MILP voltage ride-through mode parameter identification method and device, storage medium and computer equipment

Abstract

The application discloses a MILP voltage ride-through mode parameter identification method and device, a storage medium and computer equipment. The method comprises the following steps: constructing a voltage ride-through control model, wherein the voltage ride-through control model comprises linear equations corresponding to multiple control modes, and the control modes comprise active power control modes and reactive power control modes; constructing a mixed integer linear programming model, wherein the mixed integer linear programming model represents the identification of the control modes through binary integer variables, takes the minimum absolute error as a target function, linearizes the target function, and embeds at least one physical constraint; and solving the mixed integer linear programming model by using a mixed integer linear programming solver to synchronously output the identification of the activated control modes and the parameter estimation value of the corresponding voltage ride-through control model. The above method can realize the synchronous identification of the control modes and the model parameters, and improve the accuracy of the voltage ride-through control model parameter identification.

Description

Technical Field

[0001] This application relates to the field of power system modeling and simulation technology, and in particular to a method, device, storage medium and computer equipment for identifying MILP voltage ride-through mode parameters. Background Technology

[0002] As the proportion of new energy power generation such as wind power, photovoltaics, and energy storage in the power system continues to increase, the performance of wind / solar / storage converters, as core interface devices, becomes crucial to system stability. Especially when grid faults cause voltage anomalies, the converters must possess reliable voltage ride-through capabilities to support grid voltage and prevent grid disconnection. However, voltage ride-through control logic is complex, typically involving multiple active and reactive power control modes, with significant differences in response characteristics and parameter settings across different modes.

[0003] Currently, parameter identification for voltage ride-through control models generally relies on the least squares method and its improved algorithms. This type of method estimates model parameters by minimizing the squared error between the predicted output and the actual measured value. Its principle is intuitive, its computational efficiency is high, and it can obtain effective parameter identification results under ideal conditions of good data quality and low noise interference.

[0004] However, parameter identification methods based on the least squares method are highly sensitive to outliers or impulse noise in the measurement data, and are prone to biased estimations in noisy environments, resulting in insufficient model generalization ability and robustness. Furthermore, these methods cannot naturally embed engineering physical constraints into the identification process, potentially leading to identified parameters that violate actual operational limitations. In addition, existing identification methods require pre-assuming or separately judging the control mode before identifying parameters under that control mode. This step-by-step identification method not only introduces subjective errors but also struggles to effectively handle complex scenarios where control modes may switch during transit. Summary of the Invention

[0005] In view of this, embodiments of this application provide a method, apparatus, storage medium and computer device for identifying parameters of MILP voltage ride-through mode, the main purpose of which is to solve the technical problems that the control mode and model parameters of the voltage ride-through control model cannot be identified synchronously, and that the model parameter identification accuracy is low and the robustness is insufficient.

[0006] According to one aspect of this application, a method for identifying MILP voltage ride-through mode parameters is provided, the method comprising: A voltage ride-through control model is constructed, wherein the voltage ride-through control model includes linear equations corresponding to multiple control modes, and the control modes include active power control mode and reactive power control mode; A mixed-integer linear programming model is constructed, wherein the control mode identifier is represented by a binary integer variable, the minimum absolute error is used as the objective function, the objective function is linearized, and at least one physical constraint is embedded. The mixed-integer linear programming model is solved using a mixed-integer linear programming solver to simultaneously output the identifier of the activated control mode and the parameter estimates of the corresponding voltage ride-through control model.

[0007] According to another aspect of this application, a MILP voltage ride-through mode parameter identification device is provided, the device comprising: A voltage ride-through control model construction module is used to construct a voltage ride-through control model, wherein the voltage ride-through control model includes linear equations corresponding to multiple control modes, and the control modes include active power control mode and reactive power control mode; A mixed-integer linear programming model construction module is used to construct a mixed-integer linear programming model, wherein the mixed-integer linear programming model uses binary integer variables to represent the identifier of the control mode, takes the minimum absolute error as the objective function, performs linearization processing on the objective function, and embeds at least one physical constraint. The voltage ride-through mode parameter identification module is used to solve the mixed-integer linear programming model using a mixed-integer linear programming solver, so as to synchronously output the identifier of the activated control mode and the corresponding parameter estimate of the voltage ride-through control model.

