Distributed model predictive control method, device, equipment and medium for wind energy storage converter

Abstract

The present invention relates to the field of wind storage converter control technology, and discloses a distributed model predictive control method, device, equipment and medium for wind storage converters. The method includes: obtaining state data of a target converter, the target converter being any wind storage converter in a target wind power generation system; constructing a distributed predictive model of the target converter based on the state data and designing a control target of the target converter based on the model; determining an objective function of the target converter based on the control target; determining constraints based on power coordination control requirements among wind storage converters, preset grid dispatching power and preset converter power thresholds; at each control moment, optimizing and solving based on the distributed predictive model, objective function, constraints and other total optimization control targets of wind storage converters, obtaining an optimal control solution at the current control moment and controlling the target converter accordingly based on the optimal control solution. The present invention has the advantages of flexible controllability, fast response, high efficiency and excellent performance.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power storage converter control, and in particular to a distributed model predictive control method, device, equipment and medium for wind power storage converters. Background Art

[0002] At present, the converter decoupling control algorithm with PI (proportional integral) control as the core is the mainstream control method used by the grid-side converter of the wind power generation system. This method adopts a voltage outer loop-current inner loop dual-loop control structure. The voltage outer loop realizes the stable control of the DC side capacitor voltage, and its output is used as the reference value of the current inner loop. The rapidity of the current inner loop is used to timely adjust the current on the AC side, suppress the influence of load disturbances, and enable the actual current to quickly track the current reference value, thereby achieving unity power factor control. However, there is currently no simple and effective way to adjust the PI controller parameters, which often requires designers to continuously debug, and the dynamic performance of the PI controller is poor.

[0003] With the improvement of computing power of digital real-time control chips, computationally intensive model predictive control (MPC) has gradually become a research hotspot. Model predictive control is an optimal control strategy. Its finite-time rolling optimization feature can respond to changes in system states in real time and suppress various disturbances of the system online. The model predictive control algorithms currently applied to converter systems mainly include two methods, including finite control set model predictive control and continuous control set model predictive control. Specifically, the finite control set model predictive control directly models the switching devices of the converter, uses the traversal method to calculate the converter response corresponding to each switch combination, and selects the switch combination closest to the expected response for control. This method has simple modeling and fast dynamic response. However, since all switch combinations need to be traversed at each control moment, the amount of calculation is large and the switching frequency is high; the continuous control set model predictive control method adopts the classic MPC algorithm structure. After obtaining the control solution by optimizing the cost function, the actual control signal is generated through the converter modulation technology. However, due to the complex modeling process of each converter and the wind power generation system that includes a large number of wind turbines and wind storage converters, that is, a large multi-machine system mainly adopts a centralized model predictive control method, there are performance problems such as low control efficiency, poor effect, and slow response speed, which makes it difficult to ensure the stable and efficient operation of wind turbines. Summary of the invention

[0004] In view of this, the present invention provides a distributed model predictive control method, device, equipment and medium for wind energy storage inverters to solve the problems of poor adaptability and reliability of the existing centralized model predictive control method for multi-machine large systems, slow response speed, low control efficiency, and difficulty in meeting the flexible, efficient and stable control requirements of wind turbines.

[0005] In a first aspect, the present invention provides a distributed model predictive control method for a wind power storage converter, the method comprising:

[0006] Acquire state data of a target converter, where the target converter is any wind storage converter in a target wind power generation system;

[0007] Building a distributed prediction model of the target converter based on the state data, and designing a control target of the target converter based on the distributed prediction model;

[0008] determining an objective function of a target converter based on a control objective;

[0009] Determine the constraint conditions based on the power coordination control requirements among the wind-storage converters, the preset grid dispatching power and the preset converter power threshold;

[0010] At each control moment, the optimal control solution at the current control moment is obtained by optimizing and solving the distributed prediction model, objective function, constraint conditions and the overall optimal control objectives of other wind-storage converters;

[0011] The target converter is controlled accordingly based on the optimal control solution.

[0012] The present invention obtains the status data of any wind storage inverter, i.e., the target inverter, in the target wind power generation system, builds a distributed prediction model based on its status data, and determines the objective function and constraints of the target inverter respectively in consideration of the coordinated control requirements among multiple inverters, which are jointly used to optimize the control objectives and obtain the optimal control solution to control the target inverter accordingly. This can realize distributed model predictive control of large multi-machine systems, which not only realizes flexible control of wind storage inverters, but also improves their control accuracy and efficiency, improves control performance, and satisfies the flexible, efficient and stable control requirements of wind turbines to a great extent.

[0013] In an optional implementation, the state data includes a converter output current, a converter output voltage, a grid voltage, and a grid frequency; a distributed prediction model of a target converter is constructed based on the state data, and a control target of the target converter is designed based on the distributed prediction model, including:

[0014] Performing coordinate transformation on the state data to obtain an expression of the converter output current in a dq coordinate system, wherein the expression of the converter output current includes corresponding relationships between the d-axis component and the q-axis component of the converter output current and the converter output voltage, the grid voltage, the grid frequency, the preset converter inductance, and the preset resistance, respectively;

[0015] Calculating power data of the target converter, wherein the power data includes active power and reactive power;

[0016] The converter output current is set as the state variable, the converter output voltage is set as the control variable, the grid voltage is set as the disturbance, the d-axis component, the q-axis component, the active power and the reactive power of the converter output current are set as the output, and a state space equation of the target converter is established based on the state variable, the control variable, the disturbance and the output;

[0017] The state space equation is discretized and recursively processed in the time domain to obtain a distributed prediction model, wherein the control objective of the target converter is to track the state variables in the distributed prediction model and to limit the control variables.

[0018] The present invention obtains the output current expression of the converter in the dq coordinate system through coordinate transformation, constructs the state space equation of the target converter in combination with the acquired active power and reactive power, discretizes the state space equation to obtain a distributed prediction model, and converts the nonlinear mechanism model into a linear model, which helps to improve the performance of distributed model predictive control and ensure the smooth operation of the wind turbine set.

[0019] In an optional implementation, determining the objective function of the target converter based on the control objective includes:

[0020] Obtaining a tracking error of a converter output current in a prediction time domain, wherein the tracking error is determined by an actual value and a reference value of the converter output current of the target converter;

[0021] Obtaining a control increment of the converter output voltage in a control time domain, wherein the control increment is determined based on a to-be-solved value of the converter output voltage of the target converter;

[0022] The tracking error and the control increment are weightedly summed to obtain the objective function of the target converter.

