A method and system for active power distribution in a wind farm

Abstract

The present application relates to the field of wind power generation technology, and more particularly to a wind farm active power distribution method and system. The present application is based on the error between the actual value and the reference value of the wind turbine active power, the current measurement value of the wind turbine at the current time, and the current saturation value of the PI controller to construct a current reference value calculation model of the converter current loop; the current reference value calculation model of the converter current loop is converted into a linear expression in incremental form to obtain the constraint condition of the wind turbine active power reference value increment; the objective function and the constraint condition of the wind turbine active power distribution strategy are constructed to obtain the target active power distribution strategy of each prediction step. The present application improves the wind turbine active power tracking performance and enhances the stability of the power system.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of wind power generation, in particular to a wind farm active power distribution method and system. BACKGROUND

[0002] As an important part of clean energy, wind power generation has become a key force for optimizing the energy landscape due to its clean and renewable advantages. Its large-scale promotion has significantly reduced the proportion of traditional power generation technology, reduced dependence on fossil energy, and improved air quality. However, wind turbines themselves have problems such as slow power response and difficult dispatch power distribution, which limit the development of new energy generation technology.

[0003] Active power distribution of wind farms is a core link to achieve stable power supply. The power grid dispatch will issue total active power instructions to the wind farm according to the system power balance demand, and by reasonably distributing the total active power instructions to each wind turbine, efficient, stable and economic operation can be achieved. The commonly used distribution methods include equal power distribution method, capacity proportional distribution method and distributed coordinated control method based on multi-objective optimization. The equal power distribution method and the capacity proportional distribution method only make static distribution based on the rated parameters of the unit, without considering the individual differences between different units, and forcibly distributing power according to a fixed proportion may cause some units to overload.

[0004] To solve the defects of static distribution, the distributed coordinated control method based on multi-objective optimization is based on the next cycle predicted wind speed to predict the wind turbine output potential, minimizes the wind farm output change value, and minimizes the maximum value of the output change value of each wind turbine as the objective function, and takes the wind farm output constraint, wind turbine output constraint, wind farm active power change rate constraint and wind turbine output change rate constraint as the constraint condition. The active power reference value of each wind turbine in the wind farm is calculated by genetic algorithm, which can dynamically respond to real-time data and improve the stability of wind farm operation and equipment economy.

[0005] In actual operation, wind turbines need to access the wind farm through a converter, and the converter adjustment process has a dynamic response process. When the upper dispatching issues an increase in power command, the converter needs to gradually increase the current to increase the active power of the wind turbine. Due to the small capacity of early wind turbines and the fast response speed of the converter, the distributed coordination control method based on multi-objective optimization defaults the converter to an "ideal actuator", simplifies the wind turbine to a "unit that can directly adjust the output power", and considers that the randomness of wind speed is the main source of wind turbine output fluctuation. It is assumed that the output of the wind turbine is determined by the wind speed and can be adjusted in real time. However, under the current background of large-scale access of high-proportion renewable energy and power electronic equipment to the power grid, the capacity of wind turbines is increasing, and the mutual coupling between power electronic devices in the power grid produces oscillation, which significantly increases the instability of the power system and easily induces low-frequency oscillation and sub- / super-synchronous oscillation. When the power grid is unstable, active power support is needed to maintain stability. Whether the wind turbine can quickly respond to the active power command and accurately allocate the power dispatching command according to the unit capacity directly determines whether its steady-state operation capability meets the requirements. If it cannot meet the requirements, the grid connection of the wind turbine will worsen the stability of the power system. The distributed coordination control method based on multi-objective optimization ignores this delay, resulting in a time difference and amplitude difference between the theoretical prediction value and the actual value, which causes obvious lag error in the overall power tracking of the wind farm.