[0008] According to another aspect of this application, a storage medium is provided that stores a computer program thereon, which, when executed by a processor, implements the above-described MILP voltage ride-through mode parameter identification method.

[0009] According to another aspect of this application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor executes the program to implement the above-described MILP voltage ride-through mode parameter identification method.

[0010] By employing the above technical solutions, the MILP voltage ride-through mode parameter identification method, apparatus, storage medium, and computer equipment provided in this application, through the construction of a mixed-integer linear programming model and the introduction of binary integer variables into the mixed-integer linear programming model, can uniformly model the selection of control modes and parameter identification, thereby achieving synchronous output of control model selection and model parameter identification, thus avoiding the subjective errors caused by traditional step-by-step identification methods. Furthermore, by using the minimum absolute error, which is insensitive to outliers, as the objective function and performing linearization processing, the robustness of the model identification in high-noise measurement environments can be significantly enhanced. Simultaneously, by directly embedding physical constraints into the optimization model, it can be ensured that the identification results conform to actual engineering operating limitations. Based on this, the above method can achieve synchronous identification of control modes and model parameters under various complex operating conditions such as low-voltage ride-through and high-voltage ride-through, and effectively improve the accuracy and engineering applicability of voltage ride-through control model parameter estimation.

[0011] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, specific embodiments of this application are given below. Attached Figure Description

[0012] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings: Figure 1 A flowchart illustrating a MILP voltage ride-through mode parameter identification method provided in an embodiment of this application is shown. Figure 2 A flowchart illustrating another MILP voltage ride-through mode parameter identification method provided in an embodiment of this application is shown; Figure 3 This illustration shows a comparison of the parameter identification performance between a MILP voltage ride-through mode parameter identification method provided in this application and a parameter identification method based on the least squares method. Figure 4 This is the second schematic diagram comparing the parameter identification performance of a MILP voltage ride-through mode parameter identification method provided in this application embodiment with that of a parameter identification method based on the least squares method; Figure 5 This is the third schematic diagram showing a comparison of the parameter identification effects between a MILP voltage ride-through mode parameter identification method provided in this application embodiment and a parameter identification method based on the least squares method; Figure 6A schematic diagram of the structure of a MILP voltage ride-through mode parameter identification device provided in an embodiment of this application is shown. Detailed Implementation

[0013] The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in the embodiments of the present application can be combined with each other.

[0014] In one embodiment, such as Figure 1 As shown, a method for identifying MILP voltage ride-through mode parameters is provided. Taking the application of this method to a computer device as an example, the method includes the following steps: Step 101: Construct a voltage ride-through control model, which includes linear equations corresponding to various control modes, including active power control mode and reactive power control mode.

[0015] The voltage ride-through control model refers to a mathematical model describing the dynamic behavior of wind / solar / storage converters during periods of abnormal grid voltage (such as low-voltage or high-voltage ride-through). Its control modes mainly include an active power control mode for regulating active power output and a reactive power control mode for providing reactive power support to stabilize grid voltage. In this embodiment, each control mode can be characterized by a linear equation, which describes the quantitative relationship between control commands (such as active current and reactive current) and a series of input variables (such as fault voltage and pre-fault steady-state current / power).

[0016] Specifically, linear equations can be established for different control objectives based on the actual control logic of the converter and grid connection standards. For example, for active power control, equations for active current control mode and active power control mode can be established. Similarly, for reactive power control, linear equations for reactive current control mode and reactive power control mode can be established. The coefficients in these equations are the model parameters to be identified. By integrating these linear equations established for different control modes, a model set containing multiple control modes can be formed, thus providing a foundation for subsequent mode selection and parameter identification.

[0017] Step 102: Construct a mixed-integer linear programming model, wherein the mixed-integer linear programming model uses binary integer variables to represent the identifier of the control mode, takes the minimum absolute error as the objective function, performs linearization on the objective function, and embeds at least one physical constraint.

[0018] Mixed Integer Linear Programming (MILP) refers to a mathematical framework that includes both continuous and integer variables. In this model, the binary integer variables represent whether a specific control mode is activated; they can be understood as identifiers of the control mode, taking values ​​of 0 or 1. Minimum absolute error is an objective function that uses the sum of the absolute differences between predicted and measured values ​​as the loss function; it is insensitive to outliers. Physical constraints refer to operational limitations extracted from engineering practice that must be met, such as current limiting and consistency between power and current direction.

