Intelligent identification method and system for high-voltage ride-through parameters of grid-following type energy storage inverter
By extending the state-space model and using Kalman filtering and Bayesian estimation methods, inverter data is collected in real time to identify high-voltage ride-through parameters, solving the problems of response delay and large error in existing technologies, and realizing stable operation and high-precision parameter estimation of energy storage inverters under abnormal grid conditions.
Patent Information
- Application Number
- CN202511139413.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-14
- Publication Date
- 2025-12-09
AI Technical Summary
Existing high-voltage ride-through parameter identification methods rely on offline experiments and manual calculations, which have drawbacks such as response delay, cumbersome calculations, and large errors, affecting the stable operation of energy storage inverters under abnormal grid voltage conditions.
An extended state-space model is adopted, combined with Kalman filtering and Bayesian estimation methods. Inverter data is collected in real time, an extended state-space model is constructed and parameters are estimated. The accuracy of parameter estimation is optimized by Kalman gain update and Bayesian inference.
It enables real-time and accurate identification of high-voltage ride-through parameters, improves the dynamic response capability of the system, reduces computational complexity and errors, and supports the reliability of grid stable operation and simulation modeling.
Smart Images
Figure CN121097635A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of power electronics and intelligent control technology, specifically to a method and system for intelligent identification of high voltage ride-through parameters of grid-connected energy storage inverters. Background Technology
[0002] During the grid connection of new energy power generation and energy storage systems, energy storage inverters need to possess high-voltage ride-through capability to ensure stable operation under abnormal grid voltage conditions. Existing high-voltage ride-through parameter (HVRT) identification methods mainly rely on offline experiments and manual calculations, which suffer from drawbacks such as response delays, cumbersome calculations, and large errors, affecting the accurate reproduction of the system's dynamic response by electromechanical transient simulation models. Therefore, there is an urgent need for an intelligent, real-time parameter identification method to achieve accurate estimation of high-voltage ride-through parameters. Summary of the Invention
[0003] The technical problem to be solved by this invention is to provide a method and system for intelligent identification of high voltage ride-through parameters of grid-connected energy storage inverters, thereby reducing computational complexity and improving the estimation accuracy of high voltage ride-through parameters, in response to the aforementioned problems in the prior art.
[0004] To solve the above-mentioned technical problems, the technical solution adopted by the present invention is as follows: A method for intelligent identification of high-voltage ride-through parameters of grid-connected energy storage inverters includes: Real-time acquisition of operating data during inverter operation, followed by preprocessing to obtain corresponding digital signals; An extended state-space model is constructed by combining the dynamic state of the inverter with the high voltage ride-through parameters to be estimated in the digital signal. The dynamic state refers to the behavior and response characteristics of the inverter during the transition from one steady state to another when external conditions change. The extended state-space model is updated, and the updated high-voltage ride-through parameters in the extended state-space model are used as preliminary estimation results; The preliminary estimation results are then re-estimated to obtain the optimal high-voltage ride-through parameters.
[0005] Furthermore, the expression for the extended state-space model is:
[0006] In the above formula, Let k be the dynamic state of the inverter at time k. For the high-voltage ride-through control parameters to be estimated, Let be the extended state vector at time k; The extended state-space model satisfies:
[0007]
[0008] In the above formula, Let be the extended state vector at time k+1. This is the state transition function. For the input vector, For process noise, To measure the output vector, For measurement function, For measuring noise.
[0009] Furthermore, the expression for the high-voltage ride-through control parameters to be estimated is:
[0010] In the above formula, This represents the primary coefficient of the active current response during high voltage periods. These are the quadratic coefficients of the active current response during high voltage periods. This represents the active current bias value under high voltage conditions. This represents the primary coefficient of the reactive current response during high voltage periods. These are the quadratic coefficients of the reactive current response during high voltage periods. This represents the reactive current bias value under high voltage conditions.
