Photovoltaic power station 5G network security protection equipment

Abstract

The invention, which relates to the technical field of network security protection, discloses a photovoltaic power station 5G network security protection device comprising the following components: a data acquisition module, an intelligent fusion analysis and verification system and a protection mechanism triggering module. According to the invention, by integrating an intelligent fusion analysis and verification system, multi-dimensional data such as photovoltaic power station operation data, 5G network characteristic parameters, external environment factors and personnel operation behaviors can be comprehensively collected and deeply analyzed, and a potential association relationship and a mutual influence mechanism among different factors can be accurately mined; meanwhile, a dynamic association rule self-adaptive updating mechanism can adjust the association rule in real time according to newly collected data and a mined new association relationship, the accuracy and timeliness of an analysis result are ensured, the credibility of the data is improved, and network security risks can be found and coped with in time.

Description

Technical Field

[0001] The present invention relates to the field of network security protection technology, and specifically to a 5G network security protection device for a photovoltaic power station. Background Art

[0002] With the rapid development of the photovoltaic industry and the widespread application of 5G technology, photovoltaic power stations have gradually become intelligent and networked. The operation of photovoltaic power stations depends on efficient and secure 5G networks for data transmission, remote monitoring and fault warning. However, the 5G network of photovoltaic power stations faces a complex and changeable network environment, including massive data from different devices and systems, as well as the joint influence of multi-dimensional factors such as external environmental factors and personnel operation behaviors. Therefore, how to effectively integrate and analyze these data, ensure the credibility of the data, and promptly discover network security risks has become an important issue in the 5G network security protection of photovoltaic power stations.

[0003] Traditional 5G network security protection measures for photovoltaic power stations have obvious deficiencies in data association analysis and data credibility verification. First, most traditional systems rely on fixed association rules for data analysis, which is difficult to adapt to dynamic changes in data, resulting in inaccurate analysis results and affecting network security protection effects. Secondly, traditional methods lack comprehensive and in-depth analysis methods when verifying data credibility, and it is difficult to effectively identify anomalies or deviations in the data, which may lead to decisions based on erroneous data. In addition, traditional systems lack an effective fusion mechanism when processing data from different devices and systems, resulting in serious data silos and the inability to fully realize the value of data.

[0004] In view of the above problems, it is necessary to optimize the existing 5G network security protection equipment of photovoltaic power stations, integrate data association analysis and data credibility verification functions, and introduce dynamic association rule adaptive update mechanism to comprehensively improve the security of 5G network of photovoltaic power stations. Therefore, it is of great significance to develop a 5G network security protection equipment for photovoltaic power stations that can comprehensively realize the above characteristics. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the prior art and provide a 5G network security protection device for a photovoltaic power station. It can integrate data correlation analysis and data credibility verification functions and introduce a dynamic correlation rule adaptive update mechanism to comprehensively improve the security of the 5G network of the photovoltaic power station and the credibility of the data, analysis accuracy and adaptability to environmental changes. Specifically, the data acquisition module comprehensively collects multi-dimensional data, and uses an intelligent fusion analysis and verification system to perform deep fusion analysis and verification, accurately mines the correlation relationship and mutual influence mechanism between different factors, and dynamically adjusts the security protection level of the data according to the analysis and verification results. At the same time, the dynamic association rule adaptive update mechanism can continuously monitor new data and update association rules to adapt to data patterns and changes, thereby ensuring the timeliness and effectiveness of network security protection.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a 5G network security protection device for a photovoltaic power station, the device comprising the following components: a data acquisition module, an intelligent fusion analysis and verification system, and a protection mechanism triggering module;

[0007] The data acquisition module includes a photovoltaic power station operation data acquisition unit, a 5G network characteristic parameter acquisition unit, an external environmental factor acquisition unit and a personnel operation behavior acquisition unit;

[0008] The photovoltaic power station's own operation data collection unit establishes a physical connection or data communication link with the sensors and monitoring devices of various equipment in the photovoltaic power station to achieve data interaction, and collects the photovoltaic module power generation efficiency and inverter status data in real time according to the preset sampling frequency. The 5G network characteristic parameter collection unit uses network monitoring equipment to connect to the photovoltaic power station's 5G network, and monitors and collects the 5G network's signal strength, frequency band usage and data transmission rate parameters in real time. The external environmental factor collection unit is equipped with meteorological sensors and electromagnetic interference monitors. The meteorological sensors collect weather conditions in real time, and the electromagnetic interference monitor continuously monitors the surrounding electromagnetic interference intensity. The personnel operation behavior collection unit embeds a monitoring program in the photovoltaic power station's 5G network management system or uses a network audit tool to comprehensively monitor the operation records of the operation and maintenance personnel logging into the photovoltaic power station's 5G network management system, record the operation behavior data of the operation and maintenance personnel's login frequency and operation instruction type, and organize the collected data and transmit it to the intelligent fusion analysis and verification system.