[0006] In addition, although the distributed coordination control method based on multi-objective optimization sets the output constraint of the wind turbine, so that the active power reference value of the wind turbine does not exceed its upper limit value, as the active power reference value of the wind turbine increases, the current in the current loop of the converter will also increase. When the current saturation value of the PI controller in the converter is exceeded, the converter will trigger overcurrent protection, forcibly limit the power output, or even cut off the grid connection of the unit, resulting in the loss of the tracking ability of the wind turbine to the power reference value, and finally causing the actual output of the wind farm to deviate from the dispatching plan. SUMMARY

[0007] Therefore, the technical problem to be solved by the present application is to overcome the defects of the existing active power distribution method for wind farms, which ignores the dynamic response delay of the converter, and as the active power reference value of the wind turbine increases, the current loop of the converter is easy to break through the current saturation threshold of the PI controller, resulting in the loss of the tracking ability of the wind turbine to the active power reference value, and further affecting the stability of the power system.

[0008] To solve the above technical problems, the present application provides an active power distribution method for a wind farm, comprising the following steps:

[0009] The current reference value calculation model of the current loop of the converter is converted into a linear expression in incremental form, to obtain a constraint condition of the active power reference value increment of the wind turbine;

[0010] The current reference value calculation model of the current loop of the converter is converted into a linear expression in incremental form, to obtain a constraint condition of the active power reference value increment of the wind turbine;

[0011] The deviation between the sum of the active power of all wind turbines and the total active power instruction, and the weighted sum of the deviation between the actual value of the active power of each wind turbine and the active power reference value, is minimized as the objective function of the active power distribution strategy of the wind turbine;

[0012] Based on the wind farm output constraint, the wind turbine output constraint, and the constraint condition of the active power reference value increment of each wind turbine, the constraint condition of the active power distribution strategy of the wind turbine is constructed;

[0013] The actual value of the active power of all wind turbines at the current time is substituted into the discrete space state equation containing all wind turbines, to obtain the active power prediction value of each prediction step;

[0014] The wind farm data at the current time and the active power prediction value of each prediction step are substituted into the objective function of the active power distribution strategy of the wind turbine, to obtain the target active power distribution strategy of each prediction step.

[0015] Preferably, the wind farm data at the current time includes: the actual value of the active power of each wind turbine, the total active power instruction, and the measured current value of the current loop of the corresponding converter of each wind turbine.

[0016] Preferably, the process of obtaining the actual value of the active power of each wind turbine at each time includes:

[0017] Taking the first wind turbine affected by the natural inflow in the wind farm as the upstream reference, the wind speed of each wind turbine at the current time considering the wake effect is calculated based on the single-machine wake model;

[0018] Based on the wind speed of each wind turbine at the current time considering the wake effect, the mechanical energy output by each wind turbine at the current time is calculated through the wind energy formula;

[0019] According to the distance between each wind turbine and the upstream reference, the wake effect attenuation coefficient of each wind turbine is set; wherein the wake effect attenuation coefficient is inversely proportional to the distance between the wind turbine and the upstream reference;

[0020] Based on the wake effect attenuation coefficient of each wind turbine, the output mechanical energy of each wind turbine is corrected, and the corrected mechanical energy of each wind turbine at the current moment is calculated. The correction formula is as follows:

[0021] ,

[0022] Based on the corrected mechanical energy of each wind turbine at the current moment, calculate the actual active power of each wind turbine at the current moment using the following formula:

[0023] ,

[0024] in, for Wind turbine Corrected mechanical energy for Wind turbine mechanical energy, For wind turbines The wake effect attenuation coefficient, For complex variables, For wind turbine index, For a moment, The time required for the dynamic response of converting mechanical energy into active power. , for Wind turbine The actual value of active power.

[0025] Preferably, the current reference value calculation model for the converter current loop is as follows:

[0026] , ;

[0027] in, The measurement value of the current loop of the strain gauge current transformer of the wind turbine at the current moment. For the proportional controller parameters of the PI controller, This is a reference value for the active power of the wind turbine. This represents the active power value of the wind turbine. These are the integral controller parameters for the PI controller. For complex variables, This is the current reference value for the converter current loop. This is the current saturation value of the PI controller for the wind turbine's strain gauge.