[0011] Furthermore, updating the extended state-space model specifically includes: Based on the current extended state estimation vector and covariance matrix, generate several Sigma points whose number is positively correlated with the dimension of the extended state vector; Each Sigma point is predicted using a state transition function to obtain the corresponding predicted Sigma point, and the predicted mean and predicted covariance matrix are calculated. Based on the measurement function, each predicted Sigma point is mapped to the observation space to obtain the mapped value, and the predicted measurement mean, measurement covariance matrix, and state-measurement cross-covariance matrix are calculated. The Kalman gain is calculated based on the state-measurement cross-covariance matrix, and the extended state estimation vector and covariance matrix at the next time step are updated. The high-voltage ride-through parameter in the updated extended state estimation vector at the next time step is used as the preliminary estimation result.
[0012] Furthermore, the expression for generating several Sigma points based on the current extended state estimation vector and covariance matrix, the number of which is positively correlated with the dimension of the extended state vector, is as follows:
[0013] In the above formula, Let be the extended state estimation vector at time k. Let Sigma be the center point at time k. Let be the i-th positive Sigma point at time k. Let L be the i-th negative Sigma point at time k, and L be the dimension of the extended state vector. For scaling parameters, Let be the covariance matrix at time k; The expressions for calculating the predicted mean and predicted covariance matrix are as follows:
[0014]
[0015] In the above formula, Let $k$ be the predicted mean of the extended state estimation vector at time $k+1$, which is predicted at time $k$. Let m be the mean weight of the i-th Sigma point, where m is the mean. Let the i-th Sigma point at time k be propagated to the predicted Sigma point at time k+1 through the state transition function. Let be the prediction covariance matrix at time k+1 predicted at time k. Let c be the mean weight of the i-th Sigma point, and c be the covariance. The process noise covariance matrix; The expressions for calculating the predicted measurement mean, measurement covariance matrix, and state-measurement cross-covariance matrix are as follows:
[0016]
[0017]
[0018] In the above formula, Let be the mean of the predicted measurements at time k+1. To map the predicted Sigma point at time k+1 to the observation space, To measure the covariance matrix, To observe the noise covariance matrix, The state-measurement cross-covariance matrix; The expressions for calculating the Kalman gain and updating the extended state estimate vector and covariance matrix at the next time step are as follows:
[0019]
[0020]
[0021] In the above formula, Let Kalman gain be the value at time k+1. This is the final estimate of the extended state estimation vector at time k+1. This is the actual measurement value at time k+1. This is the covariance matrix updated at time k+1. To predict the observation vector The covariance matrix.
[0022] Furthermore, the preliminary estimation result is re-estimated, specifically including: To establish a priori distribution for high voltage ride-through parameters; Construct a likelihood function to evaluate the degree of matching of observed data under different parameter values; Update the posterior distribution according to Bayes' theorem; The optimal high-voltage ride-through parameters are obtained by solving the maximum a posteriori estimation.
[0023] Furthermore, the expression for setting the prior distribution of the high-voltage ride-through parameters is as follows:
[0024] In the above formula, For high voltage ride-through parameters, Let be the prior mean vector. The prior covariance matrix; The expression for the likelihood function is:
[0025] In the above formula, To observe the noise covariance matrix, This is the actual measurement value at time k+1. This is the dynamic state estimation of the inverter at time k+1. For measurement functions; The expression for updating the posterior distribution according to Bayes' theorem is:
[0026] In the above formula, This represents all actual measurements from time 1 to time k+1. These are all actual measured values from time 1 to time k; The expression for the optimal high-voltage ride-through parameters obtained by solving the maximum a posteriori estimation is as follows: .