[0009] The intelligent fusion analysis and verification system includes a data preprocessing submodule, a fusion analysis and verification submodule, a risk determination and protection level adjustment submodule, and a dynamic association rule adaptive update submodule;

[0010] The data preprocessing submodule performs data cleaning operations on the received data of various factors, and converts the data from different sources into a unified internal data format. The fusion analysis and verification submodule uses big data analysis technology and intelligent algorithms to perform deep fusion analysis and verification on the preprocessed data of various factors. In terms of mining the potential correlation and mutual influence mechanism between different factors, a data association model is constructed. Based on historical data and real-time collected data, the correlation coefficients between different factors are analyzed. The association rule mining algorithm is used to mine the logical correlation between different factors, and complex nonlinear relationships are modeled to accurately capture the potential influence mechanism between different factors. In terms of multi-source data fusion verification function, cross-analysis is performed on data from different devices and systems. By comparing the same or related data items in different data sources, the differences are analyzed, and data comparison operations are performed. According to the pre-set physical The submodule of risk determination and protection level adjustment calculates the risk probability value through the risk assessment model and compares it with the threshold value to determine whether there is a network security risk according to the mined association relationship, mutual influence mechanism and data credibility, combined with the set network security risk determination standard. At the same time, different protection measures are set according to the data credibility to dynamically adjust the security protection level of the data. The submodule of dynamic association rule adaptive update continuously monitors the newly collected data and the new association relationship mined by the fusion analysis and verification submodule. With the continuous collection and analysis of new data, when the environmental factors or the operating status of the equipment change, the update of the association rules is triggered. Specifically, the triggering condition for the update of the association rules is set. When the correlation coefficient between a certain factor and other factors in the newly collected data changes by more than the threshold value R x , using machine learning-based update algorithms to update association rules. Specifically, according to new data features and new association relationships mined, the weights and thresholds of each factor in the association rules are adjusted to adapt to new data patterns and changes;

[0011] The protection mechanism triggering module includes a data encryption strengthening unit, a network access policy adjustment unit and an early warning notification unit;

[0012] After the data encryption enhancement unit receives the signal that the intelligent fusion analysis and verification system determines that there is a network security risk or detects a signal that the data credibility is abnormal, it automatically starts the data encryption enhancement program. The network access policy adjustment unit adjusts the access policy of the 5G network accordingly based on the specific situation of the risk or the situation of the data credibility abnormality. The early warning notification unit sends an early warning notification to the operation and maintenance personnel, and the notification content contains specific information about the network security risk or data credibility abnormality.

[0013] Furthermore, in the intelligent fusion analysis and verification system, the fusion analysis and verification submodule analyzes the correlation coefficients between different factors based on historical data and real-time collected data, and the calculation formula is: Among them, x i and i Represent the specific observation values ​​in the data series of two different factors. and The data sequence x is i and i The average value of n represents the data sequence x i and i The number of data points in .

[0014] Furthermore, in the intelligent fusion analysis and verification system, the fusion analysis and verification submodule uses an association rule mining algorithm to mine the logical association relationship between different factors. The algorithm formula is: Among them, R ij represents the comprehensive correlation strength between factor i and factor j, n represents the number of factors involved in calculating the correlation, including the operating data of the PV power station itself, 5G network characteristic parameters, external environmental factors, and various factors of personnel operation behavior, and w k is the weight coefficient of the kth factor when calculating the correlation relationship, f k (x ik ,y jk ) is the correlation function corresponding to the kth factor, x ik is the specific observed value of factor i with respect to the kth factor, y jk is the specific observed value of factor j with respect to the kth factor.

[0015] Furthermore, in the intelligent fusion analysis and verification system, by comparing the same or related data items in different data sources, analyzing their differences, performing data comparison operations, and judging whether the data conforms to normal conditions according to pre-set physical laws and logical relationships, the formula is: Among them, C data represents the credibility of the data after comprehensive evaluation, M represents the number of data sources involved in data fusion verification, including the operating data of the photovoltaic power station itself, 5G network characteristic parameters, external environmental factors, and various factors of human operation behavior, a m is the weight coefficient of the mth data source when evaluating the data credibility, g m (z m ) is the credibility function corresponding to the m-th data source, which is used to describe the relationship between the data characteristics and credibility of the m-th data source. m It is the specific data value or data feature vector of the mth data source. By calculating the data credibility, it is determined whether the data conforms to the normal situation.

[0016] Furthermore, in the intelligent fusion analysis and verification system, the risk determination and protection level adjustment submodule calculates the risk probability value through the risk assessment model and compares it with the threshold to determine whether there is a network security risk. The algorithm formula of the risk assessment model is: Among them, P risk represents the probability value of network security risk after comprehensive judgment, L represents the number of factors involved in risk judgment, and b l is the weight coefficient of the lth factor in determining network security risks, μ l (v l ) is the membership function corresponding to the lth factor, v l is the specific observed value of the lth factor.