[0019] Specifically, a binary integer variable can be introduced to correspond to each predefined control mode, and the Big M method can be used to transform the linear equation corresponding to each control mode into a pair of linear inequality constraints. When the binary integer variable is 1, this pair of linear inequality constraints forces the predicted command to be equal to the output of the control mode equation; when the binary integer variable is 0, the constraint fails due to the presence of the large constant M. Simultaneously, to improve noise robustness, the objective function can be minimized using the mean absolute error. By introducing a pair of non-negative auxiliary variables representing positive and negative deviations for each data sample, the nonlinear absolute value objective can be transformed into an equivalent linear form, with corresponding constraints added. Finally, engineering physical constraints, such as the current command not exceeding the maximum allowable value, can be directly embedded into the model in the form of linear inequalities to form a complete mixed-integer linear programming model.

[0020] The above method couples the problems of mode selection and parameter fitting within a unified MILP framework. It can achieve conditional activation of control modes through the Big M method, and ensure the solvability of the problem through the linearization of the objective function. At the same time, it can ensure the engineering feasibility of the identification results by embedding physical constraints, thus fundamentally overcoming the shortcomings of traditional step-by-step identification methods.

[0021] Step 103: Solve the mixed-integer linear programming model using a mixed-integer linear programming solver to simultaneously output the identifier of the activated control mode and the parameter estimates of the corresponding voltage ride-through control model.

[0022] Among them, mixed-integer linear programming solvers refer to specialized computational software or algorithm libraries capable of solving mixed-integer linear programming problems, such as Gurobi, COPT, and CPLEX. Synchronous output refers to obtaining the optimal solutions for both discrete and continuous variables of the voltage ride-through control model simultaneously during a single solution process.

[0023] Specifically, the completed mixed-integer linear programming model, including all linear objective functions, equality and inequality constraints, can be input into a selected mixed-integer linear programming solver (hereinafter referred to as the solver). The solver can automatically and comprehensively search for the optimal solution by running its internal algorithm (such as branch and bound), while simultaneously determining the activated control mode at each data time point, i.e., finding the identifier of the control mode that minimizes the objective function, and calculating the optimal estimates of the coefficients of all linear equations under the activated control mode. The solution process does not require manual pre-specification or mode switching. After calculation, the computer can directly output two results: first, the identifiers of the control mode switching sequence throughout the entire observation period; and second, the precise set of model parameter values ​​corresponding to these modes.

[0024] The above method enables integrated and synchronous identification of control modes and model parameters. Furthermore, the solver can automatically match the actual control logic and parameters behind the data through global optimization, thereby avoiding subjective errors and mode misjudgments caused by manual step-by-step identification. Especially in complex scenarios with noise interference or dynamic switching of control modes, it can significantly improve the accuracy, robustness, and automation level of identification.

[0025] This embodiment constructs a mixed-integer linear programming model and introduces binary integer variables into it, enabling unified modeling of control mode selection and parameter identification. This achieves synchronous output of control model selection and model parameter identification, avoiding subjective errors inherent in traditional step-by-step identification methods. Furthermore, by employing the minimum absolute error, which is insensitive to outliers, as the objective function and performing linearization, the model's robustness in noisy measurement environments is significantly enhanced. Simultaneously, by directly embedding physical constraints into the optimization model, the identification results are ensured to conform to actual engineering operational limitations. Therefore, this method can achieve synchronous identification of control modes and model parameters under various complex operating conditions, such as low-voltage ride-through and high-voltage ride-through, effectively improving the accuracy and engineering applicability of voltage ride-through control model parameter estimation.

[0026] Furthermore, as a refinement and extension of the specific implementation of the above embodiments, and to fully illustrate the specific implementation process of this embodiment, a MILP voltage ride-through mode parameter identification method is provided, such as... Figure 2 As shown, the method includes: Step 201: Based on the actual control logic of the converter, establish linear equations corresponding to various control modes, and integrate the linear equations corresponding to various control models to obtain the voltage ride-through control model.