[0027] Furthermore, the intelligent identification method for high-voltage ride-through parameters of the grid-connected energy storage inverter also includes: The active current command and reactive current command are calculated based on the optimal high-voltage ride-through parameters, and the calculation expressions are as follows:
[0028]
[0029] In the above formula, This is an active current command. This represents the primary coefficient of the active current response during high voltage periods. These are the quadratic coefficients of the active current response during high voltage periods. This represents the active current bias value under high voltage conditions. The terminal voltage amplitude, This is the initial active current. This is a reactive current command. This represents the primary coefficient of the reactive current response during high voltage periods. These are the quadratic coefficients of the reactive current response during high voltage periods. This represents the reactive current bias value under high voltage conditions. To reach the high voltage ride-through threshold, This is the initial reactive current.
[0030] A smart identification system for high-voltage ride-through parameters of a grid-connected energy storage inverter includes: The data acquisition and preprocessing module is used to acquire the operating data of the inverter in real time and obtain the corresponding digital signal after preprocessing. The model building module is used to jointly construct an extended state-space model by combining the dynamic state of the inverter with the high voltage ride-through parameters to be estimated in the digital signal. The dynamic state refers to the behavior and response characteristics of the inverter during the transition from one steady state to another when external conditions change. The first parameter estimation module is used to update the extended state space model and use the updated high voltage ride-through parameters in the extended state space model as the preliminary estimation result. The second parameter estimation module is used to re-estimate the preliminary estimation results to obtain the optimal high-voltage ride-through parameters.
[0031] A computer-readable storage medium storing a computer program / instructions programmed or configured to execute, via a processor, a method for intelligent identification of high-voltage ride-through parameters of a grid-connected energy storage inverter.
[0032] Compared with the prior art, the advantages of the present invention are as follows: This invention improves the accuracy of joint estimation of nonlinear systems by constructing an extended state-space model, avoiding the accumulation of errors in parameter and state-separate estimations. It enhances identification accuracy, overcoming the problems of cumbersome calculations, large errors, and response lags inherent in traditional manual estimation methods, thereby achieving real-time and accurate identification of high-voltage ride-through parameters. The system can rapidly acquire the actual response behavior of the inverter after grid disturbances occur, completing dynamic estimation of high-voltage ride-through parameters, facilitating subsequent simulation modeling and strategy optimization to improve simulation efficiency and modeling reliability. Applicable to various operating conditions and control strategies, it possesses advantages such as clear structure, strong adaptability, and flexible deployment, demonstrating significant engineering application value in improving parameter modeling efficiency, reducing labor costs, and supporting stable grid operation. Attached Figure Description
[0033] Figure 1 This is a flowchart of the intelligent identification method for high voltage ride-through parameters of a grid-connected energy storage inverter according to an embodiment of the present invention.
[0034] Figure 2 This is a comparison chart of electromechanical transient simulation results and measured data in a specific application embodiment.
[0035] Figure 3 This is a schematic diagram of the framework of the intelligent identification system for high voltage ride-through parameters of a grid-connected energy storage inverter according to an embodiment of the present invention. Detailed Implementation
[0036] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.
[0037] Example 1 like Figure 1 As shown in this embodiment, the intelligent identification method for high voltage ride-through parameters of a grid-connected energy storage inverter includes: Step S1: Real-time acquisition of operating data during the operation of the inverter, followed by preprocessing to obtain the corresponding digital signal; Step S2: Combine the dynamic state of the inverter with the high voltage ride-through parameters to be estimated in the digital signal to construct an extended state-space model. The dynamic state is the behavior and response characteristics of the inverter during the transition from one steady state to another when external conditions change (specifically involving the instantaneous changes of parameters such as voltage, current, and frequency). Step S3: Update the extended state-space model (specifically, the unscented Kalman filter algorithm can be used), and use the updated high-voltage ride-through parameters in the extended state-space model as the preliminary estimation result; Step S4: Re-estimate the preliminary estimation results (specifically, a Bayesian inference algorithm can be used) to obtain the optimal high-voltage ride-through parameters.