[0017] Furthermore, in the intelligent fusion analysis and verification system, the risk determination and protection level adjustment submodule sets different protection measures according to the data credibility to dynamically adjust the security protection level of the data. The protection level determination formula is: Among them, PL adjust Indicates the protection level after dynamic adjustment, C data is the data credibility, C high is the high confidence threshold. When the data confidence level C data When it is greater than or equal to this threshold, it means that the data credibility is high, P risk is the probability value of network security risk, P low is the low risk threshold. When the network security risk probability value P risk When it is less than or equal to this threshold, it means that the network security risk is low. mid is the medium risk threshold. When the network security risk probability value P risk Greater than P low When the network security risk probability value P is less than or equal to this threshold, it means that the network security risk is at a medium level. risk Greater than P mid , indicating that the network security risk is high.

[0018] Furthermore, in the intelligent fusion analysis and verification system, the dynamic association rule adaptive update submodule adjusts the weight and threshold of each factor in the association rule according to the new data features and the new association relationship mined. For the weight adjustment, the adjustment formula is: in, represents the updated weight value of factor i after the t+1th iteration, is the current weight value of factor i at the tth iteration, α is the learning rate parameter, Represents the loss function L with respect to the current weight The partial derivative of , for threshold update, the update formula is: Among them, T t+1 represents the updated threshold after the t+1th iteration, T t is the current threshold at iteration t, β is the learning rate, Represents the loss function L with respect to the current threshold T t The partial derivative of .

[0019] Furthermore, in the protection mechanism triggering module, after the data encryption strengthening unit receives a signal from the intelligent fusion analysis and verification system that a network security risk exists or a signal that data credibility is abnormal, it automatically starts the data encryption strengthening program and dynamically adjusts the encryption strength. The encryption algorithm determination method is: Among them, E upgrade Indicates the encryption algorithm to be upgraded based on the risk situation and data importance. The value can be AES, 3DES or RSA, which represent encryption algorithms of different strengths. risk is the probability value of network security risk, P low is the low risk threshold, P mid is the medium risk threshold. When the network security risk probability value P risk Greater than P low When the network security risk is less than or equal to this threshold, it means that the network security risk is at a medium level. data It is a data importance indicator, which is used to measure the importance of the current transmission data to the operation and decision-making of the photovoltaic power station. low Low data importance threshold, I mid is the data importance threshold. When the data importance index I dnta Greater than I low When the data importance index I is less than or equal to this threshold, it means that the data importance is at a medium level. dnta Greater than I mid , it indicates that the data is of high importance.

[0020] Furthermore, in the protection mechanism triggering module, the network access policy adjustment unit upgrades the encryption algorithm through the formula S encryptiom =S base +k·(P risk -P ref ) dynamically adjusts the encryption strength, where S encryptiom is the adjusted encryption strength, S bure is the basic encryption strength, P risk is the network security risk probability value, k is the encryption strength adjustment coefficient, P ref is the reference risk probability value.

[0021] Compared with the existing technology, this photovoltaic power station 5G network security protection equipment has the following beneficial effects:

[0022] 1. The present invention integrates an intelligent fusion analysis and verification system to comprehensively collect and deeply analyze multi-dimensional data such as photovoltaic power station operation data, 5G network characteristic parameters, external environmental factors, and personnel operation behavior, and accurately mine the potential correlation relationships and mutual influence mechanisms between different factors. At the same time, the dynamic association rule adaptive update mechanism can adjust the association rules in real time according to the newly collected data and the newly mined correlation relationships, ensuring the accuracy and timeliness of the analysis results, which not only improves the credibility of the data, but also can timely discover and respond to network security risks.

[0023] 2. The present invention uses data feature adaptive sensing and collection technology to accurately capture and transmit feature information of various factor data, such as fluctuation amplitude, frequency, change rate, etc., providing richer and more accurate data support for the intelligent fusion analysis and verification system, enabling it to more accurately analyze the impact of environmental factors or changes in equipment operating status on photovoltaic power station operation and network security. At the same time, the dynamic association rule adaptive update mechanism can adjust the association rules in time according to environmental changes or equipment status changes to ensure the accuracy and reliability of the analysis results.

[0024] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 This is a schematic diagram of the structure of a 5G network security protection device for a photovoltaic power station;

[0027] Figure 2 This is a process operation diagram of 5G network security protection equipment for photovoltaic power stations. DETAILED DESCRIPTION

[0028] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0029] Example

[0030] This embodiment describes in detail the application of a 5G network security protection device for photovoltaic power stations in the network security protection of large-scale desert photovoltaic power stations. Through the close cooperation of various modules, accurate data collection and analysis, and targeted protection mechanism triggering, the security protection capability of the power station's 5G network is effectively improved, ensuring the stable and safe operation of the power station in harsh environments and complex operation and maintenance conditions.

[0031] The data collection module includes a photovoltaic power station's own operation data collection unit, a 5G network characteristic parameter collection unit, an external environmental factor collection unit, and a personnel operation behavior collection unit. The photovoltaic power station's own operation data collection unit establishes a physical connection or data communication link with the sensors and monitoring devices of various equipment in the photovoltaic power station to achieve data interaction, and collects the photovoltaic module power generation efficiency and inverter status data in real time according to the preset sampling frequency. The 5G network characteristic parameter collection unit uses network monitoring equipment to connect to the photovoltaic power station's 5G network, and monitors and collects the 5G network's signal strength, frequency band usage, and data transmission rate parameters in real time. The external environmental factor collection unit is equipped with meteorological sensors and electromagnetic interference monitors. The meteorological sensors collect weather conditions in real time, and the electromagnetic interference monitor continuously monitors the surrounding electromagnetic interference intensity. The personnel operation behavior collection unit embeds a monitoring program in the photovoltaic power station's 5G network management system or uses a network audit tool to comprehensively monitor the operation records of the operation and maintenance personnel logging into the photovoltaic power station's 5G network management system, record the operation and maintenance personnel's login frequency and operation instruction type operation behavior data, and organize the collected human data and transmit it to the intelligent fusion analysis and verification system.