[0027] In this embodiment, linear equations corresponding to various control modes can be established based on the actual control logic of the converter. These linear equations can be used to describe the linear relationship between control commands and input variables. The control modes mainly include active power control mode and reactive power control mode. Then, the linear equations corresponding to various control models can be integrated to obtain a voltage ride-through control model. The parameters to be identified in the voltage ride-through control model are the coefficients of each linear equation.

[0028] In this embodiment, linear equations corresponding to multiple control modes can be established based on the actual control logic of at least one of the power generation converter, photovoltaic converter, and energy storage converter, so that the mixed integer linear programming model can be applied to the model parameter identification of the low voltage ride-through control process and high voltage ride-through control process of at least one of the wind power generation converter, photovoltaic converter, and energy storage converter.

[0029] In voltage ride-through control models, active power control mode and reactive power control mode are the two main categories, each containing sub-modes based on different physical quantities or control objectives. Their specific structure is as follows: Active power control mode refers to the control strategy used to regulate the active power output of the converter during voltage ride-through. It mainly includes two sub-modes: active current control mode and active power control mode. The active current control mode targets the active component of the output current. Its model typically establishes a linear relationship between the active current command and variables such as fault voltage and initial active current. This mode is direct and fast, often used to prioritize meeting current limits or provide rapid active power support. Conversely, the active power control mode targets the output active power. Its model typically correlates the active power command with the initial power and then converts it into a corresponding current command based on the fault voltage. This mode focuses on power-level control or limitation and can be used to smooth power fluctuations or execute specific power dispatch commands.

[0030] Reactive power control mode refers to the control strategy used during voltage ride-through to regulate the reactive power output of the converter to support the grid voltage. Its sub-modes are structurally symmetrical with active power control mode, mainly including reactive current control mode and reactive power control mode. The control objective of reactive current control mode is the reactive component of the output current. Its command is usually proportional to the voltage dip depth; generally, the deeper the voltage dip, the larger the required reactive current to meet the grid connection standards for dynamic reactive power support. Correspondingly, the control objective of reactive power control mode is the output reactive power. In some scenarios, the reactive power command value may be directly given and then converted into a current command.

[0031] In this embodiment, the voltage ride-through controller of the converter can be considered as a multi-mode switching system. At any given time, its active and reactive power circuits will each select one of the aforementioned control modes to operate. One of the core objectives of the MILP voltage ride-through mode parameter identification proposed in this embodiment is to automatically and accurately identify which control mode each circuit actually uses from the measured data, and to accurately estimate the specific parameters under that control mode.

[0032] For example, based on the actual control logic, the linear equation corresponding to the active power control mode established during low-voltage ride-through and high-voltage ride-through can be shown in Equation (1): (1) in, This indicates the active current command during fault ride-through. The grid voltage during the fault is the input variable of the model. This represents the steady-state active current before the fault occurs, and is also an input variable of the model; , These represent the coefficients of the first-order terms of the fault voltage and the initial active current, respectively, and are the parameters to be identified in the model. This is a constant coefficient, representing the reference or offset of the active current, and is also a parameter to be identified.

[0033] Step 202 introduces a binary integer variable to represent the identifier of the control mode, and transforms the linear equations corresponding to each control mode into linear constraints of a mixed integer linear programming model.

[0034] In this embodiment, for the m-th control mode, the following pair of constraints can be established: (2) (3) in, The predicted value for the control command, Let X be the linear equation corresponding to the m-th control mode, X be the input variable, and θ be the parameter to be identified. Let M be a binary integer variable representing whether the m-th mode is activated, where M is a positive number much larger than the parameter to be identified.

[0035] In the above constraints, when When =1, the above constraint condition is enforced. = ;when When =0, the above constraints apply. There are no practical restrictions on the value of .

[0036] For example, the Big M method can be used to transform the equations corresponding to the control mode into the following linear constraints: (4) (5) (6) (7) (8) in, This indicates the active current command during fault ride-through. , These represent the grid voltage during the fault and the steady-state active current before the fault, respectively, and are the input variables of the model. , , These are the parameters to be identified in the voltage ride-through control model; It is a known constant representing the start-up threshold voltage during fault ride-through. , The two parameters are binary integer variables, taking values ​​of 0 or 1. When both parameters are set to 1, they represent active current control mode and active power control mode, respectively. M is a positive number that is much larger than other values ​​in the model and is used to linearize the condition constraints.