[0038] It is understood that this embodiment, by constructing an extended state-space model, can improve the accuracy of joint estimation of nonlinear systems and avoid the accumulation of errors in parameter and state-separate estimation; it can improve identification accuracy and overcome the problems of cumbersome calculation, large error, and response lag in traditional manual estimation methods, thereby achieving real-time and accurate identification of high-voltage ride-through parameters; the system can quickly acquire the actual response behavior of the inverter after a grid disturbance occurs, and complete the dynamic estimation of high-voltage ride-through parameters, which is convenient for subsequent simulation modeling and strategy optimization to improve simulation efficiency and modeling reliability; it is applicable to various operating conditions and control strategies, and has the advantages of clear structure, strong adaptability, and flexible deployment, and has significant engineering application value in improving parameter modeling efficiency, reducing labor costs, and supporting the stable operation of the power grid.
[0039] In a specific application embodiment, step S1 specifically includes: Step S1.1, Data Acquisition: Real-time acquisition of operating data of the energy storage inverter under high voltage disturbance conditions, including the voltage of each phase. U a , U b , U c Current in each phase I a , I b , I c Active power P reactive power Q Other relevant control signals, forming the original data sequence { y k}; Step S1.2, Data Preprocessing: The original data sequence is filtered, denoised, normalized, and format-converted to obtain the preprocessed digital signal. To meet the input requirements of online identification algorithms.
[0040] In this embodiment, the expression for the extended state-space model is: (1) In the above formula, Let k be the dynamic state of the inverter at time k. The high-voltage ride-through parameters to be estimated (used to accurately reflect the dynamic characteristics of the inverter under high-voltage fault conditions). Let be the extended state vector at time k; The extended state-space model satisfies the state transition equation shown in equation (2) and the measurement equation shown in equation (3): (2) (3) In equation (2), Let be the extended state vector at time k+1. This is the state transition function. For the input vector, For process noise; in equation (3), To measure the output vector, For measurement function, For measuring noise; where, , , The process noise covariance matrix is... To observe the noise covariance matrix.
[0041] It is understandable that the dynamic state of the inverter system and the HVRT parameters to be estimated are used to construct an extended state-space model. The model introduces structured state variables of the voltage disturbance evolution process to customize the nonlinear transient characteristics during high voltage ride-through, thereby achieving a more accurate description of abnormal operating conditions and improving robustness.
[0042] In this embodiment, the expression for the high-voltage ride-through control parameter to be estimated is: (4) In the above formula, This represents the primary coefficient of the active current response during high voltage periods. These are the quadratic coefficients of the active current response during high voltage periods. This represents the active current bias value under high voltage conditions. This represents the primary coefficient of the reactive current response during high voltage periods. These are the quadratic coefficients of the reactive current response during high voltage periods. This represents the reactive current bias value under high voltage conditions.
[0043] In this embodiment, updating the extended state-space model specifically includes: Based on the current extended state estimation vector and covariance matrix, generate several Sigma points whose number is positively correlated with the dimension of the extended state vector; Each Sigma point is predicted using a state transition function to obtain the corresponding predicted Sigma point, and the predicted mean and predicted covariance matrix are calculated. Based on the measurement function, each predicted Sigma point is mapped to the observation space to obtain the mapped value, and the predicted measurement mean, measurement covariance matrix, and state-measurement cross-covariance matrix are calculated. The Kalman gain is calculated based on the state-measurement cross-covariance matrix, and the extended state estimation vector and covariance matrix at the next time step are updated. The high-voltage ride-through parameter in the updated extended state estimation vector at the next time step is used as the preliminary estimation result.