[0028] Preferably, the current reference value calculation model of the converter current loop is converted into an incremental linear expression, as shown in the formula:

[0029] ,

[0030] Based on the constant relationship between the integrator and the linear expression, the integrator... Convert to The converted formula is:

[0031] ,

[0032] Based on the transformed formula, The constraint condition for obtaining the incremental value of the active power reference value of the wind turbine is:

[0033] ,

[0034] in, This represents the increment of the active power reference value for wind turbine units. This represents the increment of the active power value of the wind turbine. This represents the current saturation value of the PI controller for the wind turbine's strain gauge. The measurement value of the current loop of the strain gauge current transformer of the wind turbine at the current moment. For the proportional controller parameters of the PI controller, These are the integral controller parameters for the PI controller. The integral equivalent coefficient is... It is a complex variable.

[0035] Preferably, the objective function of the active power allocation strategy for wind turbine units is:

[0036] ,

[0037] ,

[0038] ,

[0039] in, Let be the objective function of the active power allocation strategy for wind turbine units. This represents the objective function for optimizing the deviation between the sum of the active power of all wind turbine units and the total active power command. The objective function represents optimizing the deviation between the actual active power value and the reference active power value of each wind turbine. Indicates minimization. for The weight, for The weight, This represents the total number of prediction steps. For the prediction step index, This represents the total number of wind turbine units. For wind turbine index, for a wind turbine at a prediction step an active power value increment, for a current time wind turbine an active power actual value, a total active power instruction, a norm square, for a wind turbine at a prediction step an active power reference value.

[0040] Preferably, the wind farm output constraint is constructed based on the sum of active power of all wind turbines being equal to the total active power instruction;

[0041] The wind turbine output constraint is constructed based on the active power reference value of each wind turbine being greater than or equal to 0 and less than or equal to the available capacity of the wind turbine.

[0042] Preferably, the obtaining process of the discrete space state equation of all wind turbines comprises:

[0043] A first-order delay function is used to describe the dynamic characteristics of the active power control loop of the wind turbine;

[0044] The dynamic characteristics of the active power control loop of the wind turbine are converted into a linear expression in incremental form to obtain the state space equation of the active power loop of the wind turbine;

[0045] Based on the state space equation of the active power loop of the wind turbine, a continuous state space equation containing all wind turbines is constructed by diagonal matrix expansion;

[0046] The continuous state space equation containing all wind turbines is discretized to obtain a discrete space state equation containing all wind turbines.

[0047] Preferably, the method for solving the target active power distribution strategy at each prediction step is any one of linear programming, quadratic programming, and genetic algorithm.

[0048] The application further provides a wind farm active power distribution system, comprising:

[0049] A model construction module is configured to construct a current reference value calculation model of a current loop of a corresponding converter of the wind turbine based on an error between an active power actual value and a reference value of the wind turbine, a current loop measured current value of the corresponding converter of the wind turbine at a current time, and a current saturation value of a PI controller of the corresponding converter of the wind turbine;

[0050] A constraint condition conversion module is configured to convert the current reference value calculation model of the current loop of the converter into a linear expression in incremental form to obtain a constraint condition of an active power reference value increment of the wind turbine.

[0051] a target function construction module, configured to construct a target function of the active power distribution strategy of the wind turbine generator set as a weighted sum of a deviation between a total active power of all wind turbine generator sets and a total active power instruction and a deviation between an actual value of the active power of each wind turbine generator set and a reference value of the active power of each wind turbine generator set being minimum;

[0052] a constraint condition construction module, configured to construct a constraint condition of the active power distribution strategy of the wind turbine generator set based on a wind farm output constraint, a wind turbine generator set output constraint and a constraint condition of an increment of the reference value of the active power of each wind turbine generator set;

[0053] an active power acquisition module, configured to substitute actual values of the active power of all wind turbine generator sets at a current time into a discrete space state equation containing all wind turbine generator sets to obtain predicted values of the active power at each prediction step;

[0054] a solution module, configured to substitute wind farm data at the current time and the predicted values of the active power at each prediction step into the target function of the active power distribution strategy of the wind turbine generator set to obtain a target active power distribution strategy at each prediction step.