[0032] The intelligent fusion analysis and verification system includes a data preprocessing submodule, a fusion analysis and verification submodule, a risk determination and protection level adjustment submodule, and a dynamic association rule adaptive update submodule. The data preprocessing submodule performs comprehensive preliminary processing on the various factor data received (data transmitted from the photovoltaic power station's own operation data collection unit, the 5G network characteristic parameter collection unit, the external environmental factor collection unit, and the personnel operation behavior collection unit). First, the data cleaning operation is performed, and invalid data is eliminated according to the set data filtering rules (such as removing obviously erroneous data values, such as abnormally large or abnormally small data values ​​collected due to sensor failure, and removing duplicate data records to avoid interference with subsequent analysis) to ensure data accuracy. For example, in the data received this time, it was found that a certain sensor collected a temperature value that was obviously deviated from the normal range due to dust intrusion. It was eliminated through data cleaning operations, and then the format was unified. The data from different sources were converted according to the pre-defined standard data structure to ensure that all data have a consistent structure in the subsequent analysis and verification process. The format requirements are consistent to improve data processing efficiency and accuracy. For example, the data organized in JSON format transmitted by the photovoltaic power station's own operation data collection unit, the data in a specific network data format transmitted by the 5G network characteristic parameter collection unit, the data in a general environment data format transmitted by the external environmental factor collection unit, and the data in the form of a database table structure transmitted by the personnel operation behavior collection unit are all converted into a unified internal data format (such as a structured data format defined specifically for analysis of this system) so that subsequent modules can easily process and analyze these data. The fusion analysis and verification submodule uses advanced big data analysis technology and intelligent algorithms to perform deep fusion analysis and verification on various types of pre-processed factor data. First, a data association model is constructed to analyze the correlation coefficients between different factors based on historical data and real-time collected data. For example, the correlation between the power generation efficiency of photovoltaic modules and the signal strength of 5G networks is calculated by statistical analysis methods. Assuming that relevant data for the past month have been collected, the correlation coefficient between the two is calculated by calculation. The calculation formula is: Among them, x i and i Represents the specific observations in the data series of two different factors, for example, x i It may be a series of specific measurements of the efficiency of photovoltaic modules in a photovoltaic power plant over a certain period of time, while y i It may be a series of measured values ​​of 5G network signal strength in the corresponding time period. By performing correlation analysis on the data of these two groups of different factors, the potential correlation between them can be explored, so that the dynamic association rules can be updated according to the changes in this correlation. and The data sequence x is iand i The purpose of calculating the average value is to use the average level of the data series as a reference when calculating the correlation coefficient, to more accurately measure the deviation of each data point from the average level, and thus to more accurately reflect the degree of correlation between the two sets of data. For example, for the photovoltaic module power generation efficiency data series x i , It is the average power generation efficiency of all measured values ​​of the power generation efficiency in a specific time period. For the 5G network signal strength data sequence y i , It is the average signal strength of all measured values ​​of signal strength in the corresponding time period, and n represents the data sequence x i and iThe number of data points in the data set is calculated. Secondly, the association rule mining algorithm is used to mine the possible logical associations between different factors. For example, when the external environment temperature rises, by analyzing a large amount of historical data and real-time collected data, the logical association relationship that the power generation efficiency of photovoltaic modules may decrease and the data transmission rate of 5G network may also be affected is mined. Specifically, in the analysis process, the support threshold is set to 0.3 and the confidence threshold is set to 0.7. When these threshold conditions are met, it is considered that the association rules with practical significance are mined. Through the analysis of different combinations of factors, it is found that when the temperature rises to a certain extent (such as exceeding 40°C), in most cases (meeting the confidence threshold), the power generation efficiency of photovoltaic modules will decrease significantly, and the data transmission rate of 5G network will also decrease. This association relationship is of great significance for a comprehensive understanding of the operation status of the power station and the subsequent risk determination and formulation of protective measures. Furthermore, the complex nonlinear relationship is modeled by the neural network algorithm to accurately capture the potential influence mechanism between different factors. For example, for the relationship between abnormal operation behavior of personnel and the operation data and network characteristic parameters of photovoltaic power stations, a multi-layer perceptron neural network model is constructed to determine the number of input layer nodes according to the number of features of the input data. Here, the input The input data includes the operation instruction execution time, operation instruction type, login frequency of the operation and maintenance personnel, and the corresponding photovoltaic power station operation data (such as photovoltaic module power generation efficiency, inverter status, etc.) and 5G network characteristic parameters (such as signal strength, frequency band usage, etc.). Assume that after feature extraction and screening, the number of input layer nodes is determined to be 10, and the hidden layer can be set to multiple layers. The number of nodes in each layer is determined based on experience and experiments. Here it is set to two layers, the number of nodes in the first hidden layer is 8, and the number of nodes in the second hidden layer is 6. The number of nodes in the output layer is determined according to the number of targets to be predicted. The goal here is to predict the impact of abnormal operation behavior of personnel on the output layer. The impact of the overall operation of the power station (such as whether it will cause a significant drop in power generation efficiency, whether the network will be interrupted, etc.), so the number of output layer nodes is set to 2. By training a large amount of historical data and real-time collected data, the neural network model can learn the complex relationship between different factors and accurately capture the potential impact mechanism between abnormal personnel operation behavior and photovoltaic power station operation data and network characteristic parameters. For example, when the operation and maintenance personnel perform certain specific misoperations (such as misconfiguration of network parameters), the model can predict the possible decrease in photovoltaic module power generation efficiency and the fluctuation of 5G network signal strength. At the same time, through the formula Mining the comprehensive correlation strength R between factor i and factor j ij , where n represents the number of factors involved in calculating the correlation relationship, covering various factors in multiple dimensions, such as photovoltaic module power generation efficiency, 5G network signal strength, external environment temperature, and operation and maintenance personnel operation instruction execution time, etc., w kis the weight