[0044] In this embodiment, the expression for generating several Sigma points whose number is positively correlated with the dimension of the extended state vector based on the current extended state estimation vector and covariance matrix is as follows: (5) In the above formula, Let be the extended state estimation vector at time k. Let Sigma be the center point at time k. Let be the i-th positive Sigma point at time k. Let L be the i-th negative Sigma point at time k, and L be the dimension of the extended state vector (preferably, 2L+1 Sigma points are generated). For scaling parameters, Let be the covariance matrix at time k; The predicted mean is calculated using equation (6), and the predicted covariance matrix is calculated using equation (7). The expressions are as follows: (6) (7) In equation (6), Let $k$ be the predicted mean of the extended state estimation vector at time $k+1$, which is predicted at time $k$. Let m be the mean weight of the i-th Sigma point, where m is the abbreviation for mean. Let the i-th Sigma point at time k be propagated to the predicted Sigma point at time k+1 through the state transition function; in equation (7), Let be the prediction covariance matrix at time k+1 predicted at time k. Let be the mean weight of the i-th Sigma point, and c be the abbreviation for covariance. The process noise covariance matrix; The predicted measurement mean is calculated using equation (8), the measurement covariance matrix is calculated using equation (9), and the state-measurement cross-covariance matrix is calculated using equation (10). The expression is as follows: (8) (9) (10) In equation (8), Let be the mean of the predicted measurements at time k+1. To utilize measurement functions The mapping value obtained by mapping the predicted Sigma point at time k+1 to the observation space; in equation (9), To measure the covariance matrix, For the observation noise covariance matrix; in equation (10), The state-measurement cross-covariance matrix; The Kalman gain is calculated using equation (11), the extended state estimation vector for the next time step is updated using equation (12), and the covariance matrix is updated using equation (13). The expression is as follows: (11) (12) (13) In equation (11), Let K be the Kalman gain matrix at time k+1; in equation (12), This is the final estimate of the extended state estimation vector at time k+1. The actual measured value at time k+1; in equation (13), This is the covariance matrix updated at time k+1. To predict the observation vector The covariance matrix.
[0045] It is understandable that, based on the constructed extended state-space model, parameter updates are performed. In this process, an adaptive noise covariance estimation mechanism is introduced, and Sigma point weighting or anomaly suppression factors are used to address observation anomalies caused by voltage surges. This enables the filter to track dynamic noise characteristics online and suppress abnormal observations, thereby achieving reliable estimation of the inverter's high-voltage ride-through parameters.
[0046] In this embodiment, the preliminary estimation result is re-estimated, specifically including: Set a priori distribution for high-voltage ride-through parameters; specifically, a parameter prior library can be constructed based on historical high-voltage disturbance conditions and then clustered and combined to initialize the priori distribution. Construct a likelihood function to evaluate the degree of matching of observed data under different parameter values; The posterior distribution is updated according to Bayes' theorem; the posterior update can be specifically achieved using variational methods to improve the estimation ability of non-Gaussian posterior distributions. The optimal high-voltage ride-through parameters are obtained by solving the maximum a posteriori estimation. The operating condition sensitivity factor can be introduced to weight the a posteriori results to enhance the stability of parameter estimation under different disturbance levels.
[0047] In this embodiment, the expression for setting the prior distribution of the high-voltage ride-through parameters is as follows: (14) In the above formula, For high voltage ride-through parameters, Let be the prior mean vector. The prior covariance matrix; The expression for the likelihood function is: (15) In the above formula, To observe the noise covariance matrix, This is the actual measurement value at time k+1. This is the dynamic state estimation of the inverter at time k+1. For measurement functions; The expression for updating the posterior distribution according to Bayes' theorem is: (16) In the above formula, This represents all actual measurements from time 1 to time k+1. These are all actual measured values from time 1 to time k; The expression for the optimal high-voltage ride-through parameters obtained by solving the maximum a posteriori estimation is as follows: (17) Specifically, the maximum a posteriori estimate of equation (17) can be solved using gradient descent or the quasi-Newton method.