[0055] The above technical solution of the present application has the following beneficial effects compared with the prior art:

[0056] The wind farm active power distribution method and system provided by the present application convert a current reference value calculation model of a current loop of a converter into a linear expression in an incremental form to obtain a constraint condition of an increment of a reference value of the active power of the wind turbine generator set, directly combine equipment operation restrictions with a power regulation process, ensure that an adjustment amount of the reference value of the active power of the wind turbine generator set is always within a safe current range of the converter, avoid response delay or equipment damage caused by current overrun, and significantly improve the rapidity and safety of power regulation.

[0057] The present application describes the dynamic characteristics of the active power control loop of the wind turbine generator set by using a first-order delay function, fully considers the dynamic characteristics of the regulation process of the converter, accurately depicts the delay and variation law of the power control in the actual operation of the wind turbine generator set, realizes the unified description of the overall operation state of all wind turbine generator sets by constructing a continuous state space equation through diagonal matrix expansion, obtains a discrete space state equation containing all wind turbine generator sets through discretization processing, predicts the active power values at each prediction step by using the discrete space state equation containing all wind turbine generator sets, improves the prediction accuracy of the actual value of the active power, reduces the power tracking error caused by the dynamic response delay of the converter, makes the overall power output of the wind farm closer to the dispatching instruction, and effectively improves the power tracking performance of the wind farm.

[0058] In addition, the application optimizes the process of obtaining the actual value of the active power of each wind turbine, and in view of the problems of low calculation efficiency and poor real-time performance caused by a complex control topology in the traditional method, the conversion process from mechanical energy to active power is simplified by using a first-order transfer function, and by using parameters such as complex variables and delay time, the key dynamic characteristics such as inertia of the mechanical system and energy conversion delay are effectively captured, the complex operation of solving a large number of differential equations is converted into algebraic operation, and the calculation time is greatly reduced; at the same time, considering the influence of the wake effect on wind energy capture, in order to improve the accuracy of the mechanical energy output by the wind turbine, the wind speed is calculated based on a single-machine wake model, and the mechanical energy is corrected by using a decay coefficient to compensate for the wake loss, so that the calculation result of the active power is more in line with the actual operation condition, and the active power tracking performance of the wind farm is significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0059] In order to make the content of the application more easily understood, the application will be further described in detail below according to specific embodiments of the application and in conjunction with the drawings, in which:

[0060] Figure 1 is a flowchart of a wind farm active power distribution method of the application.

[0061] Figure 2 is a comparison diagram of the wind turbine power response curves of the application method and the traditional method after an initial given active power instruction.

[0062] Figure 3 is a comparison diagram of the wind turbine power response curves of the application method and the traditional method under an initial given active power instruction and a step power instruction.

[0063] Figure 4 is a comparison diagram of the active power reference values of the application and the traditional method under different working conditions, Figure 4 (a) in the comparison diagram of the active power reference values of the application and the traditional method under a given active power instruction, Figure 4 (b) in the comparison diagram of the active power reference values of the application and the traditional method under a given step power instruction. DETAILED DESCRIPTION

[0064] The application will be further described below in conjunction with the drawings and specific embodiments, so that those skilled in the art can better understand the application and implement it, but the embodiments are not limiting to the application.