[0023] The present invention tracks the output current of the converter in the prediction time domain, takes into account the limitation of the control increment size of the converter output voltage in the control time domain, obtains the objective function of the target converter by weighted summation of the tracking error and the control increment, and can ensure the accuracy of the objective function.

[0024] In an optional implementation manner, the process of determining the reference value of the converter output current includes:

[0025] The target converter implements the virtual synchronous generator control strategy to obtain the reference value of the converter output current in the dq coordinate system, wherein the virtual synchronous generator control includes frequency modulation control of active power and frequency, and excitation regulation control of reactive power and voltage; wherein the frequency modulation control is used to obtain the optimal phase angle and perform coordinate transformation based on the optimal phase angle; the excitation regulation control is used to obtain the electromotive force, and the electromotive force is obtained through a proportional-integral controller to obtain the reference value of the converter output current in the dq coordinate system.

[0026] The present invention adopts a virtual synchronous generator control strategy to enable the target converter to have the inertial damping characteristics of the virtual synchronous generator to achieve grid-type control of the wind storage converter. This can greatly ensure the accuracy of the reference value of the converter output current, help improve the subsequent distributed model predictive control accuracy, and thus meet the flexible, efficient and stable control requirements of the wind turbine set.

[0027] In an optional implementation, the constraint condition is determined based on the power coordination control requirements among the wind power storage converters, the preset grid dispatching power and the preset converter power threshold, including:

[0028] Determine the coupling constraints of the target converter based on the power coordination control requirements between each wind storage converter and the preset grid dispatching power;

[0029] The coupling constraints are optimized based on the preset slack variables to obtain soft constraints;

[0030] determining a power constraint based on a preset converter power threshold;

[0031] Soft constraints and power constraints are determined as constraint conditions.

[0032] The present invention takes into account the coordination between the output power of each wind turbine and each wind-storage inverter in the energy storage system and the corresponding determined constraints of the wind-storage inverter output power limitation, which can ensure the feasibility of the control optimization problem and improve the efficiency of distributed model predictive control to a certain extent.

[0033] In an optional implementation, at each control moment, an optimization solution is performed based on the distributed prediction model, the objective function, the constraints and the overall optimization control objectives of other wind-storage converters to obtain the optimal control solution at the current control moment, including:

[0034] Penalize the preset slack variables to obtain the corresponding penalty items, and obtain the penalty values ​​of the penalty items in the prediction time domain;

[0035] Obtaining the objective functions corresponding to all other wind-storage converters except the target converter in the target wind power generation system, and performing weighted summation on them to obtain the overall optimization control objectives of other wind-storage converters;

[0036] Based on the objective function, penalty value and other overall optimization control objectives of the wind-storage converter, a target cost function is obtained;

[0037] At each control moment, an iterative solution is performed based on the target cost function, distributed prediction model and constraints until the preset iteration requirements are met to obtain the optimal control solution for the current control moment.

[0038] The present invention takes into account the penalty for preset slack variables in the optimization conditions, as well as the influencing factors of other wind storage converters in the target wind power generation system on the target converter, constructs a target cost function and solves the optimal control target, which can ensure the effectiveness and rationality of the control target, has the advantages of flexible controllability, fast response, high efficiency and excellent performance, and ensures the stable operation of the wind turbine set.

[0039] In an optional implementation, at each control moment, an iterative solution is performed based on the target cost function, the distributed prediction model, and the constraint conditions until the preset iteration requirements are met to obtain the optimal control solution at the current control moment, including:

[0040] At each control moment, the Nash optimal algorithm is used to iterate the control variables of the target converter, and the optimal value of the current iteration is obtained by combining the target cost function, the distributed prediction model and the constraints.

[0041] Calculate the current difference between the optimization value of the current iteration and the optimization value of the previous iteration;

[0042] Count the number of iterations of the current iteration;

[0043] Determine whether the current difference is greater than a preset accuracy threshold and whether the number of iterations is less than a preset number threshold;

[0044] When the current difference is not greater than a preset accuracy threshold, or when the number of iterations is not less than a preset number threshold, the optimized value of the current iteration is determined as the optimal control solution at the current control moment.

[0045] The present invention iteratively solves the target converter through the Nash optimal algorithm, and can also take into account the optimal solution of other wind storage converters in the target wind power generation system, which helps to improve the accuracy and efficiency of distributed model predictive control.

[0046] In a second aspect, the present invention provides a distributed model predictive control device for a wind power storage converter, the device comprising:

[0047] An acquisition module, used for acquiring status data of a target converter, where the target converter is any wind storage converter in a target wind power generation system;

[0048] A construction module, used for constructing a distributed prediction model of a target converter based on the state data, and designing a control target of the target converter based on the distributed prediction model;

[0049] A first determination module, configured to determine an objective function of a target converter based on a control objective;

[0050] The second determination module is used to determine the constraint conditions based on the power coordination control requirements between the wind storage converters, the preset grid dispatching power and the preset converter power threshold;

[0051] A solution module is used to optimize and solve at each control moment based on the distributed prediction model, objective function, constraint conditions and the overall optimization control objectives of other wind-storage converters to obtain the optimal control solution at the current control moment;

[0052] The control module is used to control the target converter accordingly based on the optimal control solution.

[0053] The distributed model predictive control device for wind storage inverters of the present invention can realize distributed model predictive control of any wind storage inverter in the target wind power generation system, has the advantages of flexible control, fast response, high efficiency and excellent performance, and meets the flexible, efficient and stable control requirements of wind turbines.

[0054] In a third aspect, the present invention provides a computer device, comprising: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to execute a distributed model predictive control method for a wind storage inverter according to the first aspect or any corresponding embodiment thereof.