[0048] In this embodiment, the intelligent identification method for high voltage ride-through parameters of grid-connected energy storage inverters further includes: The active current command is calculated using equation (18) based on the optimal high-voltage ride-through parameters, and the reactive current command is calculated using equation (19). The calculation expressions are as follows: (18) (19) In equation (18), This is an active current command. This represents the primary coefficient of the active current response during high voltage periods. These are the quadratic coefficients of the active current response during high voltage periods. This represents the active current bias value under high voltage conditions. The terminal voltage amplitude, In the initial active current equation (19), This is a reactive current command. This represents the primary coefficient of the reactive current response during high voltage periods. These are the quadratic coefficients of the reactive current response during high voltage periods. This represents the reactive current bias value under high voltage conditions. To reach the high voltage ride-through threshold, This is the initial reactive current.
[0049] It is understood that the active current command and reactive current command of the inverter during high voltage ride-through are calculated based on the optimal high voltage ride-through parameters. The command calculation expression is given according to the voltage amplitude during high voltage and the inverter model form, and is used for closed-loop control.
[0050] In an exemplary application embodiment, the high-voltage ride-through parameter identification results obtained by the intelligent identification method for high-voltage ride-through parameters of the grid-connected energy storage inverter of the present invention are as follows: (20) The parameters obtained from equation (20) are used in electromechanical transient PSASP simulation. Under rated power charging conditions, the three-phase voltage on the AC side of the energy storage converter rises to 120%Un. The simulation and test data are compared as follows: Figure 2 As shown, the data includes voltage (HIL_U, PSASP_U), active power (HIL_P, PSASP_P), reactive power (HIL_Q, PSASP_Q), active current (HIL_Id, PSASP_Id), and reactive current (HIL_Iq, PSASP_Iq). The "Hil" prefix indicates that the data is test data, and the "PSASP" prefix indicates that the data is simulation data. It can be seen that the simulation results are in good agreement with the actual measurement data, further demonstrating the high accuracy of the high-voltage ride-through parameter identification results obtained by the intelligent identification method for grid-connected energy storage inverters of this invention.
[0051] The standard requires that the simulation calculation deviations of voltage, current, reactive current, active power, and reactive power at the grid connection point of electrochemical energy storage power stations should meet the maximum allowable deviation values specified in Table 1. After simulation, the calculation results of the electromechanical transient simulation deviations are shown in Table 2. It can be seen that the results of each simulation data item are within the maximum deviation range specified in Table 1.
[0052] Table 1. Maximum allowable deviation values
[0053] Table 2 Deviation values of electromechanical transient simulation results
[0054] Example 2 like Figure 3 As shown, the intelligent identification system for high voltage ride-through parameters of the grid-connected energy storage inverter in this embodiment includes: The data acquisition and preprocessing module is used to acquire the operating data of the inverter in real time and obtain the corresponding digital signal after preprocessing. The model building module is used to jointly construct an extended state-space model by combining the dynamic state of the inverter with the high voltage ride-through parameters to be estimated in the digital signal. The dynamic state is the behavior and response characteristics of the inverter during the transition from one steady state to another when external conditions change. The first parameter estimation module is used to update the extended state space model and use the updated high voltage ride-through parameters in the extended state space model as the preliminary estimation results. The second parameter estimation module is used to re-estimate the preliminary estimation results to obtain the optimal high-voltage ride-through parameters.
[0055] In specific application embodiments, the intelligent identification system for high voltage ride-through parameters of grid-connected energy storage inverters may further include: The control feedback module feeds back the identified parameters to the electromechanical transient simulation analysis system via a communication interface for engineers to refer to. The communication and human-computer interaction module enables high-speed data transmission and real-time monitoring and display between various modules.
[0056] In addition, the present invention provides a computer-readable storage medium storing a computer program / instruction, which is programmed or configured to execute, via a processor, a method for intelligent identification of high voltage ride-through parameters of a grid-connected energy storage inverter.
[0057] The system and medium of the present invention, corresponding to the methods described above, also have the advantages described above.