coefficient of the kth factor when calculating the association relationship. The importance of mining association relationships for different factors is set. It is assumed that for the photovoltaic module power generation efficiency factor, its weight coefficient w1 = 0.4, for the 5G network signal strength factor, its weight coefficient w2 = 0.3, for the external environment temperature factor, its weight coefficient w3 = 0.2, for the operation and maintenance personnel operation instruction execution time factor, its weight coefficient w4 = 0.1, f k (x ik ,y jk ) is the correlation function corresponding to the kth factor, which is used to describe the relationship between factor i and factor j at the kth factor level. For example, for the correlation function of photovoltaic module power generation efficiency and 5G network signal strength at the external environment temperature factor level, the corresponding functional relationship can be established based on actual data and physical principles, x ik is the specific observed value of factor i with respect to the kth factor. For example, when i is the power generation efficiency of the photovoltaic module and k is the external ambient temperature, x ik It is the external ambient temperature value corresponding to the photovoltaic module power generation efficiency value observed at a certain moment, y jk is the specific observation value of factor j with respect to the kth factor. This formula can comprehensively consider multiple factors, quantify the potential correlation between the same factors, and provide an accurate basis for mining potential correlation relationships. For example, by calculating the comprehensive correlation strength R between the power generation efficiency of photovoltaic modules and the signal strength of 5G networks considering multiple factors such as the external ambient temperature and the execution time of the operation instructions of the operation and maintenance personnel, ij is 0.5, indicating that there is a certain degree of correlation between the two, and this correlation is affected by a variety of factors. Finally, using the formula Evaluate the credibility of the data after comprehensive evaluation C data , where M represents the number of data sources involved in data fusion verification, including data from different devices and systems, such as photovoltaic module monitoring data, inverter data, weather station data, 5G network tester data, operation and maintenance personnel operation record data, etc., a m The weight coefficient of the mth data source in evaluating the data credibility is set according to the importance of different data sources on the data credibility. It is assumed that for the PV module monitoring data, the weight coefficient a1 is 0.3, for the inverter data, the weight coefficient a2 is 0.25, for the weather station data, the weight coefficient a3 is 0.2, for the 5G network tester data, the weight coefficient a4 is 0.15, for the operation record data of the operation and maintenance personnel, the weight coefficient a5 is 0.1, g m (z m) is the credibility function corresponding to the mth data source, which is used to describe the relationship between the data characteristics and credibility of the mth data source. For example, for photovoltaic module monitoring data, its credibility function can establish a corresponding functional relationship based on factors such as sensor accuracy and data acquisition stability. m is the specific data value or data feature vector of the mth data source. This formula can accurately reflect the credibility of the data and provide an important basis for subsequent judgment of data reliability. For example, the credibility of the data after comprehensive evaluation can be calculated. data The value is 0.65, indicating that the credibility of the overall data is at a medium level. This factor needs to be considered in the subsequent risk assessment and adjustment of protection measures. The risk assessment and protection level adjustment submodule determines whether there is a current network security risk based on the association, mutual influence mechanism and data credibility mined by the fusion analysis and verification submodule, combined with the pre-set network security risk assessment criteria (such as setting risk level thresholds corresponding to different combinations of factors). For example, in the case of extreme high temperature weather and abnormal human operation behavior, the formula Calculate the network security risk probability value P risk It has increased significantly and exceeded the medium risk threshold. L represents the number of factors involved in risk determination, including various factors related to network security risks, such as the strength of correlation and the degree of data deviation. It is assumed that the correlation strength between the power generation efficiency of photovoltaic modules and the signal strength of 5G network, the degree of deviation between the power generation efficiency data of photovoltaic modules and the normal level, the degree of deviation between the 5G network signal strength data and the normal level, and the abnormality evaluation value of the execution time of the operation and maintenance personnel's operation instructions are considered. That is, L = 4, b l is the weight coefficient of the first factor in determining network security risks. It is set according to the importance of different factors in risk determination. Assume that for the factor of the strength of the correlation between the power generation efficiency of photovoltaic modules and the 5G network signal strength, its weight coefficient b1 = 0.3, and for the factor of the degree of deviation between the power generation efficiency data of photovoltaic modules and the normal level, its weight coefficient b2 = 0.25. The network security risk probability value P risk Significantly increased, exceeding the medium risk threshold, according to the formula Dynamically adjust the data security protection level to PL3 and take corresponding enhanced protection measures, including C duta is the data credibility, where C duta =0.65, P risk is the probability value of network security risk, which has been calculated to exceed the medium risk threshold, C high is the high confidence threshold, which is determined based on the requirements of the photovoltaic power station for data accuracy and reliability and the analysis of historical data. Assume that C high =0.8, P lowis the low risk threshold, which is determined based on the tolerance of the photovoltaic power station to network security risks and the analysis of historical data. Assuming that P low =0.2, P mid is the medium risk threshold, which is also determined based on the tolerance of the photovoltaic power station to network security risks and the analysis of historical data. Assuming P mid =0.5, so according to the above conditions, C data ≥C high And P risk >P mid Therefore, the protection level is adjusted to PL3. The dynamic association rule adaptive update submodule continuously monitors the newly collected data and the new association relationships mined by the fusion analysis and verification submodule. With the continuous collection and analysis of new data, when the environmental factors or equipment operation status change, the update of the association rules is triggered. Specifically, the triggering condition for the update of the association rules is set. When the correlation coefficient between a factor and other factors in the newly collected data changes by more than the threshold R x , an update algorithm based on machine learning is used to update the association rules. Specifically, according to the new data features and the new association relationships mined, the weights and thresholds of each factor in the association rules are adjusted to adapt to the new data patterns and changes. For weight adjustment, the adjustment formula is: in, represents the updated weight value of factor i after the t+1th iteration, is the current weight value of factor i at the tth iteration, α is the learning rate parameter, Represents the loss function L with respect to the current weight The partial derivative of , for threshold update, the update formula is: Among them, T t+1 represents the updated threshold after the t+1th iteration, T t is the current threshold at iteration t, β is the learning rate, Represents the loss function L with respect to the current threshold T t The partial derivative of .

