A control system and method for a photovoltaic frequency-modulated inverter

Abstract

This invention discloses a control system and method for photovoltaic frequency regulation inverters, relating to the field of photovoltaic power plant control technology. The invention includes: real-time monitoring of the grid frequency by a central controller, calculating the total power requiring adjustment for the entire power plant, and issuing instructions to several dispatch centers; upon receiving the instructions, each dispatch center issues instructions to its subordinate inverters; simultaneously, a security center randomly groups all inverters into several security feedback groups; the system determines the control result based on data feedback from each inverter; upon receiving a fault warning result, it locates the corresponding security feedback group of the inverter, establishes a monitoring period, and uses a data model to confirm the cause of the fault warning within the monitoring period; after confirming the cause of the fault warning, it reports it to the administrator port for fault confirmation. This invention can quickly locate the attack source or fault node, shorten fault diagnosis and recovery time, and significantly improve the operational reliability of the photovoltaic power plant frequency regulation system.

Description

Technical Field

[0001] This invention relates to the field of photovoltaic power plant control technology, specifically a control system and method for a photovoltaic frequency modulation inverter. Background Technology

[0002] With the rapid growth of installed capacity of new energy power generation, represented by photovoltaics, the frequency stability problem of the power system has become increasingly prominent. Traditional thermal power generating units naturally provide frequency support to the power grid through the inertia characteristics of rotating rotors. However, photovoltaic power generation systems are connected to the grid through inverters, and their power output lacks a physical coupling relationship with the grid frequency, making them unable to respond autonomously to frequency fluctuations. Therefore, current technologies mostly adopt a central controller to monitor the grid frequency in real time, calculate the total power that the entire power station needs to adjust, and then send instructions to each inverter through a high-speed communication network to achieve group control and adjustment. At the overall system integration level, although frequency change response is achieved, the distributed single-machine control adopted in the current solution results in each inverter being independent of each other and lacking data linkage. At the same time, the large number of communication links also poses significant transmission security risks. When subjected to network security attacks, it is difficult to determine the source of the attack in a timely manner. Even when fault warnings occur during frequency adjustment, it is difficult to distinguish in time whether it is an external network security attack or an internal communication channel risk, resulting in slow investigation, slow diagnosis, and slow response, and a lack of rapid and effective technical means. Summary of the Invention

[0003] The purpose of this invention is to provide a control system and method for a photovoltaic frequency modulation inverter to solve the problems raised in the prior art.

[0004] To achieve the above objectives, the present invention provides the following technical solution: a method for regulating a photovoltaic frequency modulation inverter, the method comprising the following steps: S1. The central controller monitors the grid frequency in real time, calculates the total power that needs to be adjusted for the entire power station, and sends instructions to several dispatch centers. S2. Each dispatch center connects to several inverters. After receiving the instruction, the dispatch center issues the instruction to its subordinate inverters. At the same time, the safety center randomly groups all inverters into several safety feedback groups. Data interaction is established between inverters in each safety feedback group, but no data interaction is established between safety feedback groups. S3. The system determines the control result based on the data feedback of each inverter. When a fault warning result is received, the system locates the safety feedback group where the corresponding inverter is located, which is called the first safety feedback group. The system sends encryption and decryption commands to the communication link between several inverters in the first safety feedback group and the dispatch center. S4. Establish a monitoring period. During the monitoring period, acquire the transmission data of the communication link that issues encryption and decryption commands and the transmission data of other communication links in the first security feedback group in real time, and use a data model to confirm whether the fault warning is caused by an external network attack. S5. After confirming the cause of the fault warning, report it to the administrator port for fault confirmation.