[0037] In the above example, formulas (4) and (5) can form the activation and deactivation constraints for the active current control mode. Wherein, The term represents the difference between the threshold voltage and the fault voltage, indicating the voltage sag depth. The active current command value is linearly related to the voltage sag depth.

[0038] when At this time, the active current control mode is activated. The term is 0. Two constraint inequalities force active current command. It must be equal to the current model calculation value on the right.

[0039] when At this time, the active current control mode is turned off. Item becomes Since M is a very large positive number, the right-hand side of constraint 1 becomes very large, and the right-hand side of constraint 2 becomes very small. This makes these two inequalities affect the active current command. The value of is essentially unrestricted, meaning that the equation corresponding to the active current control mode is shut down.

[0040] Furthermore, equations (6) and (7) can form the activation and deactivation constraints for the active power control mode. In this mode, the active current command aims to achieve or track a target power. Its constraint inequalities include... This item reflects the target active power at fault voltage. The corresponding current value.

[0041] when When the active power control mode is activated, the forced current command is equal to the output of the power model.

[0042] when At this time, the active power control mode is turned off, and the constraint fails.

[0043] Furthermore, formula (8) mandates that the sum of the two binary integer variables must be 1. This means that in any given identification scenario, only one active power control mode can be activated, and the binary integer variable corresponding to that control mode is 1. This accurately simulates the physical reality that the converter can only operate under one dominant control mode at any given time.

[0044] The above set of formulas forms the core of the MILP identification framework. By introducing a binary integer variable z and a large constant M, the discrete control mode selection problem and the continuous parameter fitting problem can be successfully coupled into the same linear optimization model. At this point, the task of the optimization solver is to adjust the continuous parameters and the discrete mode variable z, while satisfying all physical constraints (such as current limiting), to make the current command output by the model... The overall error between the solution and the measured data is minimized. During the optimization process, the solver automatically selects the best-matching pattern (the pattern with z=1) for each data point or time period and provides the optimal parameter values ​​for that pattern. This avoids the error of subjectively judging the pattern first and then fitting the parameters in the traditional two-step method, thus achieving robust and accurate integrated identification.

[0045] Step 203: Construct the objective function using the mean absolute error, and introduce auxiliary continuous variables to linearize the objective function.

[0046] In this step, the mean absolute error can be constructed as the objective function, and a pair of auxiliary continuous variables representing positive and negative deviations can be introduced for each data sample. Then, the active current error and reactive current error in the objective function are replaced with the sum of the positive and negative deviations to transform the objective function into a linear form.

[0047] In this embodiment, to enhance noise immunity, the mean absolute error (MAE) can be used as the objective function. For example, the objective function can be expressed in the following form: (9) in, The total number of data samples used for identification. Sample index, i=1,2,…,N, This represents the predicted active current command value calculated based on the voltage ride-through control model to be identified at the i-th sample point. This represents the measured value of the active current command obtained from the actual operating data of the converter at the i-th sample point. This represents the predicted reactive current command value calculated based on the model to be identified at the i-th sample point. This represents the measured value of the reactive current command at the i-th sample point.

[0048] Formula (9) defines the objective function to be minimized, which is constructed based on the mean absolute error. It can calculate the average absolute error between the active and reactive current commands output by the model and their corresponding measured values. By using the mean absolute error instead of the mean square error, this formula can effectively suppress outlier interference and improve the robustness of the identification model in noisy environments.

[0049] Furthermore, this can be achieved by introducing auxiliary continuous variables. and (representing positive and negative deviations respectively), transforming the above nonlinear objective into a linear form: (10) in, and satisfy: (11) (12) in, This represents the positive deviation of the predicted active current value from the measured value at the i-th sample point. This variable equals the difference when the predicted value is greater than the measured value; otherwise, it is 0. For the i-th sample point, the negative deviation of the predicted active current value from the measured value is given by the variable. When the predicted value is less than the measured value, this variable is equal to the absolute value of the difference; otherwise, it is 0. denoted as the positive deviation of the reactive current prediction value at the i-th sample point; The negative deviation of the reactive current prediction value at the i-th sample point.