[0055] In a fourth aspect, the present invention provides a computer-readable storage medium having computer instructions stored thereon, the computer instructions being used to enable a computer to execute a distributed model predictive control method for a wind storage inverter according to the first aspect or any corresponding embodiment thereof. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0057] Figure 1 is a flow chart of a distributed model predictive control method for a wind power storage converter according to an embodiment of the present invention;

[0058] Figure 2 is a flow chart of another wind storage converter distributed model predictive control method according to an embodiment of the present invention;

[0059] Figure 3 is a schematic structural diagram of active power-frequency control according to an embodiment of the present invention;

[0060] Figure 4 is a schematic structural diagram of reactive power-voltage control according to an embodiment of the present invention;

[0061] Figure 5 It is a schematic diagram of a process of solving Nash optimality using an iterative method according to an embodiment of the present invention;

[0062] Figure 6 is a structural block diagram of a distributed model predictive control device for a wind power storage converter according to an embodiment of the present invention;

[0063] Figure 7 It is a schematic diagram of the structure of a computer device according to an embodiment of the present invention. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0065] An embodiment of the present invention provides an embodiment of a distributed model predictive control method for a wind energy storage inverter. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0066] In this embodiment, a distributed model predictive control method for a wind power storage converter is provided. Figure 1 is a flow chart of a distributed model predictive control method for a wind power storage converter according to an embodiment of the present invention. Figure 1 As shown, the process includes the following steps:

[0067] Step S101, obtaining status data of a target converter, where the target converter is any wind-storage converter in a target wind power generation system.

[0068] It should be noted that the wind-storage converter in this embodiment is also called a grid-side converter, which is used to directly control the power exchange between the wind turbine or energy storage and the grid in the wind power generation system.

[0069] In this embodiment, the specific content and acquisition method of the status data are not limited here, and are determined based on actual project requirements and conventional data acquisition methods in the field. For example, the status data includes the current flowing into the grid from the wind storage converter, that is, the converter output current, which can be collected by a current sensor, which is only used as an exemplary description.

[0070] Step S102: constructing a distributed prediction model of the target converter based on the state data, and designing a control target of the target converter based on the distributed prediction model.

[0071] In this embodiment, the distributed prediction model characterizes the state changes of the wind storage converter at different times and can be constructed by the converter state space equation. It should be noted that the control target in this embodiment can be adaptively set according to actual needs and is not specifically limited here. For example, the control target is the output voltage of the wind storage converter, which is only used as an exemplary illustration.

[0072] Step S103: determining the objective function of the target converter based on the control objective.

[0073] It should be noted that the objective function is an optimization criterion or optimization standard in the optimization process. The objective function in this embodiment is adaptively determined based on the specific type of the actual control target.

[0074] Step S104, determining constraint conditions based on the power coordination control requirements among the wind power storage converters, the preset grid dispatching power and the preset converter power threshold.

[0075] It should be noted that the specific values ​​of the preset grid dispatching power and the preset converter power threshold in this embodiment are not limited here, and are set based on the actual power value that can be dispatched by the grid and the specific model adaptability of the wind storage converter in the wind power generation system.

[0076] Step S105: At each control moment, an optimization solution is performed based on the distributed prediction model, objective function, constraint conditions and other overall optimization control objectives of the wind-storage converter to obtain the optimal control solution at the current control moment.

[0077] In this embodiment, the specific method of optimization solution is not limited here, and can be determined based on conventional optimization algorithms in the field and adaptability to actual needs. For example, optimization algorithms include genetic algorithms, particle swarm algorithms, ant colony algorithms, simulated annealing algorithms, Nash equilibrium theory algorithms, etc., which are only used as exemplary illustrations.

[0078] Step S106: performing corresponding control on the target converter based on the optimal control solution.

[0079] In this embodiment, assuming that the control target is the output voltage of the converter, the optimal converter output voltage is modulated to obtain a PWM (Pulse Width Modulation) signal, and the switch tube of the grid-side converter is controlled based on the PWM signal to achieve corresponding control of the target converter. In the embodiment of the present invention, by obtaining the state data of any wind storage converter, that is, the target converter, in the target wind power generation system, a distributed prediction model is constructed based on its state data, and the objective function and constraint conditions of the target converter are determined, which are used together to optimize the control target and obtain the optimal control solution to control the target converter accordingly, so as to realize the distributed model predictive control of a large multi-machine system, which not only realizes the flexible control of the wind storage converter, but also improves its control accuracy and efficiency, improves the control performance, and greatly meets the flexible, efficient and stable control requirements of the wind turbine.

[0080] In this embodiment, a distributed model predictive control method for a wind power storage converter is provided. Figure 2 FIG. 1 is a flow chart of another wind storage converter distributed model predictive control method according to an embodiment of the present invention. Figure 2 As shown, the process includes the following steps:

[0081] Step S201, obtaining the state data of the target converter, the target converter is any wind storage converter in the target wind power generation system. Figure 1 Step S101 of the illustrated embodiment will not be described in detail here.

[0082] Step S202 : constructing a distributed prediction model of the target converter based on the state data, and designing a control target of the target converter based on the distributed prediction model.

[0083] In this embodiment, the state data includes the converter output current, the converter output voltage, the grid voltage and the grid frequency. Specifically, the above step 202 includes:

[0084] Step A1, transform the state data into coordinates to obtain the converter output current expression in the dq coordinate system, wherein the converter output current expression includes the correspondence between the d-axis component and the q-axis component of the converter output current and the converter output voltage, the grid voltage, the grid frequency, the preset converter inductance and the preset resistance, respectively.

[0085] In this embodiment, grid voltage oriented vector control is adopted to convert the three-phase AC quantity (ie, state data) of the target converter into a two-phase rotating coordinate system (ie, dq coordinate system).

[0086] It should be noted that the preset converter inductance and preset resistance in this embodiment are respectively the inductance and resistance inside the target converter, and their specific data are fixed values, which can be determined based on the adaptability of the specific model of the converter actually used.

[0087] Step A2: Calculate power data of the target converter, wherein the power data includes active power and reactive power.

[0088] In this embodiment, the power data is the active power and reactive power absorbed by the target converter from the power grid.

[0089] Step A3, let the converter output current be the state variable, the converter output voltage be the control variable, the grid voltage be the disturbance, the d-axis component, q-axis component, active power and reactive power of the converter output current be the output quantities, and establish the state space equation of the target converter based on the state variable, control variable, disturbance and output quantity.

[0090] Step A4, discretizing the state space equation and recursively extrapolating it in the time domain to obtain a distributed prediction model, wherein the control objective of the target converter is to track the state variables in the distributed prediction model and to limit the control variables.

[0091] It should be noted that, in this embodiment, it is considered that the default grid voltage does not change within the short control time domain, so the state space equation can be discretized based on this.

[0092] In a specific embodiment, the target converter output current expression in the dq coordinate system is obtained by adopting the grid voltage oriented vector control, which is specifically:

[0093] (1)

[0094] in, , They are the d-axis component and q-axis component of the grid-side current (i.e., the converter output current) respectively; , are the d-axis component and q-axis component of the converter output voltage respectively; , They are the d-axis component and q-axis component of the grid voltage respectively; , They are filter inductance (i.e. converter inductance) and resistance, is the grid frequency.