[0005] According to the above technical solution, the central controller monitors the power grid frequency in real time and calculates the total power that needs to be adjusted for the entire power station, including: The central controller acquires the three-phase voltage signal at the grid connection point in real time through a high-precision voltage transformer. The three-phase voltage signals are converted using the Clarke transform: The Clarke transformed signals are obtained respectively. Axial components, Axis components; the Clarke transform obtained by Park transform. Axial components, The axial components are converted into the active / amplitude components and reactive / phase error components of the voltage vector in the rotating coordinate system. By controlling the reactive / phase error components of the voltage vector to 0 using a PI regulator, the phase-locked loop output angular frequency tracks the grid angular frequency, and the grid regulation frequency is calculated. By calculating the difference between the grid regulation frequency and the rated frequency, a frequency dead zone is set to correct the difference, forming an effective frequency deviation, and the total power that the power station needs to adjust is calculated:

[0006] in, The total power that the power plant needs to adjust; This refers to the upper limit of the adjustable capacity of the power plant; The adjustment rate; The rated frequency; This represents the effective frequency deviation.

[0007] According to the above technical solution, the system determines the control result based on the data feedback of each inverter. When a fault warning result is received, the system locates the safety feedback group to which the corresponding inverter belongs, denoted as the first safety feedback group, and sends encryption / decryption commands to the communication links between several inverters in the first safety feedback group and the dispatch center. The control results include control success and control failure. Control failure includes control delay and control command mismatch. When the feedback is control failure, a fault warning result is generated. When the system receives a fault warning result from a certain inverter, it locates the safety feedback group to which the corresponding inverter belongs and starts to send encryption and decryption commands to the communication link between several inverters and the dispatch center. The encryption / decryption instruction refers to the dispatch center using a hash algorithm to generate a unique digital digest for the instruction, signing it with a private key, and setting a public key and computing port at the inverter receiver. Upon receiving the instruction, the public key is used to verify the signature and confirm the integrity of the instruction. The computing port confirms the instruction data, which includes the instruction loss percentage and the amount of instruction data. At the same time, a timer is set at the inverter receiver to obtain the instruction delay time.

[0008] According to the above technical solution, the establishment of a monitoring period, during which the transmission data of the communication link issuing encryption / decryption commands and the transmission data of other communication links in the first security feedback group are acquired in real time, includes: The system establishes a time interval as a monitoring period. Within the monitoring period, it acquires the transmission data of the communication link that issues encryption and decryption commands. The transmission data includes command data and command delay time. At the same time, the inverter receiver of other communication links in the first security feedback group sets up a computing port to acquire the transmission data of other communication links in the first security feedback group.

[0009] According to the above technical solution, the fault warning method used to confirm whether the fault is caused by an external network attack includes: The monitoring period is randomly divided into several sub-intervals. For each sub-interval, the percentage of instruction loss in the instruction data is obtained. The system sets an instruction loss percentage threshold. If the instruction loss percentage exceeds the threshold, the current transmission is directly defined as abnormal and marked as abnormal transmission data. For each sub-interval, obtain the instruction delay time and instruction data volume; The first test data under a simulated environment is obtained. This simulated test data is generated by constructing a secure transmission channel in a laboratory, controlling variables to ensure a consistent network environment, and recording the instruction delay time for each transmission of instructions with added encryption / decryption commands. Several scattered points are formed in a coordinate system, with the data corresponding to different instruction data volumes as the x-axis and the data corresponding to the instruction delay time as the y-axis. A functional relationship between the instruction data volume and the instruction delay time is constructed through linear fitting.

[0010] in, The dependent variable represents the instruction delay time output; The independent variable represents the amount of instruction data input; This represents the average delay time of the instructions involved in the linear fitting; This represents the average amount of instruction data involved in the linear fitting; , These represent the amount of instruction data and the instruction delay time of the i-th group participating in the linear fitting, respectively. If the actual instruction delay time exceeds the predicted instruction delay time, it is marked as abnormal data transmission; Obtain the total amount of abnormal data transmission within the monitoring period. If the total amount exceeds the threshold for the number of abnormal data transmissions, it is defined as a fault warning caused by an external network attack.