[0065] Referring to Figure 1 , the embodiment one provides a wind farm active power distribution method, which is suitable for a wind farm composed of large-capacity wind turbines based on a conventional phase-locked loop (PLL) and connected to a grid, and includes the following steps:

[0066] Step S1: based on the error between the actual value and the reference value of the wind turbine active power, the current loop measured current value of the converter corresponding to the wind turbine at the current time, the current saturation value of the PI controller of the converter corresponding to the wind turbine, using the PI controller characteristics, a current reference value calculation model of the converter current loop is constructed;

[0067] In this embodiment, preferably, the current reference value calculation model of the converter current loop is:

[0068]

[0069] wherein, is the current loop measured current value of the converter corresponding to the wind turbine at the current time, is the proportional controller parameter of the PI controller, is the wind turbine active power reference value, is the wind turbine active power value, is the integral controller parameter of the PI controller, is a complex variable, is the current reference value of the converter current loop, is the current saturation value of the PI controller of the converter corresponding to the wind turbine.

[0070] By introducing the PI controller, in combination with the error between the wind turbine active power value and the reference value, a dynamic correlation between the current loop measured current value of the converter and the wind turbine active power can be established, which lays a solid foundation for deriving the wind turbine active power reference value increment constraint condition.

[0071] Step S2: the current reference value calculation model of the converter current loop is converted into a linear expression in the form of increment, and the constraint condition of the wind turbine active power reference value increment is obtained;

[0072] The linear expression in the form of increment of the current reference value calculation model of the converter current loop is:

[0073]

[0074] In the optimization problem, the constraint condition is generally in the linear form containing the control quantity, but the integral element of the PI controller is contained in the above formula Based on the integral characteristics of the dynamic variable, for the dynamic variable , the integral from 0 to time has a fixed constant relationship with , The relationship is applied to the integral element in the model, and is obtained, and the formula after optimization is: ​​

[0075] ,

[0076] And according to ,get Therefore, the constraint conditions for the increment of the active power reference value of the wind turbine are derived as follows:

[0077] ,

[0078] in, This represents the increment of the active power reference value for wind turbine units. This represents the increment of the active power value of the wind turbine. This represents the current saturation value of the PI controller for the wind turbine's strain gauge. The measurement value of the current loop of the strain gauge current transformer of the wind turbine at the current moment. For the proportional controller parameters of the PI controller, These are the integral controller parameters for the PI controller. is the integral equivalent coefficient.

[0079] Considering that the active power reference value will continuously increase to accelerate the response speed of wind turbine active power, but when the active power reference value is too large, it will exceed the capacity of the wind turbine itself and cause the PI controller of the current loop to saturate, causing the wind turbine to lose its ability to track the power reference value, which to some extent restricts the grid's rapid adjustment capability. In order to optimize the performance of wind turbines and improve the overall stability of the power system, this invention constrains the current reference value calculation model of the converter current loop to not exceed the current saturation value of the PI controller. The current reference value calculation model of the converter current loop is converted into an incremental linear expression, and the constraint condition of the incremental active power reference value of the wind turbine is derived. This ensures that the active power reference value is within the saturation value of the PI controller, which not only ensures the safe operation of the wind turbine within the maximum current limit, but also avoids large fluctuations in power commands through incremental constraints, significantly improving the stability and reliability of wind farm power dispatch. The introduction of the PI controller enhances the system's anti-interference capability, while the linear incremental expression simplifies the engineering implementation of the constraint condition.

[0080] Step S3: The objective function of the wind turbine active power allocation strategy is to minimize the weighted sum of the deviations between the total active power of all wind turbines and the total active power command, and the deviations between the actual active power of each wind turbine and the active power reference value.

[0081] In this embodiment, specifically, the objective function of the active power allocation strategy for wind turbine generators is:

[0082] ,

[0083] ,

[0084] ,

[0085] wherein, is a target function of the wind turbine active power distribution strategy, represents a target function of optimizing the deviation between the sum of all wind turbine active power and the total active power instruction, represents a target function of optimizing the deviation between the actual value of each wind turbine active power and the active power reference value, represents minimizing, is a weight of is a weight of is a total prediction step number, is a prediction step index, is a total wind turbine number, is a wind turbine index, is an active power value increment of the wind turbine at the prediction step , , , is an active power actual value of the wind turbine at the current moment, is an active power prediction value of the wind turbine at the prediction step , is a total active power instruction, is a norm square, is an active power reference value of the wind turbine at the prediction step . Step S4: based on the wind farm output constraint, the wind turbine output constraint, and the constraint condition of the active power reference value increment of each wind turbine, a constraint condition of the wind turbine active power distribution strategy is constructed;