[0050] Formula (10) is the linearized MILP objective function, and formulas (11) and (12) are constraints defining the deviation variables. These two equality constraints establish the mathematical relationship between the prediction error and the auxiliary deviation variables, and together with formula (10), they constitute a complete linear objective function system. The above formulas are also the linear equivalent of formula (9). By introducing a pair of non-negative auxiliary variables ( ) for each data sample's active and reactive current errors respectively. and The absolute value operation can be transformed into a linear summation of these variables, thereby transforming the originally nonlinear optimization objective into a linear objective function that can be directly processed by the MILP solver.

[0051] Step 204: Embed physical constraints, including current limiting constraints and power and current change direction consistency constraints, to complete the construction of the mixed integer linear programming model.

[0052] In this step, after completing the control mode modeling and objective function linearization, physical constraints to ensure the engineering rationality of the model can be embedded into the MILP framework. These embedded physical constraints can include at least one of the following: current limiting constraints, meaning the absolute value of the commanded current must not exceed the maximum value allowed by the converter hardware; power and current change direction consistency constraints, ensuring that the change direction of active / reactive power commands is consistent with the change direction of their corresponding current command components; command value sign matching constraints, for example, requiring the sign of the reactive current command supporting the grid voltage to match the sign of the voltage deviation; and current change rate limiting constraints, to smooth the command output, which can be represented by the change in command values ​​between consecutive sampling points not exceeding a given threshold. These physical constraints can all be expressed as linear equations or inequalities and added as hard constraints to the mixed-integer linear programming model, thus completing the construction from a mathematical model to an engineering-practical model.

[0053] By using the aforementioned physical constraints as hard boundary conditions, it can be fundamentally ensured that the identified control modes and parameter combinations can meet the safety limits and basic physical laws of actual equipment operation, eliminating the possibility of generating engineering-infeasible or dangerous parameters, and greatly improving the practicality and reliability of the identification results.

[0054] Step 205: Solve the mixed-integer linear programming model using a mixed-integer linear programming solver to simultaneously output the identifier of the activated control mode and the parameter estimates of the corresponding voltage ride-through control model.

[0055] In this step, after the model is built, a professional mixed-integer linear programming solver, such as Gurobi, COPT, or CPLEX, can be called. The complete mixed-integer linear programming model, including the linear objective function, mixed-integer variables, and all linear constraints, is input into the solver. The solver can automatically perform a global search using algorithms such as branch and bound and cutting planes, and simultaneously determine the optimal control mode at each moment of the entire observation period during a single solution process. This determines which binary integer variables are set to 1, and the optimal estimates of the coefficients of all linear equations under the corresponding mode. After the solution is completed, two key results can be output: first, the activation sequence of the control mode, which identifies the control mode adopted by the converter at different stages; and second, the estimated model parameters, i.e., the precise set of coefficients. Together, these two results constitute a complete, accurate, and reproducible voltage ride-through control model.

[0056] The above method can avoid subjective errors caused by prior model assumptions and can automatically handle the dynamic switching of control modes during fault conditions. For example, Figure 3 , Figure 4 and Figure 5 This is a comparison chart of the parameter identification results between the MILP voltage ride-through mode parameter identification method proposed in this embodiment and the existing least squares (LS) identification method. Figure 3 , Figure 4 and Figure 5 As can be seen, under multiple operating conditions, the identification method proposed in this embodiment can construct a voltage ride-through control model that is closer to the actual characteristics.

[0057] The above embodiments, by constructing a mixed-integer linear programming model and introducing binary integer variables into it, can uniformly model the selection of control modes and parameter identification, thereby achieving synchronous output of control model selection and model parameter identification, thus avoiding the subjective errors caused by traditional step-by-step identification methods. Furthermore, by using the minimum absolute error, which is insensitive to outliers, as the objective function and performing linearization, the robustness of the model identification in noisy measurement environments can be significantly enhanced. Simultaneously, by directly embedding physical constraints into the optimization model, it can be ensured that the identification results conform to actual engineering operating limitations. Based on this, the above method can achieve synchronous identification of control modes and model parameters under various complex operating conditions such as low-voltage ride-through and high-voltage ride-through, and effectively improve the accuracy and engineering applicability of voltage ride-through control model parameter estimation.