[0011] According to the above technical solution, the setting of the threshold for the number of abnormal transmitted data includes: Obtain the number of data entries t generated within the monitoring period, where one inverter generates one data entry in each sub-interval; Acquire second test data under a simulated environment. The second test data under the simulated environment refers to the number of abnormal transmission data under each number of data transmissions recorded by constructing a secure transmission channel in the laboratory and randomly setting the number of data transmissions in the system. In the second test data under the simulated environment, test data with a data transmission count of t is selected. The number of data groups participating in the test is determined, and the number of abnormal transmission data in each group is marked in sequence. The initial trend is set to 0, and a state vector matrix is ​​defined, which includes the current level and the current trend. The initial noise of the state change and the initial observation noise are set to form the prediction result of the next round of observations. The difference is formed by comparing the number of abnormal transmission data of the prediction result with the actual observation of the next round. The Kalman gain is calculated, and the predicted state is corrected by using the Kalman gain to update the uncertainty. Repeat the prediction for each round until the (t+1)th round is reached. At this point, output the predicted value of the observation and round it up as the threshold for the number of abnormal data transmissions.

[0012] A control system for a photovoltaic frequency modulation inverter, the system comprising: a central control module, a scheduling and processing module, a network security monitoring module, and a fault diagnosis module; The central control module has a built-in central controller, which monitors the grid frequency in real time, calculates the total power that needs to be adjusted for the entire power station, and sends instructions to the scheduling and processing module. The scheduling and processing module has several built-in scheduling centers, each of which is connected to several inverters. After receiving an instruction, the scheduling center issues instructions to its subordinate inverters. The network security monitoring module is connected to the scheduling and processing module. While the scheduling and processing module issues instructions, the network security monitoring module randomly groups all inverters through the security center inside the module to form several security feedback groups. The network security monitoring module also receives the control results of each inverter from the system through the monitoring center inside the module. When a fault warning result is received, the module locates the security feedback group where the corresponding inverter is located, which is recorded as the first security feedback group. The module then sends encryption and decryption commands to the communication links between several inverters in the first security feedback group and the dispatch center. The fault diagnosis module establishes a monitoring period. Within the monitoring period, it acquires the transmission data of the communication link that issues encryption and decryption commands and the transmission data of other communication links in the first security feedback group in real time, and uses a data model to confirm whether the fault warning is caused by an external network attack.

[0013] According to the above technical solution, the central controller collects the three-phase voltage signal of the grid connection point in real time through a high-precision voltage transformer, converts the three-phase voltage signal, and calculates the grid regulation frequency. By calculating the difference between the grid regulation frequency and the rated frequency, a frequency dead zone is set to correct the difference, forming an effective frequency deviation, and the total power that the power station needs to adjust is calculated.

[0014] According to the above technical solution, the method of using a data model to confirm whether the fault warning is caused by an external network attack also includes: Obtain the total amount of abnormal data transmission within the monitoring period. If the total amount exceeds the threshold for the number of abnormal data transmissions, it is defined as a fault warning caused by an external network attack.

[0015] According to the above technical solution, the system also includes an administrator communication module. After the fault diagnosis module confirms the cause of the fault warning, it reports to the administrator communication module. The administrator communication module issues instructions to the administrator through the built-in administrator port to confirm the fault.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: By constructing a centralized and collaborative control architecture at the power station level, the present invention can realize the overall frequency regulation control of the power station based on a central controller, and has a complete control system, enabling all inverters to operate synchronously according to a unified strategy, eliminating the incoordination phenomenon of multi-machine operation. Moreover, when expanding the power station, adding inverters, or replacing equipment of different brands, the central controller can be directly used to provide a standardized dispatch interface, which can be conveniently connected to the power grid dispatch center system and quickly adapt to the differentiated requirements of different power grids for frequency regulation functions. A security feedback group is established during system operation to achieve data linkage and information interaction between inverters. In the face of network security attacks, the inverters no longer operate independently, thereby quickly identifying the attacked link by differentiation, enabling the entire system to quickly identify the attack source and trace the fault. At the same time, the built-in network security monitoring module and fault diagnosis module can distinguish the source of frequency response anomalies or warning signals in real time. When an anomaly occurs, the system can quickly locate the attack source or fault node, solving the problems of "slow investigation, slow diagnosis, and slow response" in existing technologies, greatly improving fault response efficiency, shortening fault diagnosis and recovery time, and significantly improving the operational reliability of the photovoltaic power station frequency regulation system. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the control method of a photovoltaic frequency modulation inverter according to the present invention. Detailed Implementation

