on the permission change instruction, the permission control engine calls the following sub-modules:

[0098] Permission publisher: Updates permission policies to user terminals and platform accounts.

[0099] Logger: Records the source of changes, the reason for changes, and timestamps.

[0100] Notification Center: Sends notifications to users regarding permission changes and the reasons for such changes.

[0101] Recovery trigger: Automatically attempt to restore permission level once behavior returns to normal.

[0102] Step 6: Continuous monitoring and periodic evaluation. Recalculate S and R every 5 / 15 / 30 minutes (configurable) to perform real-time dynamic maintenance of permission levels. Real-time permission adjustments can be triggered by abnormal events (such as overstepping boundaries, device malfunctions, etc.).

[0103] The input variables for the behavior context awareness mechanism include: meteorological information, equipment operating status, historical scheduling frequency, operation period, terminal status, and user behavior frequency.

[0104] The error detection module sets a fault tolerance threshold. If the number of errors does not exceed the preset value within a unit of time, the original permission level is maintained and the error is logged.

[0105] The duration of the observation period and the frequency of behavior monitoring are dynamically adjusted based on the user's risk level. If the behavior remains stable during the observation period, the original permissions will be automatically restored.

[0106] The access control engine supports five levels of permission configuration: P0 blacklist blocking, P1 read-only permission, P2 operation requiring authorization, P3 normal operation permission, and P4 full control permission.

[0107] A user behavior-based access control system includes: a behavior collection module, a profile modeling module, a scoring engine, a context-aware module, an access control engine, an access control module, and a log recording and backtracking module.

[0108] The context awareness module interfaces with the meteorological platform, load monitoring system, and power equipment status monitoring system to obtain real-time context data for behavior analysis.

[0109] The permission execution module supports remote blocking, dynamic permission issuance, behavior auditing, and user application retrospective interfaces to ensure the prudence and rationality of permission changes.

[0110] The core function of the behavior acquisition module is to collect various user behavior data in real time. This behavior data includes, but is not limited to, power consumption curves, equipment operation logs, dispatch response records, out-of-bounds control behaviors, abnormal alarm history, and collaborative control indicators.

[0111] Data source: The behavior acquisition module periodically acquires users' electricity consumption data through interfaces of devices such as sensors and smart meters. Simultaneously, the module collaborates with equipment management systems and dispatch control systems to collect real-time operation logs and response records.

[0112] Data collection frequency: The module configures the data collection cycle, which is once per minute by default; for important events, such as system anomalies or changes in user behavior, it provides second-level collection functionality.

[0113] Electricity consumption data (such as power curves) is stored in a time-series data format indexed by timestamps, supporting high-frequency data acquisition and storage.

[0114] Device operation logs, abnormal alarms, and scheduling responses are stored in JSON format for easy parsing and processing later.

[0115] The user profile modeling module uses collected behavioral data and data mining and machine learning algorithms to build user behavior profiles and generate corresponding behavioral feature vectors.

[0116] Data cleaning and preprocessing: The collected raw data is cleaned to remove outliers and noisy data, and the time series data is smoothed.

[0117] Feature engineering: Extract important features from user behavior data, such as electricity consumption regularity, energy-saving performance score, response timeliness, operation stability, and scheduling coordination, and normalize the data to the [0,1] interval through standardization processing.

[0118] Modeling method: A behavioral profile model suitable for this user group is trained by using machine learning models such as decision trees and random forests, combined with historical data and user behavior patterns.

[0119] Model update mechanism: The behavioral profile model is retrained periodically or based on abnormal events, according to long-term changes in user behavior.

[0120] The profile feature vectors are stored in a distributed database to ensure efficient retrieval and updates.

[0121] The update cycle and triggering mechanism for user profiles can be customized, such as monthly regular updates or real-time updates triggered when significant changes occur in user behavior.

[0122] The rating engine generates a user's behavioral score and risk coefficient based on the input behavioral feature vector.

[0123] The context-aware module acquires real-time external context data (such as meteorological information, power load data, etc.) and combines it with user behavior data to provide more accurate behavior analysis and risk assessment.

[0124] Data source interface: The context-aware module interfaces with the following platforms:

[0125] Meteorological platform: Acquires weather forecast information, temperature, humidity, wind speed and other data, and analyzes the impact of weather on electricity consumption behavior.

[0126] Load monitoring system: acquires power grid load information, determines the correlation between user behavior and power grid load, and predicts the risks brought about by load fluctuations.

[0127] Equipment status monitoring system: Acquires information on the operating status of power equipment and analyzes the impact of equipment health status on user behavior.

[0128] Data fusion and processing: The external context information and user behavior data are fused in multiple dimensions, and user behavior is predicted and warned based on big data analysis models.

[0129] The context-aware module interacts with external systems via RESTful APIs, supporting real-time data acquisition and periodic updates.

[0130] The access control engine dynamically configures user permission levels based on the behavior scores and risk coefficients output by the scoring engine, ensuring system security and flexibility.

[0131] Access control mapping rules: Based on scores and risk coefficients, the control engine generates target access levels using predefined mapping rules (such as a score-risk matrix).

[0132] Permission adjustment mechanism: The engine triggers permission changes through set rules and interfaces with user devices and the system to issue permission adjustment instructions in real time.

[0133] Gradual adjustment: When user ratings do not change significantly, permission adjustments are made gradually to prevent drastic fluctuations.

[0134] Anomaly monitoring: When user behavior is abnormal, the system will automatically enter an observation period, and after a certain period, permission restoration or downgrade operations will be performed.