[0033] The protection mechanism trigger module includes a data encryption enhancement unit, a network access policy adjustment unit and an early warning notification unit. The data encryption enhancement unit upgrades the original AES encryption algorithm to a 3DES encryption algorithm according to the preset encryption policy due to the increased risk level. While upgrading the encryption algorithm, the formula S encryptiom =S base +k·(P risk -P ref ) dynamically adjusts the encryption strength to ensure that data is not stolen or altered during transmission, thus ensuring data security in high-risk situations. bureis the basic encryption strength, assuming it is set to a fixed value, such as S base =100 (the value here is only an example, and can be actually determined according to the initial setting of the encryption algorithm, etc.), k is the encryption strength adjustment coefficient, which is set according to the requirements for data security and the security strategy of the photovoltaic power station, assuming that k = 0.5; P risk is the probability value of network security risk, which has been calculated to exceed the medium risk threshold, P ref As a reference risk probability value, it can be set based on historical data or experience. Assume that P ref =0.3, through calculation we can get S encryption =100+0.5·(P risk -0.3), the specific P risk The adjusted encryption strength can be obtained by substituting the value, thereby realizing dynamic adjustment of data encryption strength to adapt to high-risk network environments. The network access policy adjustment unit restricts the access of relevant IP addresses through the network firewall for suspicious IP address access, and adds identity authentication links such as fingerprint verification and facial recognition. For example, when a suspicious IP address is detected trying to frequently access the key network resources of the power station, the network firewall immediately blocks the access request of the IP address, and pops up the fingerprint verification and facial recognition window to require the visitor to authenticate the identity. Only the verified visitor can continue to access the relevant resources. At the same time, according to the data credibility, the accessible scope of some low-credibility data is limited, and only specific operation and maintenance personnel roles are allowed to access the relevant data, effectively preventing unauthorized access and data leakage risks. For example, for those data whose credibility is lower than a certain threshold (such as C data <0.5), such as some temporarily recorded equipment debugging parameters, etc., which are only accessible to operation and maintenance personnel with high-level permissions (such as the technical supervisor of the power station, etc.), and ordinary operation and maintenance personnel cannot access these data, thereby ensuring the security and confidentiality of the data. The early warning notification unit determines that the warning level is high risk and immediately sends an early warning notification to all operation and maintenance personnel via SMS. The notification content details the existing network security risks, including the impact of extreme high temperature weather on the power generation efficiency of photovoltaic modules and 5G networks, abnormal operating behaviors of personnel, and adjustments to data encryption and network access strategies, so that the operation and maintenance personnel can take corresponding measures in time to deal with them, such as on-site inspection of equipment, troubleshooting, etc. The operation and maintenance personnel can clearly understand the current status of the power station and the measures that need to be taken, so as to effectively respond to network security risks and ensure the normal operation of the power station.