[0058] The present invention can implement all or part of the processes in the methods of the above embodiments, or it can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of the above method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. Computer-readable media include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. The memory is used to store computer programs and / or modules. The processor implements various functions by running or executing the computer programs and / or modules stored in the memory, and by calling data stored in the memory. The memory may include high-speed random access memory, as well as non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.
[0059] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should also be considered within the scope of protection of the present invention.
Claims
1. A method for intelligent identification of high-voltage ride-through parameters in a grid-connected energy storage inverter, characterized in that, include: Real-time acquisition of operating data during inverter operation, followed by preprocessing to obtain corresponding digital signals; An extended state-space model is constructed by combining the dynamic state of the inverter with the high voltage ride-through parameters to be estimated in the digital signal. The dynamic state refers to the behavior and response characteristics of the inverter during the transition from one steady state to another when external conditions change. The extended state-space model is updated, and the updated high-voltage ride-through parameters in the extended state-space model are used as preliminary estimation results; The preliminary estimation results are then re-estimated to obtain the optimal high-voltage ride-through parameters.
2. The intelligent identification method for high voltage ride-through parameters of grid-connected energy storage inverters according to claim 1, characterized in that, The expression for the extended state-space model is: In the above formula, Let k be the dynamic state of the inverter at time k. For the high-voltage ride-through control parameters to be estimated, Let be the extended state vector at time k; The extended state-space model satisfies: In the above formula, Let be the extended state vector at time k+1. This is the state transition function. For the input vector, For process noise, To measure the output vector, For measurement function, For measuring noise.
3. The intelligent identification method for high voltage ride-through parameters of grid-connected energy storage inverters according to claim 2, characterized in that, The expression for the high-voltage ride-through control parameters to be estimated is: In the above formula, This represents the primary coefficient of the active current response during high voltage periods. These are the quadratic coefficients of the active current response during high voltage periods. This represents the active current bias value under high voltage conditions. This represents the primary coefficient of the reactive current response during high voltage periods. These are the quadratic coefficients of the reactive current response during high voltage periods. This represents the reactive current bias value under high voltage conditions.
4. The intelligent identification method for high voltage ride-through parameters of grid-connected energy storage inverters according to claim 2, characterized in that, Updating the extended state-space model specifically includes: Based on the current extended state estimation vector and covariance matrix, generate several Sigma points whose number is positively correlated with the dimension of the extended state vector; Each Sigma point is predicted using a state transition function to obtain the corresponding predicted Sigma point, and the predicted mean and predicted covariance matrix are calculated. Based on the measurement function, each predicted Sigma point is mapped to the observation space to obtain the mapped value, and the predicted measurement mean, measurement covariance matrix, and state-measurement cross-covariance matrix are calculated. The Kalman gain is calculated based on the state-measurement cross-covariance matrix, and the extended state estimation vector and covariance matrix at the next time step are updated. The high-voltage ride-through parameter in the updated extended state estimation vector at the next time step is used as the preliminary estimation result.
5. The intelligent identification method for high voltage ride-through parameters of grid-connected energy storage inverters according to claim 4, characterized in that, The expression for generating several Sigma points that are positively correlated with the dimension of the extended state vector based on the current extended state estimation vector and covariance matrix is as follows: In the above formula, Let be the extended state estimation vector at time k. Let Sigma be the center point at time k. Let be the i-th positive Sigma point at time k. Let L be the i-th negative Sigma point at time k, and L be the dimension of the extended state vector. For scaling parameters, Let be the covariance matrix at time k; The expressions for calculating the predicted mean and predicted covariance matrix are as follows: In the above formula, Let $k$ be the predicted mean of the extended state estimation vector at time $k+1$, which is predicted at time $k$. Let m be the mean weight of the i-th Sigma point, where m is the mean. Let the i-th Sigma point at time k be propagated to the predicted Sigma point at time k+1 through the state transition function. Let be the prediction covariance matrix at time k+1 predicted at time k. Let c be the mean weight of the i-th Sigma point, and c be the covariance. The process noise covariance matrix; The expressions for calculating the predicted measurement mean, measurement covariance matrix, and state-measurement cross-covariance matrix are as follows: In the above formula, Let be the mean of the predicted measurements at time k+1. To map the predicted Sigma point at time k+1 to the observation space, To measure the covariance matrix, To observe the noise covariance matrix, The state-measurement cross-covariance matrix; The expressions for calculating the Kalman gain and updating the extended state estimate vector and covariance matrix at the next time step are as follows: In the above formula, Let Kalman gain be the value at time k+1. This is the final estimate of the extended state estimation vector at time k+1. This is the actual measurement value at time k+1. This is the covariance matrix updated at time k+1. To predict the observation vector The covariance matrix.