[0018] Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] Example: Figure 1 As shown, the present invention provides a method for regulating a photovoltaic frequency modulation inverter, the method comprising the following steps: The central controller monitors the grid frequency in real time, calculates the total power that the entire power station needs to adjust, and issues instructions to several dispatch centers. The central controller monitors the grid frequency in real time and calculates the total power that needs to be adjusted for the entire power station. This includes: the central controller acquiring the three-phase voltage signal at the grid connection point in real time through a high-precision voltage transformer; and calculating the real-time grid frequency based on the acquired voltage signal using a phase-locked loop (PLL) method. The three-phase voltage is converted using the Clarke transform:

[0020]

[0021] in, , These represent the results obtained after Clarke transform. Axial components, Axial components; , , These represent the three-phase voltage signals respectively; Through Park transformation , Convert to rotating coordinate system , ;

[0022]

[0023] in, The active / amplitude component representing the voltage vector; Represents the reactive / phase error component of the voltage vector; Controlled by PI regulator This makes the phase-locked loop output angular frequency Tracking the grid angular frequency:

[0024] in, The feedforward angular frequency is set to [value] in this application. rad / s, corresponding to 50Hz; For PI regulator output; Further calculation of power grid regulation frequency :

[0025] Calculate the difference between the grid regulation frequency and the rated frequency, set a frequency dead zone to correct the difference, forming an effective frequency deviation, and calculate the total active power that the power station needs to adjust:

[0026] in, The total power that the power plant needs to adjust; This refers to the upper limit of the adjustable capacity of the power plant; The adjustment rate; The rated frequency is 50Hz in this application; This represents the effective frequency deviation.

[0027] In actual production, the calculated power adjustment amount needs to be limited due to the actual adjustable capacity of the power plant.

[0028] Each dispatch center connects to several inverters. After receiving instructions, the dispatch center issues instructions to its subordinate inverters. Simultaneously, the security center randomly groups all inverters into several security feedback groups. Inverters within each security feedback group establish data interaction, but there is no data interaction between security feedback groups. Data interaction refers to the existence of data channels between inverters, i.e., data ports are reserved between inverters. In this application, an unprotected data channel is used. When subjected to a network attack, inverters with data channels may be affected by collateral damage. However, due to the grouping setting, the entire group of inverters can be managed and controlled by the dispatch center. The system determines the control result based on the data feedback of each inverter. When a fault warning result is received, the system locates the safety feedback group where the corresponding inverter is located, which is denoted as the first safety feedback group. The system then sends encryption and decryption commands to the communication links between several inverters in the first safety feedback group and the dispatch center. The control results include control success and control failure. Control failure includes control delay and control command mismatch. When the feedback is control failure, a fault warning result is generated. When the system receives a fault warning result from a certain inverter, it locates the safety feedback group to which the corresponding inverter belongs and starts to send encryption and decryption commands to the communication link between several inverters and the dispatch center. The encryption / decryption instruction refers to the dispatch center using a hash algorithm to generate a unique digital digest for the instruction, signing it with a private key, and setting a public key and computing port at the inverter receiver. Upon receiving the instruction, the public key is used to verify the signature and confirm the integrity of the instruction. The computing port confirms the instruction data, which includes the instruction loss percentage and the amount of instruction data. At the same time, a timer is set at the inverter receiver to obtain the instruction delay time.