[0034] In summary, in the above embodiments, through the close cooperation of various modules, accurate data collection and analysis, and targeted protection mechanism triggering, the security protection capability of the power station 5G network is effectively improved, and the stable and safe operation of the power station in harsh environments and complex operation and maintenance conditions is ensured, which fully reflects the practicality and effectiveness of the present invention in this specific application field.

[0035] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A 5G network security protection device for a photovoltaic power station, characterized in that: The system includes the following components: data acquisition module, intelligent fusion analysis and verification system and protection mechanism triggering module; The data acquisition module includes a photovoltaic power station operation data acquisition unit, a 5G network characteristic parameter acquisition unit, an external environmental factor acquisition unit and a personnel operation behavior acquisition unit; The photovoltaic power station's own operation data collection unit establishes a physical connection or data communication link with the sensors and monitoring devices of various equipment in the photovoltaic power station to achieve data interaction, and collects the photovoltaic module power generation efficiency and inverter status data in real time according to the preset sampling frequency. The 5G network characteristic parameter collection unit uses network monitoring equipment to connect to the photovoltaic power station's 5G network, and monitors and collects the 5G network's signal strength, frequency band usage and data transmission rate parameters in real time. The external environmental factor collection unit is equipped with meteorological sensors and electromagnetic interference monitors. The meteorological sensors collect weather conditions in real time, and the electromagnetic interference monitor continuously monitors the surrounding electromagnetic interference intensity. The personnel operation behavior collection unit embeds a monitoring program in the photovoltaic power station's 5G network management system or uses a network audit tool to comprehensively monitor the operation records of the operation and maintenance personnel logging into the photovoltaic power station's 5G network management system, record the operation behavior data of the operation and maintenance personnel's login frequency and operation instruction type, and organize the collected data and transmit it to the intelligent fusion analysis and verification system. The intelligent fusion analysis and verification system includes a data preprocessing submodule, a fusion analysis and verification submodule, a risk determination and protection level adjustment submodule, and a dynamic association rule adaptive update submodule; The data preprocessing submodule performs data cleaning operations on the received data of various factors, and converts the data from different sources into a unified internal data format. The fusion analysis and verification submodule uses big data analysis technology and intelligent algorithms to perform deep fusion analysis and verification on the preprocessed data of various factors. In terms of mining the potential correlation and mutual influence mechanism between different factors, a data association model is constructed. Based on historical data and real-time collected data, the correlation coefficients between different factors are analyzed. The association rule mining algorithm is used to mine the logical correlation between different factors, and complex nonlinear relationships are modeled to accurately capture the potential influence mechanism between different factors. In terms of multi-source data fusion verification function, cross-analysis is performed on data from different devices and systems. By comparing the same or related data items in different data sources, the differences are analyzed, and data comparison operations are performed. According to the pre-set physical The submodule of risk determination and protection level adjustment calculates the risk probability value through the risk assessment model and compares it with the threshold value to determine whether there is a network security risk according to the mined association relationship, mutual influence mechanism and data credibility, combined with the set network security risk determination standard. At the same time, different protection measures are set according to the data credibility to dynamically adjust the security protection level of the data. The submodule of dynamic association rule adaptive update continuously monitors the newly collected data and the new association relationship mined by the fusion analysis and verification submodule. With the continuous collection and analysis of new data, when the environmental factors or the operating status of the equipment change, the update of the association rules is triggered. Specifically, the triggering condition for the update of the association rules is set. When the correlation coefficient between a certain factor and other factors in the newly collected data changes by more than the threshold value R x , using machine learning-based update algorithms to update association rules. Specifically, according to new data features and new association relationships mined, the weights and thresholds of each factor in the association rules are adjusted to adapt to new data patterns and changes; The protection mechanism triggering module includes a data encryption strengthening unit, a network access policy adjustment unit and an early warning notification unit; After the data encryption enhancement unit receives the signal that the intelligent fusion analysis and verification system determines that there is a network security risk or detects a signal that the data credibility is abnormal, it automatically starts the data encryption enhancement program. The network access policy adjustment unit adjusts the access policy of the 5G network accordingly based on the specific situation of the risk or the situation of the data credibility abnormality. The early warning notification unit sends an early warning notification to the operation and maintenance personnel, and the notification content contains specific information about the network security risk or data credibility abnormality.

2. A photovoltaic power station 5G network security protection device according to claim 1, characterized in that: In the intelligent fusion analysis and verification system, the fusion analysis and verification submodule analyzes the correlation coefficients between different factors based on historical data and real-time collected data, and the calculation formula is: Among them, x i and i Represent the specific observation values ​​in the data series of two different factors. and They are the data series x i and i The average value of n represents the data sequence x i and i The number of data points in .

