Identity authentication method and system for electric power inspection terminal

Abstract

The application discloses a kind of identity authentication methods for electric power inspection terminal, including obtaining the historical inspection data information of electric power system and carrying out data preprocessing to obtain data set;Setting electric power inspection terminal fingerprint parameter probability model, and the relationship between the device state and data based on the obtained data set is expressed;Based on bayesian inference, the obtained model is updated to posterior distribution;According to the posterior distribution obtained, the fingerprint data of electric power inspection terminal is generated;According to the fingerprint data obtained, the identity authentication of actual electric power inspection terminal is completed.The application also discloses a kind of system for implementing the identity authentication method for electric power inspection terminal.The application is processed, modeling and fingerprint generation to the historical inspection data of electric power system, and according to the fingerprint of electric power inspection equipment generated, the identification of electric power inspection terminal is carried out;Therefore, the application not only can realize the identity authentication of electric power inspection terminal, but also has higher reliability and better security.

Description

Technical Field

[0001] The present invention belongs to the field of electrical automation, and in particular relates to an identity authentication method and system for a power inspection terminal. Background Art

[0002] With the development of economy and technology and the improvement of people's living standards, electricity has become an indispensable secondary energy source in people's production and life, bringing endless convenience to people's production and life. Therefore, ensuring a stable and reliable supply of electricity has become one of the most important tasks of the power system.

[0003] Power system inspections are crucial for ensuring stable and reliable operation. With the rapid development of intelligent technology, an increasing number of inspection terminals, such as inspection robots and drones, are being used in power system inspections. While these terminals offer significant convenience for power system personnel, data security concerns are also a growing concern.

[0004] The data security of power inspection terminals is directly related to the data security of the power system. Identity authentication of these terminals is paramount in ensuring their data security. Currently, the most commonly used terminal identity authentication schemes in power systems are interactive authentication schemes based on the communication process. However, these schemes pose the risk of attacks and decryption during communication, and their reliability and security are limited. Summary of the Invention

[0005] One of the objectives of the present invention is to provide an identity authentication method for a power inspection terminal with high reliability and good security.

[0006] A second object of the present invention is to provide a system for implementing the identity authentication method for power inspection terminals.

[0007] The identity authentication method for a power inspection terminal provided by the present invention comprises the following steps:

[0008] S1. Obtain historical inspection data information of the power system;

[0009] S2. Preprocess the data information obtained in step S1 to obtain a data set;

[0010] S3 sets the power inspection terminal fingerprint parameter probability model, and based on the data set obtained in step S2, represents the relationship between the device status and data;

[0011] S4. Based on Bayesian inference, update the posterior distribution of the model obtained in step S3;

[0012] S5. Generate fingerprint data of the power inspection terminal according to the posterior distribution obtained in step S4;

[0013] S6. Based on the fingerprint data obtained in step S5, the actual identity authentication of the power inspection terminal is completed.

[0014] The acquisition of historical inspection data information of the power system described in step S1 specifically includes the following steps:

[0015] Obtain historical inspection data information of the power system;

[0016] The inspection data information includes static data and dynamic data;

[0017] The static data includes the ID of the main control chip of the power inspection terminal, the sensor serial number of the power inspection terminal, the operating system version of the power inspection terminal, the designated driver list of the power inspection terminal, and the designated application list of the power inspection terminal;

[0018] Dynamic data includes network characteristic data and inspection characteristic data; among them, network characteristic data includes transmission rate and transmission delay; inspection characteristic data includes PID control parameters, inspection mode and GPS positioning data.

[0019] The pre-processing described in step S2 specifically includes the following steps:

[0020] Preprocessing includes missing data processing, abnormal data processing and data unification;

[0021] Missing data processing: If the amount of missing data is less than or equal to the set value, interpolation is used to fill the data; if the amount of missing data is greater than the set value, the corresponding data is directly discarded;

[0022] Abnormal data processing: Abnormal data is detected using a detection method and directly deleted; the detection method includes the Z-Score method or the IQR method;

[0023] Data unification: The data obtained after missing data processing and abnormal data processing are unified into a standard time axis to achieve synchronization of the time corresponding to the data; for the dynamic inspection data in the data, the data is divided into several segments and stored separately; the data segmentation specifically includes sliding window segmentation.