[0029] In this embodiment, a monitoring period is also established. During the monitoring period, the transmission data of the communication link that issues encryption and decryption commands and the transmission data of other communication links in the first security feedback group are acquired in real time. A data model is used to confirm whether the fault warning is caused by an external network attack. The system establishes a time interval as a monitoring period. Within the monitoring period, it acquires the transmission data of the communication link that issues encryption / decryption commands. The transmission data includes command data and command delay time. Simultaneously, the inverter receivers of other communication links in the first security feedback group set up computing ports to acquire the transmission data of other communication links in the first security feedback group. The monitoring period is generally set in minutes. When a warning is issued, encryption / decryption commands are urgently set. By observing the delay changes, it is determined whether there is a continuous network attack.

[0030] The method of using data models to confirm whether the fault warning is caused by an external network attack includes: The monitoring period is randomly divided into several sub-intervals. For each sub-interval, the percentage of instruction loss in the instruction data is obtained. The system sets an instruction loss percentage threshold. If the instruction loss percentage exceeds the threshold, the current transmission is directly defined as abnormal and marked as abnormal transmission data. For each sub-interval, obtain the instruction delay time and instruction data volume; The first test data under a simulated environment is obtained. This simulated test data is generated by constructing a secure transmission channel in a laboratory, controlling variables to ensure a consistent network environment, and recording the instruction delay time for each transmission of instructions with added encryption / decryption commands. Several scattered points are formed in a coordinate system, with the data corresponding to different instruction data volumes as the x-axis and the data corresponding to the instruction delay time as the y-axis. A functional relationship between the instruction data volume and the instruction delay time is constructed through linear fitting.

[0031] in, The dependent variable represents the instruction delay time output; The independent variable represents the amount of instruction data input; This represents the average delay time of the instructions involved in the linear fitting; This represents the average amount of instruction data involved in the linear fitting; , These represent the amount of instruction data and the instruction delay time of the i-th group participating in the linear fitting, respectively. If the actual instruction delay time exceeds the predicted instruction delay time, it is marked as abnormal data transmission; Obtain the total amount of abnormal data transmission within the monitoring period. If the total amount exceeds the threshold for the number of abnormal data transmissions, it is defined as a fault warning caused by an external network attack.

[0032] The setting of the threshold for the number of abnormal transmitted data includes: Obtain the number of data entries t generated within the monitoring period, where one inverter generates one data entry in each sub-interval; Acquire second test data under a simulated environment. The second test data under the simulated environment refers to the number of abnormal transmission data under each number of data transmissions recorded by constructing a secure transmission channel in the laboratory and randomly setting the number of data transmissions in the system. In the second test data under the simulated environment, test data with a data transmission count of t is selected. If there is not enough test data in the system, an early warning will be issued and the data will be supplemented. Under normal circumstances, the system will store the common data counts when it is set up, determine the number of data groups participating in the test, and mark the number of abnormal transmission data in each group in order. The initial trend is set to 0, and a state vector matrix is ​​defined. The state vector matrix includes the current level and the current trend. The initial noise of the state change and the initial observation noise are set to form the prediction result of the next round of observations. The difference is formed by comparing the number of abnormal transmission data with the actual observations in the next round. The Kalman gain is calculated, and the predicted state is corrected by using the Kalman gain to update the uncertainty. Repeat the prediction for each round until the (t+1)th round is reached. At this point, output the predicted value of the observation and round it up as the threshold for the number of abnormal data transmissions.