[0095] Active power absorbed by the target converter from the grid and reactive power , specifically expressed as:

[0096] (2).

[0097] Let the state variable (i.e. grid-side current) be expressed as , the control variable (i.e., the converter output voltage) is expressed as , the disturbance (i.e., grid voltage) is expressed as , the output (including grid-side current, and active power and reactive power absorbed by the converter from the grid) is expressed as , then the state space equation of the target converter is established as:

[0098] (3)

[0099] Among them, the system matrix , the input matrix , the perturbation matrix , It contains the nonlinear relationship shown in formula (2).

[0100] Furthermore, considering that the grid voltage does not change in the short control time domain, the following can be obtained by discretizing equation (3):

[0101] (4)

[0102] in, , , , , k For any moment, T s is the discrete sampling time. For formula (4), it is continuously recursively forward in the time domain to obtain the prediction model of the target converter state. In actual engineering, the model parameters of the wind power grid side converter and the energy storage system converter are different, but the prediction model form remains consistent without loss of generality. In this embodiment, it is uniformly represented by formula (4).

[0103] In the embodiment of the present invention, the output current expression of the converter in the dq coordinate system is obtained by coordinate transformation, the state space equation of the target converter is constructed in combination with the acquired active power and reactive power, and the state space equation is discretized to obtain a distributed prediction model, and the nonlinear mechanism model is converted into a linear model, which helps to improve the performance of distributed model predictive control and ensure the smooth operation of the wind turbine.

[0104] Step S203: determining the objective function of the target converter based on the control objective.

[0105] Specifically, the above step 203 includes:

[0106] Step B1, obtaining a tracking error of a converter output current in a prediction time domain, wherein the tracking error is determined by an actual value and a reference value of the converter output current of a target converter.

[0107] In this embodiment, the process of determining the reference value of the converter output current includes: the target converter implements the virtual synchronous generator control strategy, and the corresponding reference value of the converter output current in the dq coordinate system is obtained, wherein the virtual synchronous generator control includes the frequency modulation control of active power and frequency, and the excitation regulation control of reactive power and voltage; wherein the frequency modulation control is used to obtain the optimal phase angle, and coordinate conversion is performed based on the optimal phase angle; the excitation regulation control is used to obtain the electromotive force, and the electromotive force is obtained through the proportional integral controller to obtain the reference value of the converter output current in the dq coordinate system. Specifically, the virtual synchronous generator control strategy is adopted to enable the target converter to have the inertial damping characteristics of the virtual synchronous generator to realize the grid-type control of the wind storage converter, which can greatly ensure the accuracy of the reference value of the converter output current, help to improve the accuracy of the subsequent distributed model prediction control, and thus meet the flexible, efficient and stable control requirements of the wind turbine.

[0108] Step B2, obtaining a control increment of the converter output voltage in the control time domain, wherein the control increment is determined based on a to-be-solved value of the converter output voltage of the target converter.

[0109] Step B3, performing weighted summation of the tracking error and the control increment to obtain the objective function of the target converter.

[0110] In a specific embodiment, the control target of the target converter is to track the actual three-phase current (i.e., the grid-side current) to the current reference value, while considering the limitation on the size of the control increment (i.e., the converter output voltage), and adopting the distributed model predictive control algorithm, the predictive controller objective function of each converter is l for:

[0111] (5)

[0112] Among them, the subscript Indicates converter (i.e., target converter); weight coefficient and They are used to set the weights of the current tracking error and the voltage control increment in the cost function and balance the dimensions of the two. Their values ​​are adaptively set based on actual needs. N p For the prediction time domain, N c To control the time domain, j The value range is natural numbers; , , and They are the d-axis reference value, d-axis actual value, q-axis reference value and q-axis actual value of the grid-side current respectively; , are the q-axis reference value and q-axis actual value of the converter output voltage respectively; k + j | k express k Always k + j Prediction of the moment.

[0113] In a specific embodiment, a virtual synchronous generator control strategy is adopted to enable the converter to have the inertial damping characteristics of a synchronous generator, so as to realize the grid-type control of the wind storage converter. The control structure is as follows: Figure 3 and 4 It should be noted that Figure 3 P and are the actual value and reference value of active power, ω and ω ref are the actual value and reference value of the grid frequency, D is the damping coefficient, J is the virtual moment of inertia, s is the differential operator, θ is the phase angle; Figure 4 Q and are the actual value and reference value of reactive power, V and V ref are the actual value and reference value of the converter output voltage, respectively, and the coefficient K u Indicates the reactive power adjustment value and coefficient corresponding to unit voltage change K is the integral coefficient, s is the differential operator, E abc is the three-phase control voltage (i.e. electromotive force) in the abc coordinate system. Specifically, Figure 3 The active power-frequency control shown simulates the rotor motion and primary frequency modulation process of the synchronous generator, and is used to characterize the active power-frequency droop characteristic, and changes the virtual mechanical power output according to the detected active power deviation, thereby achieving frequency regulation. Figure 4 The reactive power-voltage control shown simulates the excitation regulation process of the synchronous generator, which is used to characterize the reactive power-voltage droop characteristics and adjust the output voltage according to the reactive power deviation and voltage deviation detected. Specifically, according to the phase angle obtained by the active power-frequency control θ It is used to convert the three-phase AC quantity of the target converter into dq grid-side components, electromotive force E abcIt is used to obtain the reference value of the dq axis grid side current, and the reference value of the dq axis grid side current is obtained by using the proportional integral controller PI as follows:

[0114] (6)

[0115] (7).

[0116] In the embodiment of the present invention, by tracking the output current of the converter in the prediction time domain and taking into account the limitation of the control increment size of the converter output voltage in the control time domain, the objective function of the target converter is obtained by weighted summation of the tracking error and the control increment, thereby ensuring the accuracy of the objective function.

[0117] Step S204, determining constraint conditions based on the power coordination control requirements among the wind power storage converters, the preset grid dispatching power and the preset converter power threshold.

[0118] Specifically, the above step S204 includes:

[0119] Step S2041, determining the coupling constraint of the target converter based on the power coordination control requirements between the wind power storage converters and the preset grid dispatching power.