[0024] The setting of the fingerprint parameter probability model of the power inspection terminal in step S3 and the representation of the relationship between the device status and the data based on the data set obtained in step S2 specifically include the following steps:

[0025] The following normal distribution model is used as the prior distribution, and the initial parameter model is constructed:

[0026] p(μ)=N(μ1|μ0,(δ0) 2 )

[0027] Where p(μ) is the normal distribution of parameter μ; N() is the symbol for normal distribution; μ1 is a random variable; μ0 is the expected value of normal distribution; (δ0) 2 is the variance of the normal distribution;

[0028] According to the prior distribution, the following formula is used as the likelihood function:

[0029]

[0030] Where p(X|μ) is the likelihood function; X is the observed quantity; μ is the expected value of the prior normal distribution; δ is the standard deviation of the prior normal distribution; N is the total number of observations; x n is the nth observation in the data set.

[0031] The posterior distribution of the model obtained in step S3 is updated based on Bayesian inference in step S4, which specifically includes the following steps:

[0032] Based on Bayesian inference, the following formula is used as the posterior distribution:

[0033] p(μ|X)=N(μ|μ N ,(δ N ) 2 )

[0034]

[0035]

[0036] Where p(μ|X) is the posterior distribution; μ N is the expected value of the posterior distribution; (δ N ) 2 is the variance of the posterior distribution;

[0037] According to the obtained posterior distribution, the root mean square error of the parameters in the Bayesian model is calculated and the root mean square error is normalized; finally, the normalized value is used as the credibility of the parameter.

[0038] Step S5, based on the posterior distribution obtained in step S4, generates fingerprint data of the power inspection terminal, which specifically includes the following steps:

[0039] Based on the credibility of the parameters in the Bayesian model obtained in step S4, several parameters with the lowest credibility parameters are selected as input features for generating fingerprints;

[0040] The obtained input features are input into the transformer model to generate fingerprint data of the power inspection terminal.

[0041] Step S6, based on the fingerprint data obtained in step S5, completes the actual identity authentication of the power inspection terminal, which specifically includes the following steps:

[0042] Obtain inspection data information of target power inspection terminals in real time;

[0043] Generate a fingerprint sequence of the target power inspection terminal based on the acquired inspection data information;

[0044] Match the fingerprint sequence of the target power inspection terminal with the fingerprint data of the power inspection terminal generated in step S5:

[0045] If the match is successful, the target power inspection terminal is determined to be a legitimate terminal;

[0046] If the matching is unsuccessful, the target power inspection terminal is determined to be an illegal terminal.

[0047] The present invention also provides a system for implementing the identity authentication method for power inspection terminals, comprising a data acquisition module, a data processing module, a model construction module, a distribution update module, a fingerprint generation module and an identity authentication module; the data acquisition module, the data processing module, the model construction module, the distribution update module, the fingerprint generation module and the identity authentication module are connected in series in sequence; the data acquisition module is used to acquire historical inspection data information of the power system and upload the data information to the data processing module; the data processing module is used to perform data preprocessing on the acquired data information according to the received data information to obtain a data set, and upload the data information to the model construction module; the model construction module is used to generate a fingerprint based on the received data information. According to the received data information, a fingerprint parameter probability model of the power inspection terminal is set, and based on the obtained data set, the relationship between the device status and the data is represented, and the data information is uploaded to the distribution update module; the distribution update module is used to perform a posterior distribution update on the obtained model based on the received data information and Bayesian inference, and upload the data information to the fingerprint generation module; the fingerprint generation module is used to generate the fingerprint data of the power inspection terminal according to the received data information and the obtained posterior distribution, and upload the data information to the identity authentication module; the identity authentication module is used to complete the actual identity authentication of the power inspection terminal according to the received data information and the obtained fingerprint data.

[0048] The identity authentication method and system for electric power inspection terminals provided by the present invention process, model and generate fingerprints of historical inspection data of the power system, and identify the electric power inspection terminal based on the generated fingerprint of the electric power inspection equipment; therefore, the present invention can not only realize the identity authentication of the electric power inspection terminal, but also has higher reliability and better security. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Figure 1 Schematic diagram of the process flow of the present invention.