3. A photovoltaic power station 5G network security protection device according to claim 1, characterized in that: In the intelligent fusion analysis and verification system, the fusion analysis and verification submodule uses the association rule mining algorithm to mine the logical association relationship between different factors. The algorithm formula is: Among them, R ij represents the comprehensive correlation strength between factor i and factor j, n represents the number of factors involved in calculating the correlation, including the operating data of the PV power station itself, 5G network characteristic parameters, external environmental factors, and various factors of personnel operation behavior, and w k is the weight coefficient of the kth factor when calculating the correlation relationship, f k (x ik ,y jk ) is the correlation function corresponding to the kth factor, x ik is the specific observed value of factor i with respect to the kth factor, y jk is the specific observed value of factor j with respect to the kth factor.

4. A photovoltaic power station 5G network security protection device according to claim 1, characterized in that: In the intelligent fusion analysis and verification system, by comparing the same or related data items in different data sources, analyzing their differences, and performing data comparison operations, it is determined whether the data conforms to normal conditions based on pre-set physical laws and logical relationships. The formula is: Among them, C data represents the credibility of the data after comprehensive evaluation, M represents the number of data sources involved in data fusion verification, including the operating data of the photovoltaic power station itself, 5G network characteristic parameters, external environmental factors, and various factors of human operation behavior, a m is the weight coefficient of the mth data source when evaluating the data credibility, g m (z m ) is the credibility function corresponding to the m-th data source, which is used to describe the relationship between the data characteristics and credibility of the m-th data source. m It is the specific data value or data feature vector of the mth data source. By calculating the data credibility, it is determined whether the data conforms to the normal situation.

5. A photovoltaic power station 5G network security protection device according to claim 1, characterized in that: In the intelligent fusion analysis and verification system, the risk determination and protection level adjustment submodule calculates the risk probability value through the risk assessment model and compares it with the threshold to determine whether there is a network security risk. The algorithm formula of the risk assessment model is: Among them, P risk represents the probability value of network security risk after comprehensive judgment, L represents the number of factors involved in risk judgment, and b l is the weight coefficient of the lth factor in determining network security risks, μ l (v l ) is the membership function corresponding to the lth factor, v l is the specific observed value of the lth factor.

6. A photovoltaic power station 5G network security protection device according to claim 1, characterized in that: In the intelligent fusion analysis and verification system, the risk determination and protection level adjustment submodule sets different protection measures according to the data credibility to dynamically adjust the security protection level of the data. The protection level determination formula is: Among them, PL adjust Indicates the protection level after dynamic adjustment, C data is the data credibility, C high is the high confidence threshold. When the data confidence level C data When it is greater than or equal to this threshold, it means that the data credibility is high, P risk is the probability value of network security risk, P low is the low risk threshold. When the network security risk probability value P risk When it is less than or equal to this threshold, it means that the network security risk is low. mid is the medium risk threshold. When the network security risk probability value P risk Greater than P low When the network security risk probability value P is less than or equal to this threshold, it means that the network security risk is at a medium level. risk Greater than P mid , indicating that the network security risk is high.

7. A photovoltaic power station 5G network security protection device according to claim 1, characterized in that: In the intelligent fusion analysis and verification system, the dynamic association rule adaptive update submodule adjusts the weight and threshold of each factor in the association rule according to the new data features and the new association relationship mined. The adjustment formula for weight adjustment is: in, represents the updated weight value of factor i after the t+1th iteration, is the current weight value of factor i at the tth iteration, α is the learning rate parameter, Represents the loss function L with respect to the current weight The partial derivative of , for threshold update, the update formula is: Among them, T t+1 represents the updated threshold after the t+1th iteration, T t is the current threshold at iteration t, β is the learning rate, Represents the loss function L with respect to the current threshold T t The partial derivative of .

8. A photovoltaic power station 5G network security protection device according to claim 1, characterized in that: In the protection mechanism trigger module, after the data encryption enhancement unit receives a signal from the intelligent fusion analysis and verification system that a network security risk exists or a signal that data credibility is abnormal, it automatically starts the data encryption enhancement program and dynamically adjusts the encryption strength. The encryption algorithm determination method is as follows: Among them, E upgrade Indicates the encryption algorithm to be upgraded based on the risk situation and data importance. The value can be AES, 3DES or RSA, which represent encryption algorithms of different strengths. risk is the probability value of network security risk, P low is the low risk threshold, P mid is the medium risk threshold. When the network security risk probability value P risk Greater than P low When the network security risk is less than or equal to this threshold, it means that the network security risk is at a medium level. data: It is a data importance indicator, which is used to measure the importance of the current transmission data to the operation and decision-making of the photovoltaic power station. low Low data importance threshold, I mid is the data importance threshold. When the data importance index I dnta Greater than I low When the data importance index I is less than or equal to this threshold, it means that the data importance is at a medium level. dnta Greater than I mid , it indicates that the data is of high importance.

9. A photovoltaic power station 5G network security protection device according to claim 8, characterized in that: In the protection mechanism triggering module, the network access policy adjustment unit upgrades the encryption algorithm through the formula S encryptiom =S base +k·(P risk -P ref ) dynamically adjusts the encryption strength, where S encryptiom is the adjusted encryption strength, S base is the basic encryption strength, P risk is the network security risk probability value, k is the encryption strength adjustment coefficient, P ref is the reference risk probability value.

Images