[0120] In a specific embodiment, in order to achieve power system frequency support, the output power of each wind turbine and the energy storage system converter needs to be coordinated with each other, so as to ensure that the total power output of all network control units can match the active power demand of the power system. This embodiment uses the control requirement of the total output power being a fixed value as the coupling constraint of the distributed wind storage converter control. Converters The coupling constraints that the target converter needs to satisfy are expressed as:

[0121] (8)

[0122] in, M is the number of wind turbines and energy storage systems connected in the wind farm (i.e. the total number of wind-storage converters in the target wind power generation system), P WF,ref It is the active power instruction issued by the power grid dispatcher.

[0123] Step S2042: Optimize the coupling constraints based on the preset slack variables to obtain soft constraints.

[0124] In a specific embodiment, in the process of solving the predictive control optimization problem, the equality constraint shown in formula (8) is used, which easily causes the recursion to be infeasible. Therefore, when establishing the optimization problem, a slack variable needs to be introduced. The strict equality constraints are relaxed into inequality constraints, i.e. soft constraints, which are expressed as:

[0125] (9)

[0126] in, represents the set of positive real numbers.

[0127] It should be noted that after the hard constraint is modified to a soft constraint, When designing a controller for a converter, these slack variables need to be additionally defined as penalty terms in the objective function to control the variables not to violate the constraints, that is, the penalty term It is expressed as:

[0128] (10)

[0129] in, is the weight coefficient of the penalty item, and its specific value can be set based on actual demand adaptability.

[0130] Step S2043: determining a power constraint based on a preset converter power threshold.

[0131] In a specific embodiment, due to the limitation of wind speed or energy storage unit capacity, there is an upper limit on the power that can be output by the grid-side converter. The converter also needs to meet the power constraint, which is expressed as:

[0132] (11)

[0133] in, Indicates The maximum power that a converter can output.

[0134] Step S2044: determine the soft constraint and the power constraint as constraint conditions.

[0135] The embodiments of the present invention take into account the coordination between the output powers of each wind turbine and each wind-storage inverter in the energy storage system and the corresponding determined constraints for the limitation of the output power of the wind-storage inverter, which can ensure the feasibility of the control optimization problem and improve the efficiency of distributed model predictive control to a certain extent.

[0136] Step S205: At each control moment, an optimization solution is performed based on the distributed prediction model, objective function, constraint conditions and other overall optimization control objectives of the wind-storage converter to obtain the optimal control solution at the current control moment.

[0137] Specifically, the above step S205 includes:

[0138] Step S2051, penalize the preset slack variable to obtain the corresponding penalty item, and obtain the penalty value of the penalty item in the prediction time domain.

[0139] Step S2052, obtaining the objective functions corresponding to all other wind-storage converters except the target converter in the target wind power generation system, and performing weighted summation on them to obtain the overall optimization control target of other wind-storage converters.

[0140] It should be noted that the determination of the objective functions corresponding to all other wind-storage converters in this embodiment can be obtained by referring to the corresponding formula (5) above.

[0141] Step S2053, based on the objective function, the penalty value and the overall optimization control objectives of other wind-storage converters, obtain the target cost function.

[0142] In this embodiment, the target cost function J Establish the The distributed model predictive control optimization problem of a converter can be specifically expressed as:

[0143]

[0144] st

[0145]

[0146] ,

[0147]

[0148] It should be noted that in the above formula st A common abbreviation in mathematics, used in mathematical proofs and formulas to indicate that certain variables or expressions need to satisfy specific conditions or constraints.

[0149] in, m For the The other converters connected to the converter, M is the total number of converter units contained in the distributed control system. From the above formula, we can see that In addition to its own objective function, the optimization problem of a converter also includes the weighted sum of the objective functions of other converters, that is, the total optimization control objective of other wind-storage converters. ( is the weight coefficient of the objective function of other converters. It should be noted that the sum of the weight coefficients of all objective functions of other converters needs to be a fixed value, that is, ), that is, when solving the optimal control quantity, the control information of other converters is still associated, so the optimal control solution of the remaining converters needs to be determined in advance.

[0150] Step S2054, at each control moment, iterative solution is performed based on the target cost function, the distributed prediction model and the constraint conditions until the preset iteration requirements are met to obtain the optimal control solution at the current control moment.

[0151] Specifically, the above step S2054 includes:

[0152] Step C1, at each control moment, the control variables of the target converter are iterated using the Nash optimal algorithm, and the target cost function, the distributed prediction model and the constraint conditions are combined to obtain the optimized value of the current iteration.

[0153] Step C2, calculating the current difference between the optimization value of the current iteration and the optimization value of the previous iteration.

[0154] Step C3, counting the number of iterations of the current iteration.

[0155] Step C4, determining whether the current difference is greater than a preset accuracy threshold and whether the number of iterations is less than a preset number threshold.

[0156] In this embodiment, specific values ​​of the preset accuracy threshold and the preset number threshold can be adaptively adjusted based on actual needs.

[0157] Step C5, when the current difference is not greater than the preset accuracy threshold, or when the number of iterations is not less than the preset number threshold, the optimized value of the current iteration is determined as the optimal control solution at the current control moment.

[0158] In a specific embodiment, since all converter controllers perform optimization calculations in parallel at the same time, this embodiment introduces the Nash optimal algorithm. , assuming that the Nash optimal solutions of other subsystems are known, the controller only uses the control quantity of this system u i (i.e., transformer output voltage) optimizes its objective function to obtain the Nash optimal solution at the current moment. At each control moment, the above solution process is iterated multiple times so that all converter controllers can reach the Nash optimal solution through competition, and the entire wind power generation system is in a Nash equilibrium state. The specific solution process is as follows: Figure 5 As shown. It should be noted that, for each moment, all converters need to be iteratively solved for Nash optimality, that is, optimization is performed at each moment to achieve the effect of feedback correction, aiming to improve control accuracy. Specifically, the target converter is iteratively solved by the Nash optimal algorithm, while taking into account the optimal solution of other converters in the target wind power generation system, which helps to improve the accuracy and efficiency of distributed model predictive control.

[0159] The embodiment of the present invention takes into account the penalty for preset slack variables in the optimization conditions, as well as the influencing factors of other wind storage converters in the target wind power generation system on the target converter, constructs a target cost function and solves the optimal control target, which can ensure the effectiveness and rationality of the control target, has the advantages of flexible controllability, fast response, high efficiency and excellent performance, and ensures the stable operation of the wind turbine set.

[0160] Step S206: Control the target converter accordingly based on the optimal control solution. Figure 1 Step S106 of the illustrated embodiment will not be described in detail here.