[0050] Figure 2 Schematic diagram of the functional modules of the system of the present invention. DETAILED DESCRIPTION

[0051] like Figure 1 The figure shows a flow chart of the method of the present invention: The identity authentication method for a power inspection terminal disclosed in the present invention comprises the following steps:

[0052] S1. Obtain historical inspection data of the power system; specifically, the steps include:

[0053] Obtain historical inspection data information of the power system;

[0054] The inspection data information includes static data and dynamic data;

[0055] The static data includes the ID of the main control chip of the power inspection terminal, the sensor serial number of the power inspection terminal, the operating system version of the power inspection terminal, the designated driver list of the power inspection terminal, and the designated application list of the power inspection terminal;

[0056] Dynamic data includes network characteristic data and inspection characteristic data; among which, network characteristic data includes transmission rate and transmission delay; inspection characteristic data includes PID control parameters, inspection mode and GPS positioning data;

[0057] S2. Preprocess the data information obtained in step S1 to obtain a data set; specifically, the steps include:

[0058] Preprocessing includes missing data processing, abnormal data processing and data unification;

[0059] Missing data processing: If the amount of missing data is less than or equal to the set value, interpolation is used to fill the data; if the amount of missing data is greater than the set value, the corresponding data is directly discarded;

[0060] Abnormal data processing: Abnormal data is detected using a detection method and directly deleted; the detection method includes the Z-Score method or the IQR method;

[0061] Data unification: The data obtained after missing data processing and abnormal data processing are unified to the standard time axis to achieve synchronization of the corresponding time of the data; for the dynamic inspection data in the data, the data is divided into several segments and stored separately; the data segmentation specifically includes sliding window segmentation;

[0062] S3 sets the power inspection terminal fingerprint parameter probability model, and based on the data set obtained in step S2, represents the relationship between the device status and the data; specifically comprising the following steps:

[0063] The following normal distribution model is used as the prior distribution, and the initial parameter model is constructed:

[0064] p(μ)=N(μ1|μ0,(δ0) 2 )

[0065] Where p(μ) is the normal distribution of parameter μ; N() is the symbol for normal distribution; μ1 is a random variable; μ0 is the expected value of normal distribution; (δ0) 2 is the variance of the normal distribution; choosing such a weak information prior provides a certain degree of flexibility while avoiding introducing too much prior bias into the model parameters;

[0066] According to the prior distribution, the following formula is used as the likelihood function:

[0067]

[0068] Where p(X|μ) is the likelihood function; X is the observed quantity; μ is the expected value of the prior normal distribution; δ is the standard deviation of the prior normal distribution; N is the total number of observations; x n is the nth observation in the data set;

[0069] S4. Based on Bayesian inference, update the posterior distribution of the model obtained in step S3; specifically, the steps include:

[0070] Based on Bayesian inference, the following formula is used as the posterior distribution:

[0071] p(μ|X)=N(μ|μ N ,(δ N ) 2 )

[0072]

[0073] Where p(μ|X) is the posterior distribution; μ N is the expected value of the posterior distribution; (δ N ) 2 is the variance of the posterior distribution;

[0074] According to the obtained posterior distribution, the root mean square error of the parameters in the Bayesian model is calculated and the root mean square error is normalized; finally, the normalized value is used as the credibility of the parameter;

[0075] In specific implementation, the reliability of parameters with large errors is low, and the reliability of parameters with small errors is high. The Z-score method can be used for normalization. After normalization, the smaller the reliability value, the higher the reliability of the parameter.

[0076] S5. Generate fingerprint data of the power inspection terminal according to the posterior distribution obtained in step S4; specifically comprising the following steps:

[0077] Based on the credibility of the parameters in the Bayesian model obtained in step S4, several parameters with the lowest credibility parameters are selected as input features for generating fingerprints;

[0078] The obtained input features are input into the transformer model to generate fingerprint data of the power inspection terminal;

[0079] In practice, the input features are converted into high-dimensional vectors through an embedding layer, and position encoding is set to enable the model to capture the temporal order information in the input data. The input data is then passed through the Transformer model to gradually generate a fingerprint sequence. The decoder generates characters at each position until the fingerprint sequence is complete. During training, the cross-entropy loss function is used to measure the difference between the actual sequence and the predicted sequence.