[0058] It should be noted that the user information (including but not limited to device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the labels corresponding to each step in the above embodiments are only for identification purposes and are not intended to limit the order of execution of the steps. The order of execution of the steps in each embodiment can be set according to the actual situation.

[0059] Furthermore, as Figures 1 to 5 To specifically implement the method, this application provides a MILP voltage ride-through mode parameter identification device, such as... Figure 6 As shown, the device includes: The voltage ride-through control model construction module 31 can be used to construct a voltage ride-through control model, wherein the voltage ride-through control model includes linear equations corresponding to multiple control modes, and the control modes include active power control mode and reactive power control mode. The mixed-integer linear programming model construction module 32 can be used to construct a mixed-integer linear programming model, wherein the mixed-integer linear programming model uses binary integer variables to represent the identifier of the control mode, takes the minimum absolute error as the objective function, performs linearization processing on the objective function, and embeds at least one physical constraint. The voltage ride-through mode parameter identification module 33 can be used to solve the mixed-integer linear programming model using a mixed-integer linear programming solver, so as to synchronously output the identifier of the activated control mode and the parameter estimate of the corresponding voltage ride-through control model.

[0060] In specific application scenarios, the voltage ride-through control model construction module 31 can be used to establish linear equations corresponding to multiple control modes based on the actual control logic of the converter. The linear equations are used to describe the linear relationship between control commands and input variables. The control modes include active power control mode and reactive power control mode. The linear equations corresponding to the multiple control modes are integrated to obtain the voltage ride-through control model. The parameters to be identified in the voltage ride-through control model are the coefficients of each linear equation.

[0061] In specific application scenarios, the voltage ride-through control model construction module 31 can be used to establish linear equations corresponding to multiple control modes based on the actual control logic of at least one converter among the power generation converter, photovoltaic converter, and energy storage converter, so that the mixed integer linear programming model can be applied to the identification of model parameters for the low voltage ride-through control process and the high voltage ride-through control process of at least one converter among the wind power generation converter, photovoltaic converter, and energy storage converter.

[0062] In specific application scenarios, the mixed-integer linear programming model construction module 32 can be used to introduce binary integer variables to represent the identifier of the control mode, and transform the linear equations corresponding to each control mode into the linear constraints of the mixed-integer linear programming model; construct the objective function using the mean absolute error, and introduce auxiliary continuous variables to linearize the objective function; embed physical constraints including current limiting constraints and power and current change direction consistency constraints to complete the construction of the mixed-integer linear programming model.

[0063] In specific application scenarios, the mixed-integer linear programming model construction module 32 can be used to establish the following pair of constraints for the m-th control mode: ; ; in, The predicted value for the control command, Let X be the linear equation corresponding to the m-th control mode, X be the input variable, and θ be the parameter to be identified. M is a binary integer variable representing whether the m-th mode is activated, where M is a positive number much larger than the parameter to be identified. when When =1, the constraint condition is enforced. = ;when When =0, the constraint condition is... There are no practical restrictions on the value of .

[0064] In specific application scenarios, the mixed-integer linear programming model construction module 32 can be used to construct the mean absolute error as the objective function; introduce a pair of auxiliary continuous variables representing positive and negative deviations for each data sample; and replace the active current error and reactive current error in the objective function with the sum of positive and negative deviations so that the objective function is transformed into a linear form.

[0065] In specific application scenarios, the mixed-integer linear programming model construction module 32 can be used to embed instruction value sign matching constraints and / or current change rate limit constraints into the physical constraints of the mixed-integer linear programming model.

[0066] It should be noted that other corresponding descriptions of the functional units involved in the MILP voltage ride-through mode parameter identification device provided in this application embodiment can be found in [reference]. Figures 1 to 5 The corresponding descriptions in the methods will not be repeated here.

[0067] This application also provides a computer device, specifically a personal computer, server, network device, etc. The computer device includes a bus, processor, memory, and communication interface, and may also include input / output interfaces and a display device. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device stores location information. The network interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements the steps in the various method embodiments.

[0068] Those skilled in the art will understand that the structure of the computer device described above is only a partial structure related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. A specific computer device may include more or fewer components, or combine certain components, or have different component arrangements.