[0161] The distributed model predictive control method for wind-storage inverters of the embodiment of the present invention, as a flexible, controllable and fast-responding wind-storage integrated grid-type inverter control technology, is an effective means to solve the problems of large fluctuations in wind turbine output power, insufficient operating stability, and reduced equivalent inertia. It can be used for the scenario of distributed grid-connected multiple wind-storage integrated grid-type inverters, and a distributed model predictive control algorithm can be used to solve the problem of coordinated control of multiple units of wind-storage integrated grid-type inverters. Each unit and energy storage unit can coordinate active power output according to the system frequency regulation requirements and under the constraints of its own available power, thereby improving the reliability and economy of the grid-type control of the multi-machine system.

[0162] In summary, the embodiments of the present invention can adaptively adjust the output power of the corresponding converter online according to the changes in grid frequency and the capacity limitations of each wind turbine and energy storage unit, so that the output powers of the converters can be coordinated with each other, and the total output of all network control units can be kept to match the active power demand of the power system, so as to maintain the active power balance of the system; the control accuracy is high, the control algorithm structure is simple, the scalability is strong, and the fault tolerance rate is high; and the control solution is distributed in the controllers of each converter, without the need for a central control node, which can greatly reduce the burden of online control.

[0163] In this embodiment, a distributed model predictive control device for a wind storage inverter is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and will not be repeated hereafter. As the term "module" is used below, it can implement a combination of software and / or hardware of a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0164] The present invention provides a distributed model predictive control device for a wind power storage converter, such as Figure 6 As shown, the device comprises:

[0165] The acquisition module 601 is used to acquire the status data of the target converter, where the target converter is any wind-storage converter in the target wind power generation system.

[0166] The construction module 602 is used to construct a distributed prediction model of the target converter based on the state data, and to design a control target of the target converter based on the distributed prediction model.

[0167] The first determination module 603 is configured to determine an objective function of a target converter based on a control objective.

[0168] The second determination module 604 is used to determine the constraint conditions based on the power coordination control requirements among the wind power storage converters, the preset grid dispatching power and the preset converter power threshold.

[0169] The solution module 605 is used to optimize and solve at each control moment based on the distributed prediction model, objective function, constraints and other overall optimization control objectives of the wind-storage converter to obtain the optimal control solution at the current control moment.

[0170] The control module 606 is used to control the target converter accordingly based on the optimal control solution.

[0171] In some optional embodiments, the construction module 602 includes: a first construction submodule, a second construction submodule, a third construction submodule and a fourth construction submodule; wherein the first construction submodule is used to transform the state data to obtain the converter output current expression in the dq coordinate system, wherein the converter output current expression includes the correspondence between the d-axis component and the q-axis component of the converter output current and the converter output voltage, the grid voltage, the grid frequency, the preset converter inductance and the preset resistance respectively; the second construction submodule is used to calculate the power data of the target converter, wherein the power data includes the active power and reactive power; a third construction submodule, used to make the converter output current as the state variable, the converter output voltage as the control variable, the grid voltage as the disturbance, the d-axis component, the q-axis component, the active power and the reactive power of the converter output current as the output, and to establish the state space equation of the target converter based on the state variable, the control variable, the disturbance and the output; a fourth construction submodule, used to discretize the state space equation and recursively in the time domain to obtain a distributed prediction model, wherein the control objective of the target converter is to track the state variables in the distributed prediction model and to limit the control variables.

[0172] In some optional embodiments, the first determination module 603 includes: a first determination submodule, a second determination submodule and a third determination submodule; wherein the first determination submodule is used to obtain the tracking error of the converter output current in the prediction time domain, wherein the tracking error is determined by the actual value and reference value of the converter output current of the target converter; the second determination submodule is used to obtain the control increment of the converter output voltage in the control time domain, wherein the control increment is determined based on the to-be-solved value of the converter output voltage of the target converter; the third determination submodule is used to perform weighted summation of the tracking error and the control increment to obtain the objective function of the target converter.

[0173] In some optional embodiments, the first determination submodule also includes: a judgment unit, which is used for the target converter to implement the virtual synchronous generator control strategy, and obtains the reference value of the converter output current in the dq coordinate system, wherein the virtual synchronous generator control includes frequency modulation control of active power and frequency, and excitation regulation control of reactive power and voltage; wherein the frequency modulation control is used to obtain the optimal phase angle, and perform coordinate transformation based on the optimal phase angle; the excitation regulation control is used to obtain the electromotive force, and obtain the reference value of the converter output current in the dq coordinate system through a proportional-integral controller.

[0174] In some optional embodiments, the second determination module 604 includes: a first determination submodule, a second determination submodule, a third determination submodule and a fourth determination submodule; wherein the first determination submodule is used to determine the coupling constraints of the target converter based on the power coordination control requirements between each wind storage converter and the preset grid dispatching power; the second determination submodule is used to optimize the coupling constraints based on preset slack variables to obtain soft constraints; the third determination submodule is used to determine the power constraints based on the preset converter power threshold; the fourth determination submodule is used to determine the soft constraints and the power constraints as constraint conditions.

[0175] In some optional embodiments, the solution module 605 includes: a first solution submodule, a second solution submodule, a third solution submodule and a fourth solution submodule; wherein the first solution submodule is used to penalize the preset slack variables, obtain the corresponding penalty items, and obtain the penalty values ​​of the penalty items in the prediction time domain; the second solution submodule is used to obtain the objective functions corresponding to all wind-storage converters except the target converter in the target wind power generation system, and perform weighted summation to obtain the overall optimization control target of other wind-storage converters; the third solution submodule is used to obtain the target cost function based on the objective function, the penalty value and the overall optimization control target of other wind-storage converters; the fourth solution submodule is used to perform iterative solution based on the target cost function, the distributed prediction model and the constraints at each control moment until the preset iteration requirements are met to obtain the optimal control solution at the current control moment.

[0176] In some optional embodiments, the fourth solution submodule includes: a first solution unit, a second solution unit, a third solution unit, a fourth solution unit and a fifth solution unit; wherein the first solution unit is used to iterate the control variables of the target converter at each control moment using the Nash optimal algorithm, and obtain the optimized value of the current iteration in combination with the target cost function, the distributed prediction model and the constraints; the second solution unit is used to calculate the current difference between the optimized value of the current iteration and the optimized value of the previous iteration; the third solution unit is used to count the number of iterations of the current iteration; the fourth solution unit is used to determine whether the current difference is greater than the preset accuracy threshold and whether the number of iterations is less than the preset number threshold; the fifth solution unit is used to determine the optimized value of the current iteration as the optimal control solution at the current control moment when the current difference is not greater than the preset accuracy threshold, or when the number of iterations is not less than the preset number threshold.