[0135] The permission execution module is responsible for actually performing permission change operations, including remote blocking, dynamic permission issuance, behavior auditing and backtracking interfaces, to ensure the prudence and rationality of permission changes.

[0136] Permission issuance: Permission changes are issued to user devices in real time through the interface between the permission management system and the user terminal.

[0137] Remote blocking: For high-risk users, a blocking operation will be performed to prevent the user from performing any operations.

[0138] Behavior auditing: The permission execution module records all permission change logs and supports user requests for retrospective queries, ensuring the transparency and traceability of permission changes.

[0139] Remote control interface: Communicates with user terminals and executes access control commands.

[0140] Audit Log Interface: Supports audit query interface, records all operations and changes, and provides a visual query interface.

[0141] The logging and backtracking module is used to record user behavior data, permission change history, abnormal events, etc., providing complete auditing and backtracking capabilities for permission management.

[0142] Log collection: The module collects user behavior data, system warnings, permission changes and other events in real time and stores them in a distributed log system.

[0143] Retrospective function: Users can query permission change records within a specific time range and view the detailed reasons for each change.

[0144] Log data storage employs a combination of distributed databases and time-series databases to ensure efficient querying and persistent storage.

[0145] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely preferred examples of the present invention and are not intended to limit the present invention. Various changes and modifications can be made to the present invention without departing from the spirit and scope thereof, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A user behavior-based access control method, characterized in that, Includes the following steps: S1. Obtain behavioral data of electricity users and perform preliminary data cleaning; S2. Perform multi-dimensional feature extraction on the cleaned behavioral data and generate a user behavior profile by combining it with user identity tags; the user behavior profile includes: electricity consumption pattern characteristics, energy-saving execution score, response latency level, operation stability score and scheduling coordination index; S3. Input the user behavior profile into the behavior scoring model and output the behavior score and risk coefficient. S4. The permission control engine dynamically configures the permission level according to the scoring risk mapping rules; Its characteristic is that the electricity consumption pattern features Used to assess the periodicity and volatility of user electricity consumption: ， in Indicates daily average power. Indicates the daily power standard deviation; Energy-saving performance score Used to measure whether users participate in energy-saving plans or peak-shifting instructions as required: ， in This indicates the number of times energy-saving control was successfully executed during the scheduling response. This indicates the total number of energy-saving control commands issued by the system. The response latency level Measure the average response time after a user receives a scheduling command: ， in Indicates the number of scheduling instructions. Indicates the first Response time of a scheduling command Indicates the maximum allowable response time of the system; The operational stability score Reflects the reliability and error rate of equipment control operations: ， in Indicates the number of incorrect operations. Indicates the total number of operations; The scheduling coordination index Reflects a user's ability to collaborate and respond with other users: ， in, This indicates the number of times users effectively collaborated in group scheduling. This indicates the number of requests the user made to participate in group scheduling; The final behavioral profile is represented as a feature vector. : ， Among them, all eigenvalues ; The comprehensive behavioral score S in S3 is calculated as follows: ， Among them, weight Based on business importance and strategy configuration, these settings are configured by administrators themselves. Risk coefficient R: ， in This indicates the number of abnormal behaviors within the recent observation period. This represents the total number of behavioral records within the observation period. Represents behavioral profile vectors The standard deviation reflects the degree of behavioral fluctuation. Indicates the weighting factor, recommended value: .

2. The user behavior-based access control method according to claim 1, characterized in that, The behavior context awareness mechanism introduced before S4 determines whether the behavior constitutes a misoperation or an occasional behavior based on the input variables. If the behavior is identified as a misoperation or an occasional behavior, it enters an observation period, records a fault tolerance log, and temporarily suspends permission adjustment. If the abnormality continues during the observation period, the permission downgrade strategy is executed. If the behavior is not judged as a misoperation, or is verified as a high-risk behavior during the observation period, the permission control engine will dynamically configure the permission level according to the scoring risk mapping rules.

3. The user behavior-based access control method according to claim 2, characterized in that, The input variables of the behavior context awareness mechanism include: meteorological information, equipment operating status, historical scheduling frequency, operation period, terminal status, and user behavior frequency.

4. A user behavior-based access control system, used to implement the user behavior-based access control method according to any one of claims 1 to 3, characterized in that, include: The behavior acquisition module is used to acquire the behavior data of electricity users and perform preliminary data cleaning. The user profile modeling module is used to extract multi-dimensional features from the cleaned behavioral data and generate user behavior profiles by combining user identity tags. The scoring engine is used to input user behavior profiles into the behavior scoring model and output behavior scores and risk coefficients. The access control engine is used to dynamically configure permission levels based on the scoring risk mapping rules.

5. A user behavior-based access control system according to claim 4, characterized in that, The system also includes a context awareness module, which interfaces with the meteorological platform, load monitoring system, and power equipment status monitoring system to obtain real-time context data for behavior analysis. The context awareness module is used to determine whether the behavior constitutes a misoperation or an occasional behavior based on the current system load, equipment status, scheduling plan, and weather information. If the behavior is identified as a misoperation or an occasional occurrence, it will enter an observation period, be logged in a fault tolerance log, and the permission adjustment will be temporarily suspended; if the abnormality continues during the observation period, the permission downgrade policy will be implemented. If the behavior is not judged as a misoperation, or is verified as a high-risk behavior during the observation period, the permission control engine will dynamically configure the permission level according to the scoring risk mapping rules.

6. A user behavior-based access control system according to claim 4, characterized in that, The system also includes a permission execution module, which supports remote blocking, dynamic permission issuance, behavior auditing, and user application retrospective interfaces to ensure the prudence and rationality of permission changes.

Images