[0069] In one embodiment, a computer-readable storage medium is provided, which may be non-volatile or volatile, having stored thereon a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0070] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0071] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, graphics processors, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0072] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0073] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for identifying MILP voltage ride-through mode parameters, characterized in that, The method includes: A voltage ride-through control model is constructed, wherein the voltage ride-through control model includes linear equations corresponding to multiple control modes, and the control modes include active power control mode and reactive power control mode; A mixed-integer linear programming model is constructed, wherein the control mode identifier is represented by a binary integer variable, the minimum absolute error is used as the objective function, the objective function is linearized, and at least one physical constraint is embedded. The mixed-integer linear programming model is solved using a mixed-integer linear programming solver to simultaneously output the identifier of the activated control mode and the parameter estimates of the corresponding voltage ride-through control model.

2. The method according to claim 1, characterized in that, The construction of the voltage ride-through control model includes: Based on the actual control logic of the converter, linear equations corresponding to various control modes are established. These linear equations describe the linear relationship between control commands and input variables. The control modes include active power control mode and reactive power control mode. The linear equations corresponding to the various control models are integrated to obtain the voltage ride-through control model, wherein the parameters to be identified in the voltage ride-through control model are the coefficients of each linear equation.

3. The method according to claim 2, characterized in that, Based on the actual control logic of the converter, linear equations corresponding to various control modes are established, including: Based on the actual control logic of at least one of the power generation converters, photovoltaic converters, and energy storage converters, linear equations corresponding to multiple control modes are established so that the mixed integer linear programming model can be applied to the model parameter identification of the low voltage ride-through control process and the high voltage ride-through control process of at least one of the power generation converters, photovoltaic converters, and energy storage converters.

4. The method according to any one of claims 1 to 3, characterized in that, The construction of the mixed-integer linear programming model includes: A binary integer variable is introduced to represent the identifier of the control mode, and the linear equations corresponding to each control mode are transformed into linear constraints of the mixed integer linear programming model. The objective function is constructed using the mean absolute error, and auxiliary continuous variables are introduced to linearize the objective function. The mixed-integer linear programming model is constructed by embedding physical constraints, including current limiting constraints and power-current change direction consistency constraints.

5. The method according to claim 4, characterized in that, The introduction of binary integer variables to represent the identifiers of the control modes, and the transformation of the linear equations corresponding to each control mode into linear constraints of the mixed-integer linear programming model, includes: For the m-th control mode, the following pair of constraints are established: ； ； in, The predicted value for the control command, Let X be the linear equation corresponding to the m-th control mode, X be the input variable, and θ be the parameter to be identified. M is a binary integer variable representing whether the m-th mode is activated, where M is a positive number much larger than the parameter to be identified. when When =1, the constraint condition is enforced. = ;when When =0, the constraint condition is... There are no practical restrictions on the value of .

6. The method according to claim 4, characterized in that, The step of constructing the objective function using the mean absolute error and introducing auxiliary continuous variables to linearize the objective function includes: Construct the mean absolute error as the objective function; For each data sample, introduce a pair of auxiliary continuous variables representing positive and negative deviations; The active current error and reactive current error in the objective function are replaced with the sum of positive and negative deviations, so that the objective function is transformed into a linear form.

7. The method according to claim 4, characterized in that, The physical constraints embedded in the mixed-integer linear programming model also include instruction value sign matching constraints and / or current change rate limit constraints.

8. A MILP voltage ride-through mode parameter identification device, characterized in that, The device includes: A voltage ride-through control model construction module is used to construct a voltage ride-through control model, wherein the voltage ride-through control model includes linear equations corresponding to multiple control modes, and the control modes include active power control mode and reactive power control mode; A mixed-integer linear programming model construction module is used to construct a mixed-integer linear programming model, wherein the mixed-integer linear programming model uses binary integer variables to represent the identifier of the control mode, takes the minimum absolute error as the objective function, performs linearization processing on the objective function, and embeds at least one physical constraint. The voltage ride-through mode parameter identification module is used to solve the mixed-integer linear programming model using a mixed-integer linear programming solver, so as to synchronously output the identifier of the activated control mode and the corresponding parameter estimate of the voltage ride-through control model.

9. A storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the method of any one of claims 1 to 7.

10. A computer device, comprising a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, characterized in that, When the processor executes the computer program, it implements the method of any one of claims 1 to 7.

Images