[0177] The further functional description of each of the above modules is the same as that of the above corresponding embodiments and will not be repeated here.

[0178] The distributed model predictive control device for the wind storage inverter in this embodiment is presented in the form of a functional unit, where the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and memory that executes one or more software or fixed programs, and / or other devices that can provide the above functions.

[0179] The distributed model predictive control device for wind storage inverters of the embodiment of the present invention can realize distributed model predictive control of any wind storage inverter in the target wind power generation system, has the advantages of flexible control, fast response, high efficiency and excellent performance, and meets the flexible, efficient and stable control requirements of wind turbines.

[0180] The embodiment of the present invention also provides a computer device, see Figure 7 , Figure 7 is a schematic diagram of the structure of the above-mentioned computer device provided by an optional embodiment of the present invention, such as Figure 7As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting various components, including high-speed interfaces and low-speed interfaces. Various components are connected to each other using different buses for communication, and can be installed on a common mainboard or installed in other ways as needed. The processor can process the instructions executed in the computer device, including instructions stored in or on the memory to display the graphical information of the GUI on an external input / output device (such as, a display device coupled to the interface). In some optional embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (for example, as a server array, a group of blade servers, or a multi-processor system). Figure 7 A processor 10 is taken as an example.

[0181] The processor 10 may be a central processing unit, a network processor or a combination thereof. The processor 10 may further include a hardware chip. The hardware chip may be a dedicated integrated circuit, a programmable logic device or a combination thereof. The programmable logic device may be a complex programmable logic device, a field programmable gate array, a general purpose array logic or any combination thereof.

[0182] The memory 20 stores instructions executable by at least one processor 10, so that at least one processor 10 executes the method shown in the above embodiment.

[0183] The memory 20 may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 20 may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some optional embodiments, the memory 20 may optionally include a memory remotely arranged relative to the processor 10, and these remote memories may be connected to the computer device via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0184] The memory 20 may include a volatile memory, such as a random access memory; the memory may also include a non-volatile memory, such as a flash memory, a hard disk or a solid state drive; the memory 20 may also include a combination of the above types of memory.

[0185] The computer device also includes a communication interface 30, which is used for the main control chip to communicate with other devices or a communication network.

[0186] The embodiment of the present invention also provides a computer-readable storage medium. The method according to the embodiment of the present invention can be implemented in hardware, firmware, or can be implemented as a computer code that can be recorded in a storage medium, or can be implemented as a computer code that is originally stored in a remote storage medium or a non-temporary machine-readable storage medium and will be stored in a local storage medium through a network download, so that the method described herein can be stored in such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only storage memory, a random access memory, a flash memory, a hard disk or a solid-state hard disk, etc.; further, the storage medium can also include a combination of the above types of memories. It can be understood that a computer, a processor, a microprocessor main control chip or programmable hardware includes a storage component that can store or receive software or computer code. When the software or computer code is accessed and executed by a computer, a processor or hardware, the method shown in the above embodiment is implemented.

[0187] Although the embodiments of the present invention have been described in conjunction with the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A distributed model predictive control method for a wind power storage converter, characterized in that: The method comprises: Acquiring state data of a target converter, wherein the target converter is any wind storage converter in a target wind power generation system; Building a distributed prediction model of the target converter based on the state data, and designing a control target of the target converter based on the distributed prediction model; Determining the objective function of the target converter based on the control objective includes: Obtaining a tracking error of the converter output current in a prediction time domain, wherein the tracking error is determined by an actual value and a reference value of the converter output current of the target converter; wherein the process of determining the reference value of the converter output current includes: the target converter implements a virtual synchronous generator control strategy, and correspondingly obtains a reference value of the converter output current in a dq coordinate system, wherein the virtual synchronous generator control includes frequency modulation control of active power and frequency, and excitation regulation control of reactive power and voltage; wherein the frequency modulation control is used to obtain an optimal phase angle, and coordinate transformation is performed based on the optimal phase angle; the excitation regulation control is used to obtain an electromotive force, and obtains a reference value of the converter output current in a dq coordinate system by using the electromotive force through a proportional-integral controller; Acquire a control increment of the converter output voltage in a control time domain, wherein the control increment is determined based on a to-be-solved value of the converter output voltage of the target converter; The tracking error and the control increment are weightedly summed to obtain the objective function of the target converter. The objective function of the target converter is for: Among them, the subscript Indicates A converter, namely, a target converter; and are to set the weight coefficients of current tracking error and voltage control increment in the cost function respectively; Np For the prediction time domain, Nc To control the time domain, j The value range is natural numbers; The grid-side current is The d-axis reference value, d-axis actual value, q-axis reference value and q-axis actual value; The converter output voltage is The actual value of the d-axis and the actual value of the q-axis; k + j | k express k Always k + j Prediction of the moment; The constraints are determined based on the power coordination control requirements between the wind power storage converters, the preset grid dispatching power and the preset converter power threshold, including: The coupling constraint of the target converter is determined based on the power coordination control requirements between the wind-storage converters and the preset grid dispatching power. The coupling constraint is expressed as: in, M is the total number of converters, Active power instructions issued by the power grid dispatcher. is the output power of the target converter; The coupling constraints are optimized based on preset slack variables to obtain soft constraints, which are expressed as: in, is the preset slack variable, represents the set of positive real numbers; A power constraint is determined based on a preset converter power threshold, and the power constraint is expressed as: in, is a preset converter power threshold; determining the soft constraint and the power constraint as constraint conditions; At each control moment, an optimization solution is performed based on the distributed prediction model, the objective function, the constraint conditions and the overall optimization control objective of other wind-storage converters to obtain the optimal control solution at the current control moment, including: Penalize the preset slack variable to obtain the corresponding penalty term, and obtain the penalty value of the penalty term in the prediction time domain. It is expressed as: in, is the weight coefficient of the penalty term; Obtain the objective functions corresponding to all other wind storage converters except the target converter in the target wind power generation system, and perform weighted summation to obtain the total optimization control target of other wind storage converters. The total optimization control target of other wind storage converters is expressed as: in, m are other converters connected to the target converter, is the weight coefficient of the objective function of other converters. The sum of the weight coefficients of all objective functions of other converters satisfies ; Based on the objective function, the penalty value and the overall optimization control target of the other wind-storage converters, a target cost function is obtained. The target cost function J It is expressed as: ； At each control moment, iteratively solving is performed based on the target cost function, the distributed prediction model and the constraint conditions until the preset iteration requirements are met to obtain the optimal control solution at the current control moment; The target converter is controlled accordingly based on the optimal control solution.