[0033] Specifically: Define the state vector matrix as follows:

[0034] Where t represents the order of the markings, Represents the current level; Represents the current trend; The state transition is performed using a state transition equation, and the next round of observation predictions is generated using an observation equation. The state transition equation is as follows:

[0035] Where F is the state transition matrix; Noise representing state changes; Observation equation:

[0036] Where H is the observation matrix; Based on a comparison of the number of anomalous transmission data between the predicted observations and the actual observations in the next round, a difference is calculated. The Kalman gain is then used to correct the predicted state, and the uncertainty equation is updated as follows:

[0037]

[0038] in, This represents the updated covariance after the t-th observation is introduced; Let t be the covariance of the prediction. The process noise covariance matrix; Represents the Kalman gain at the t-th iteration; Represents the identity matrix; When the process reaches round t, since there are no actual observations in round t+1, the predicted value is used as the final output.

[0039] In this embodiment, a control system for a photovoltaic frequency modulation inverter is also included, the system comprising: a central control module, a scheduling and processing module, a network security monitoring module, and a fault diagnosis module; The central control module has a built-in central controller, which monitors the grid frequency in real time, calculates the total power that needs to be adjusted for the entire power station, and sends instructions to the scheduling and processing module. The scheduling and processing module has several built-in scheduling centers, each of which is connected to several inverters. After receiving an instruction, the scheduling center issues instructions to its subordinate inverters. The network security monitoring module is connected to the scheduling and processing module. While the scheduling and processing module issues instructions, the network security monitoring module randomly groups all inverters through the security center inside the module to form several security feedback groups. The network security monitoring module also receives the control results of each inverter from the system through the monitoring center inside the module. When a fault warning result is received, the module locates the security feedback group where the corresponding inverter is located, which is recorded as the first security feedback group. The module then sends encryption and decryption commands to the communication links between several inverters in the first security feedback group and the dispatch center. The fault diagnosis module establishes a monitoring period. Within the monitoring period, it acquires the transmission data of the communication link that issues encryption and decryption commands and the transmission data of other communication links in the first security feedback group in real time, and uses a data model to confirm whether the fault warning is caused by an external network attack.

[0040] The central controller collects the three-phase voltage signal at the grid connection point in real time through a high-precision voltage transformer. After converting the three-phase voltage signal, it calculates the grid regulation frequency. By calculating the difference between the grid regulation frequency and the rated frequency, a frequency dead zone is set to correct the difference, forming an effective frequency deviation, and the total power that the power station needs to adjust is calculated.

[0041] The method of using data models to confirm whether a fault warning is caused by an external network attack also includes: Obtain the total amount of abnormal data transmission within the monitoring period. If the total amount exceeds the threshold for the number of abnormal data transmissions, it is defined as a fault warning caused by an external network attack.

[0042] The system also includes an administrator communication module. After the fault diagnosis module confirms the cause of the fault warning, it reports to the administrator communication module. The administrator communication module issues instructions to the administrator through the built-in administrator port to confirm the fault.

[0043] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method of regulating a photovoltaic frequency-regulated inverter, the method comprising: The method includes the following steps: ​ S1. The central controller monitors the grid frequency in real time, calculates the total power that needs to be adjusted for the entire power station, and sends instructions to several dispatch centers. S2. Each dispatch center connects to several inverters. After receiving the instruction, the dispatch center issues the instruction to its subordinate inverters. At the same time, the safety center randomly groups all inverters into several safety feedback groups. Data interaction is established between inverters in each safety feedback group, but no data interaction is established between safety feedback groups. S3. The system determines the control result based on the data feedback of each inverter. When a fault warning result is received, the system locates the safety feedback group where the corresponding inverter is located, which is called the first safety feedback group. The system sends encryption and decryption commands to the communication link between several inverters in the first safety feedback group and the dispatch center. S4. Establish a monitoring period. During the monitoring period, acquire the transmission data of the communication link that issues encryption and decryption commands and the transmission data of other communication links in the first security feedback group in real time, and use a data model to confirm whether the fault warning is caused by an external network attack. S5. After confirming the cause of the fault warning, report it to the administrator port for fault confirmation.