[0080] S6. According to the fingerprint data obtained in step S5, the actual identity authentication of the power inspection terminal is completed; specifically comprising the following steps:

[0081] Obtain inspection data information of target power inspection terminals in real time;

[0082] Generate a fingerprint sequence of the target power inspection terminal based on the acquired inspection data information; in specific implementation, classification methods such as machine learning can be used to classify the fingerprint sequence, and then match it based on the classification results;

[0083] Match the fingerprint sequence of the target power inspection terminal with the fingerprint data of the power inspection terminal generated in step S5:

[0084] If the match is successful, the target power inspection terminal is determined to be a legitimate terminal;

[0085] If the matching is unsuccessful, the target power inspection terminal is determined to be an illegal terminal.

[0086] The present invention applies Bayesian inference and transformer identity recognition technology to the authentication of power inspection terminals. This authentication is performed based on universal hardware data and inspection interaction data unique to the power inspection terminals. The authentication process only requires communication to obtain the power inspection terminal data, without involving fingerprint transmission, ensuring the security of fingerprint data. The present invention constructs a parameter credibility model using Bayesian inference technology. When generating a sequence, the top n highly reliable feature parameters are automatically selected as the basis for constructing the fingerprint sequence, reducing labor costs and the risk of manually adjusting model parameters. Compared to solutions that verify device identity based on fixed hardware data, the present invention integrates dynamic data into the fingerprint generation and verification scheme. Therefore, the present invention can adapt to dynamic changes in power inspection terminal devices, such as system upgrades, application changes, and inspection task changes. Bayesian inference technology is used to dynamically update the credibility of feature parameters, selecting highly reliable parameters to generate fingerprint sequences. Finally, the fingerprint sequence generated for the power inspection terminal by the present invention is dynamically changeable, and fingerprint recognition is achieved through classification technology. Even if a malicious attacker obtains the fingerprint hardware sequence, they cannot forge the existing sequence to obtain important business data from the power inspection terminal device.

[0087] like Figure 2 The figure shows a functional module diagram of the system of the present invention: the system disclosed by the present invention for realizing the identity authentication method for the power inspection terminal includes a data acquisition module, a data processing module, a model construction module, a distribution update module, a fingerprint generation module and an identity authentication module; the data acquisition module, the data processing module, the model construction module, the distribution update module, the fingerprint generation module and the identity authentication module are connected in series in sequence; the data acquisition module is used to obtain historical inspection data information of the power system and upload the data information to the data processing module; the data processing module is used to perform data preprocessing on the acquired data information according to the received data information to obtain a data set, and upload the data information to the model construction module; the model The model construction module is used to set the fingerprint parameter probability model of the power inspection terminal according to the received data information, and based on the obtained data set, express the relationship between the device status and the data, and upload the data information to the distribution update module; the distribution update module is used to perform posterior distribution update on the obtained model based on the received data information and Bayesian inference, and upload the data information to the fingerprint generation module; the fingerprint generation module is used to generate the fingerprint data of the power inspection terminal according to the received data information and the obtained posterior distribution, and upload the data information to the identity authentication module; the identity authentication module is used to complete the actual identity authentication of the power inspection terminal according to the received data information and the obtained fingerprint data.

Claims

1. An identity authentication method for a power inspection terminal, comprising the following steps: S1. Obtain historical inspection data information of the power system; S2. Preprocess the data information obtained in step S1 to obtain a data set; S3 sets the power inspection terminal fingerprint parameter probability model, and based on the data set obtained in step S2, represents the relationship between the device status and the data; specifically comprising the following steps: The following normal distribution model is used as the prior distribution, and the initial parameter model is constructed: p(μ)=N(μ1|μ0,(δ0) 2 ) Where p(μ) is the normal distribution of parameter μ; N() is the symbol for normal distribution; μ1 is a random variable; μ0 is the expected value of normal distribution; (δ0) 2 is the variance of the normal distribution; According to the prior distribution, the following formula is used as the likelihood function: Where p(X|μ) is the likelihood function; X is the observed quantity; μ is the expected value of the prior normal distribution; δ is the standard deviation of the prior normal distribution; N is the total number of observations; x n is the nth observation in the data set; S4. Based on Bayesian inference, update the posterior distribution of the model obtained in step S3; S5. Generate fingerprint data of the power inspection terminal according to the posterior distribution obtained in step S4; S6. Based on the fingerprint data obtained in step S5, the actual identity authentication of the power inspection terminal is completed.