2. The distributed model predictive control method for wind power storage converter according to claim 1 is characterized in that: The state data includes a converter output current, a converter output voltage, a grid voltage, and a grid frequency; constructing a distributed prediction model of the target converter based on the state data, and designing a control target of the target converter based on the distributed prediction model, including: Performing coordinate transformation on the state data to obtain an expression of the converter output current in a dq coordinate system, wherein the expression of the converter output current includes corresponding relationships between the d-axis component and the q-axis component of the converter output current and the converter output voltage, the grid voltage, the grid frequency, the preset converter inductance, and the preset resistance, respectively; Calculating power data of the target converter, wherein the power data includes active power and reactive power; Let the converter output current be a state variable, the converter output voltage be a control variable, the grid voltage be a disturbance, the d-axis component and q-axis component of the converter output current, the active power and the reactive power be output quantities, and establish a state space equation of the target converter based on the state variable, the control variable, the disturbance and the output quantity; The state space equation is discretized and recursively processed along the time domain to obtain a distributed prediction model, wherein the control objective of the target converter is to track the state variables in the distributed prediction model and to limit the control variables.

3. The distributed model predictive control method for wind power storage converter according to claim 1 is characterized in that: At each control moment, iteratively solving based on the target cost function, the distributed prediction model and the constraint conditions until the preset iteration requirements are met to obtain the optimal control solution at the current control moment, including: At each control moment, the control variable of the target converter is iterated by using the Nash optimal algorithm, and the optimized value of the current iteration is obtained by combining the target cost function, the distributed prediction model and the constraint condition; Calculate the current difference between the optimization value of the current iteration and the optimization value of the previous iteration; Count the number of iterations of the current iteration; Determine whether the current difference is greater than a preset accuracy threshold and whether the number of iterations is less than a preset number threshold; When the current difference is not greater than the preset accuracy threshold, or when the number of iterations is not less than the preset number threshold, the optimized value of the current iteration is determined as the optimal control solution at the current control moment.

4. A distributed model predictive control device for a wind power storage converter, characterized in that: The device comprises: An acquisition module, used for acquiring status data of a target converter, wherein the target converter is any wind storage converter in a target wind power generation system; A construction module, configured to construct a distributed prediction model of the target converter based on the state data, and to design a control target of the target converter based on the distributed prediction model; A first determination module, configured to determine an objective function of the target converter based on the control target, comprises: Obtaining a tracking error of the converter output current in a prediction time domain, wherein the tracking error is determined by an actual value and a reference value of the converter output current of the target converter; wherein the process of determining the reference value of the converter output current includes: the target converter implements a virtual synchronous generator control strategy, and correspondingly obtains a reference value of the converter output current in a dq coordinate system, wherein the virtual synchronous generator control includes frequency modulation control of active power and frequency, and excitation regulation control of reactive power and voltage; wherein the frequency modulation control is used to obtain an optimal phase angle, and coordinate transformation is performed based on the optimal phase angle; the excitation regulation control is used to obtain an electromotive force, and obtains a reference value of the converter output current in a dq coordinate system by using the electromotive force through a proportional-integral controller; Acquire a control increment of the converter output voltage in a control time domain, wherein the control increment is determined based on a to-be-solved value of the converter output voltage of the target converter; The tracking error and the control increment are weightedly summed to obtain the objective function of the target converter. The objective function of the target converter is for: Among them, the subscript Indicates A converter, namely, a target converter; and are to set the weight coefficients of current tracking error and voltage control increment in the cost function respectively; Np For the prediction time domain, Nc To control the time domain, j The value range is natural numbers; The grid-side current is The d-axis reference value, d-axis actual value, q-axis reference value and q-axis actual value; The converter output voltage is The actual value of the d-axis and the actual value of the q-axis; k + j | k express k Always k + j Prediction of the moment; The second determination module is used to determine the constraint conditions based on the power coordination control requirements between the wind power storage converters, the preset grid dispatching power and the preset converter power threshold, including: The coupling constraint of the target converter is determined based on the power coordination control requirements between the wind-storage converters and the preset grid dispatching power. The coupling constraint is expressed as: in, M is the total number of converters, Active power instructions issued by the power grid dispatcher. is the output power of the target converter; The coupling constraints are optimized based on preset slack variables to obtain soft constraints, which are expressed as: in, is the preset slack variable, represents the set of positive real numbers; A power constraint is determined based on a preset converter power threshold, and the power constraint is expressed as: in, is a preset converter power threshold; determining the soft constraint and the power constraint as constraint conditions; A solution module is used to optimize and solve at each control moment based on the distributed prediction model, the objective function, the constraint conditions and the overall optimization control objective of other wind-storage converters to obtain the optimal control solution at the current control moment, including: Penalize the preset slack variable to obtain the corresponding penalty term, and obtain the penalty value of the penalty term in the prediction time domain. It is expressed as: in, is the weight coefficient of the penalty term; Obtain the objective functions corresponding to all other wind storage converters except the target converter in the target wind power generation system, and perform weighted summation to obtain the total optimization control target of other wind storage converters. The total optimization control target of other wind storage converters is expressed as: in, m are other converters connected to the target converter, is the weight coefficient of the objective function of other converters. The sum of the weight coefficients of all objective functions of other converters satisfies ; Based on the objective function, the penalty value and the overall optimization control target of the other wind-storage converters, a target cost function is obtained. The target cost function J It is expressed as: ； At each control moment, iteratively solving is performed based on the target cost function, the distributed prediction model and the constraint conditions until the preset iteration requirements are met to obtain the optimal control solution at the current control moment; A control module is used to control the target converter accordingly based on the optimal control solution.

5. A computer device, characterized in that: The device includes: a memory and a processor, the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the distributed model predictive control method for wind energy storage inverters according to any one of claims 1 to 3 by executing the computer instructions.

6. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a computer to execute the distributed model predictive control method for a wind energy storage converter according to any one of claims 1 to 3.

Images