2. The method of claim 1, wherein the method further comprises: The central controller monitors the grid frequency in real time and calculates the total power that needs to be adjusted for the entire power station, including: The central controller acquires the three-phase voltage signal at the grid connection point in real time through a high-precision voltage transformer. The three-phase voltage signal is converted by Clarke transformation to obtain Clarke-transformed axis components, axis components; the Clarke-transformed axis components, axis components are converted into active / magnitude component of voltage vector and reactive / phase error component of voltage vector in a rotating coordinate system by Park transformation. By controlling the reactive / phase error components of the voltage vector to 0 using a PI regulator, the phase-locked loop output angular frequency tracks the grid angular frequency, and the grid regulation frequency is calculated. By calculating the difference between the grid regulation frequency and the rated frequency, a frequency dead zone is set to correct the difference, forming an effective frequency deviation, and the total power that the power station needs to adjust is calculated: wherein, is the total power needed for the power plant to adjust; is the upper limit of the adjustable capacity of the power plant; is the regulation difference rate; is the rated frequency; is the effective frequency deviation.

3. The method of claim 1, wherein the method further comprises: The system determines the control result based on the data feedback from each inverter. Upon receiving a fault warning result, it locates the safety feedback group to which the corresponding inverter belongs, denoted as the first safety feedback group, and issues encryption / decryption commands to the communication links between several inverters in the first safety feedback group and the dispatch center. The control results include control success and control failure. Control failure includes control delay and control command mismatch. When the feedback is control failure, a fault warning result is generated. When the system receives a fault warning result from a certain inverter, it locates the safety feedback group to which the corresponding inverter belongs and starts to send encryption and decryption commands to the communication link between several inverters and the dispatch center. The encryption / decryption instruction refers to the dispatch center using a hash algorithm to generate a unique digital digest for the instruction, signing it with a private key, and setting a public key and computing port at the inverter receiver. Upon receiving the instruction, the public key is used to verify the signature and confirm the integrity of the instruction. The computing port confirms the instruction data, which includes the instruction loss percentage and the amount of instruction data. At the same time, a timer is set at the inverter receiver to obtain the instruction delay time.

4. The method of claim 3, wherein the method further comprises: The establishment of a monitoring period, during which the transmission data of the communication link issuing encryption / decryption commands and the transmission data of other communication links in the first security feedback group are acquired in real time, includes: The system establishes a time interval as a monitoring period. Within the monitoring period, it acquires the transmission data of the communication link that issues encryption and decryption commands. The transmission data includes command data and command delay time. At the same time, the inverter receiver of other communication links in the first security feedback group sets up a computing port to acquire the transmission data of other communication links in the first security feedback group.

5. The method of claim 4, wherein the method further comprises: The method of using data models to confirm whether the fault warning is caused by an external network attack includes: The monitoring period is randomly divided into several sub-intervals. For each sub-interval, the percentage of instruction loss in the instruction data is obtained. The system sets an instruction loss percentage threshold. If the instruction loss percentage exceeds the threshold, the current transmission is directly defined as abnormal and marked as abnormal transmission data. For each sub-interval, obtain the instruction delay time and instruction data volume; The first test data under a simulated environment is obtained. This simulated test data is generated by constructing a secure transmission channel in a laboratory, controlling variables to ensure a consistent network environment, and recording the instruction delay time for each transmission of instructions with added encryption / decryption commands. Several scattered points are formed in a coordinate system, with the data corresponding to different instruction data volumes as the x-axis and the data corresponding to the instruction delay time as the y-axis. A functional relationship between the instruction data volume and the instruction delay time is constructed through linear fitting. wherein, is the dependent variable, representing instruction latency time output; is the independent variable, representing instruction data size input; represents the average of instruction latency time participating in linear fitting; represents the average of instruction data size participating in linear fitting; , respectively represent the i-th group of instruction data size, instruction latency time participating in linear fitting; If the actual instruction delay time exceeds the predicted instruction delay time, it is marked as abnormal data transmission; Obtain the total amount of abnormal data transmission within the monitoring period. If the total amount exceeds the threshold for the number of abnormal data transmissions, it is defined as a fault warning caused by an external network attack.