6. The intelligent identification method for high voltage ride-through parameters of grid-connected energy storage inverters according to claim 1, characterized in that, The preliminary estimation results are then re-estimated, specifically including: To establish a priori distribution for high voltage ride-through parameters; Construct a likelihood function to evaluate the degree of matching of observed data under different parameter values; Update the posterior distribution according to Bayes' theorem; The optimal high-voltage ride-through parameters are obtained by solving the maximum a posteriori estimation.
7. The intelligent identification method for high voltage ride-through parameters of a grid-connected energy storage inverter according to claim 6, characterized in that, The expression for setting the prior distribution of high-voltage ride-through parameters is as follows: In the above formula, For high voltage ride-through parameters, Let be the prior mean vector. The prior covariance matrix; The expression for the likelihood function is: In the above formula, To observe the noise covariance matrix, This is the actual measurement value at time k+1. This is the dynamic state estimation of the inverter at time k+1. For measurement functions; The expression for updating the posterior distribution according to Bayes' theorem is: In the above formula, This represents all actual measurements from time 1 to time k+1. These are all actual measured values from time 1 to time k; The expression for the optimal high-voltage ride-through parameters obtained by solving the maximum a posteriori estimation is as follows: 。 8. The intelligent identification method for high voltage ride-through parameters of grid-connected energy storage inverters according to claim 1, characterized in that, The intelligent identification method for high-voltage ride-through parameters of grid-connected energy storage inverters also includes: The active current command and reactive current command are calculated based on the optimal high-voltage ride-through parameters, and the calculation expressions are as follows: In the above formula, This is an active current command. This represents the primary coefficient of the active current response during high voltage periods. These are the quadratic coefficients of the active current response during high voltage periods. This represents the active current bias value under high voltage conditions. This refers to the terminal voltage amplitude. This is the initial active current. This is a reactive current command. This represents the primary coefficient of the reactive current response during high voltage periods. These are the quadratic coefficients of the reactive current response during high voltage periods. This represents the reactive current bias value under high voltage conditions. To reach the high voltage ride-through threshold, This is the initial reactive current.
9. A smart identification system for high-voltage ride-through parameters of a grid-connected energy storage inverter, characterized in that, include: The data acquisition and preprocessing module is used to acquire the operating data of the inverter in real time and obtain the corresponding digital signal after preprocessing. The model building module is used to jointly construct an extended state-space model by combining the dynamic state of the inverter with the high voltage ride-through parameters to be estimated in the digital signal. The dynamic state refers to the behavior and response characteristics of the inverter during the transition from one steady state to another when external conditions change. The first parameter estimation module is used to update the extended state space model and use the updated high voltage ride-through parameters in the extended state space model as the preliminary estimation result. The second parameter estimation module is used to re-estimate the preliminary estimation results to obtain the optimal high-voltage ride-through parameters.
10. A computer-readable storage medium storing a computer program / instructions, characterized in that, The computer program / instructions are programmed or configured to execute, via a processor, the intelligent identification method for high voltage ride-through parameters of grid-connected energy storage inverters as described in any one of claims 1 to 8.