2. The identity authentication method for a power inspection terminal according to claim 1, characterized in that The acquisition of historical inspection data information of the power system described in step S1 specifically includes the following steps: Obtain historical inspection data information of the power system; The inspection data information includes static data and dynamic data; The static data includes the ID of the main control chip of the power inspection terminal, the sensor serial number of the power inspection terminal, the operating system version of the power inspection terminal, the designated driver list of the power inspection terminal, and the designated application list of the power inspection terminal; Dynamic data includes network characteristic data and inspection characteristic data; among them, network characteristic data includes transmission rate and transmission delay; inspection characteristic data includes PID control parameters, inspection mode and GPS positioning data.

3. The identity authentication method for a power inspection terminal according to claim 2 is characterized in that The pre-processing described in step S2 specifically includes the following steps: Preprocessing includes missing data processing, abnormal data processing and data unification; Missing data processing: If the amount of missing data is less than or equal to the set value, interpolation is used to fill the data; if the amount of missing data is greater than the set value, the corresponding data is directly discarded; Abnormal data processing: Abnormal data is detected using a detection method and directly deleted; the detection method includes the Z-Score method or the IQR method; Data unification: The data obtained after missing data processing and abnormal data processing are unified into a standard time axis to achieve synchronization of the time corresponding to the data; for the dynamic inspection data in the data, the data is divided into several segments and stored separately; the data segmentation specifically includes sliding window segmentation.

4. The identity authentication method for a power inspection terminal according to claim 3 is characterized in that The posterior distribution of the model obtained in step S3 is updated based on Bayesian inference in step S4, which specifically includes the following steps: Based on Bayesian inference, the following formula is used as the posterior distribution: p(μ|X)=N(μ|μ N ,(d N ) 2 ) Where p(μ|X) is the posterior distribution; μ N is the expected value of the posterior distribution; (δ N ) 2 is the variance of the posterior distribution; According to the obtained posterior distribution, the root mean square error of the parameters in the Bayesian model is calculated and the root mean square error is normalized; Finally, the normalized value is used as the credibility of the parameter.

5. The identity authentication method for a power inspection terminal according to claim 4 is characterized in that Step S5, based on the posterior distribution obtained in step S4, generates fingerprint data of the power inspection terminal, specifically including the following steps: Based on the credibility of the parameters in the Bayesian model obtained in step S4, several parameters with the lowest credibility parameters are selected as input features for generating fingerprints; The obtained input features are input into the transformer model to generate fingerprint data of the power inspection terminal.

6. The identity authentication method for a power inspection terminal according to claim 5, characterized in that Step S6, based on the fingerprint data obtained in step S5, completes the actual identity authentication of the power inspection terminal, which specifically includes the following steps: Obtain inspection data information of target power inspection terminals in real time; Generate a fingerprint sequence of the target power inspection terminal based on the acquired inspection data information; Match the fingerprint sequence of the target power inspection terminal with the fingerprint data of the power inspection terminal generated in step S5: If the match is successful, the target power inspection terminal is determined to be a legitimate terminal; If the matching is unsuccessful, the target power inspection terminal is determined to be an illegal terminal.

7. A system for implementing the identity authentication method for a power inspection terminal according to any one of claims 1 to 6, characterized in that It includes a data acquisition module, a data processing module, a model building module, a distribution update module, a fingerprint generation module and an identity authentication module; the data acquisition module, the data processing module, the model building module, the distribution update module, the fingerprint generation module and the identity authentication module are connected in series in sequence; the data acquisition module is used to obtain historical inspection data information of the power system and upload the data information to the data processing module; The data processing module is used to perform data preprocessing on the acquired data information according to the received data information to obtain a data set, and upload the data information to the model construction module; The model building module is used to set the fingerprint parameter probability model of the power inspection terminal according to the received data information, and based on the obtained data set, it represents the relationship between the device status and data, and uploads the data information to the distribution update module; The distribution update module is used to update the posterior distribution of the obtained model based on the received data information and Bayesian inference, and upload the data information to the fingerprint generation module; The fingerprint generation module is used to generate fingerprint data of the power inspection terminal based on the received data information and the obtained posterior distribution, and upload the data information to the identity authentication module; The identity authentication module is used to complete the identity authentication of the actual power inspection terminal based on the received data information and the obtained fingerprint data.

Images