6. The method of claim 5, wherein the method further comprises: The setting of the threshold for the number of abnormal transmitted data includes: Obtain the number of data entries t generated within the monitoring period, where one inverter generates one data entry in each sub-interval; Acquire second test data under a simulated environment. The second test data under the simulated environment refers to the number of abnormal transmission data under each number of data transmissions recorded by constructing a secure transmission channel in the laboratory and randomly setting the number of data transmissions in the system. In the second test data under the simulated environment, test data with a data transmission count of t is selected. The number of data groups participating in the test is determined, and the number of abnormal transmission data in each group is marked in sequence. The initial trend is set to 0, and a state vector matrix is ​​defined, which includes the current level and the current trend. The initial noise of the state change and the initial observation noise are set to form the prediction result of the next round of observations. The difference is formed by comparing the number of abnormal transmission data of the prediction result with the actual observation of the next round. The Kalman gain is calculated, and the predicted state is corrected by using the Kalman gain to update the uncertainty. Repeat the prediction for each round until the (t+1)th round is reached. At this point, output the predicted value of the observation and round it up as the threshold for the number of abnormal data transmissions.

7. A control system of a photovoltaic frequency-regulated inverter, for implementing a control method of a photovoltaic frequency-regulated inverter according to claim 1, characterized in that: The system includes: a central control module, a scheduling and processing module, a network security monitoring module, and a fault diagnosis module; The central control module has a built-in central controller, which monitors the grid frequency in real time, calculates the total power that needs to be adjusted for the entire power station, and sends instructions to the scheduling and processing module. The scheduling and processing module has several built-in scheduling centers, each of which is connected to several inverters. After receiving an instruction, the scheduling center issues instructions to its subordinate inverters. The network security monitoring module is connected to the scheduling and processing module. While the scheduling and processing module issues instructions, the network security monitoring module randomly groups all inverters through the security center inside the module to form several security feedback groups. The network security monitoring module also receives the control results of each inverter from the system through the monitoring center inside the module. When a fault warning result is received, the module locates the security feedback group where the corresponding inverter is located, which is recorded as the first security feedback group. The module then sends encryption and decryption commands to the communication links between several inverters in the first security feedback group and the dispatch center. The fault diagnosis module establishes a monitoring period. Within the monitoring period, it acquires the transmission data of the communication link that issues encryption and decryption commands and the transmission data of other communication links in the first security feedback group in real time, and uses a data model to confirm whether the fault warning is caused by an external network attack.

8. The system for regulating a photovoltaic frequency inverter according to claim 7, wherein: The central controller collects the three-phase voltage signal at the grid connection point in real time through a high-precision voltage transformer. After converting the three-phase voltage signal, it calculates the grid regulation frequency. By calculating the difference between the grid regulation frequency and the rated frequency, a frequency dead zone is set to correct the difference, forming an effective frequency deviation. The total power that the power station needs to adjust is then calculated.

9. The regulating system of photovoltaic frequency-regulated inverter according to claim 7, characterized in that: The method of using data models to confirm whether the fault warning is caused by an external network attack also includes: Obtain the total amount of abnormal data transmission within the monitoring period. If the total amount exceeds the threshold for the number of abnormal data transmissions, it is defined as a fault warning caused by an external network attack.

10. The system of claim 7, wherein: The system also includes an administrator communication module. After the fault diagnosis module confirms the cause of the fault warning, it reports to the administrator communication module. The administrator communication module issues instructions to the administrator through the built-in administrator port to confirm the fault.

Images