[0086] In this embodiment, specifically, the wind farm output constraint is constructed based on the sum of all wind turbine active power being equal to the total active power instruction;

[0087] The wind farm output constraint is:

[0088] ,

[0089] ,

[0090] The wind turbine output constraint is constructed based on the active power reference value of each wind turbine being greater than or equal to 0 and less than or equal to the available capacity of the wind turbine;

[0091] ​The wind turbine output constraint is:

[0092] ,

[0093] wherein, is the available capacity of the wind turbine .

[0094] Step S5: substituting the actual value of the active power of all wind turbines at the current time into the discrete space state equation containing all wind turbines to obtain the active power prediction value of each prediction step;

[0095] In this embodiment, the process of obtaining the discrete space state equation containing all wind turbines includes:

[0096] The dynamic characteristics of the wind turbine active power control loop are described by using a first-order delay function as:

[0097] ,

[0098] wherein, and are the increments of the actual value of the active power of the wind turbine and the active power reference value respectively, , is the actual value of the active power of the wind turbine at the current time, is the active power reference value of the wind turbine, is the time constant of the active power control loop of the wind turbine, is the power instruction tracking time constant.

[0099] The dynamic characteristics of the wind turbine active power control loop are converted into a linearized expression in the incremental form to obtain the state space equation of the wind turbine active power loop;

[0100] The state space equation of the wind turbine active power loop is:

[0101] ,

[0102] wherein, is the rate of change of the active power value increment of the wind turbine with time, is the time constant of the active power control loop of the wind turbine.

[0103] Based on the state space equation of the wind turbine active power loop, a continuous state space equation containing all wind turbines is constructed by extending a diagonal matrix;

[0104] Based on the state space equation of the wind turbine active power loop, the continuous state space equation of all wind turbines is obtained as:

[0105] ,

[0106] wherein, , is the total number of wind turbines, , , , is a diagonal matrix, representing state equation coefficients, , , , is the wind turbine time constant of the active power control loop, represents constructing a diagonal matrix.

[0107] Discretize the continuous state space equation containing all wind turbines to obtain a discrete space state equation containing all wind turbines.

[0108] ,

[0109] wherein, , , , is the sampling time.

[0110] Step S6: Substitute the current time wind farm data and the active power prediction value of each prediction step into the objective function of the wind turbine active power distribution strategy to obtain the target active power distribution strategy of each prediction step.

[0111] In this embodiment, specifically, the current time wind farm data includes: the actual value of the active power of each wind turbine, the total active power instruction, and the current loop measured current value of the corresponding converter of each wind turbine.

[0112] In order to solve the problem that the original active power of the wind turbine needs to be obtained through a complex control topology, involves multiple coupling links, the steps are complicated, the parameters are numerous, and the calculation efficiency is low, the real-time performance is poor, and it is difficult to meet the demand of power grid rapid scheduling, the following acquisition method is proposed:

[0113] In this embodiment, preferably, the process of obtaining the actual value of the active power of each wind turbine at each time point includes:

[0114] Taking the first wind turbine affected by the natural inflow in the wind farm as the upstream reference, based on the single-machine wake model, the wind speed of each wind turbine considering the wake effect at the current time is calculated respectively;

[0115] Considering the actual wake effect in the wind farm, the wind speed of the downstream wind turbine at the distance of the upstream reference will be attenuated to:

[0116] ,

[0117] in, for Wind turbine Taking into account the wind speed of the wake effect, for Wind turbine wind speed, For thrust coefficient, For wind turbines The radius of the wind turbine, The wake expansion coefficient is... For wind turbines Distance from the upstream benchmark;

[0118] Based on the wind speed of each wind turbine at the current moment, taking into account the wake effect, the mechanical energy output of each wind turbine at the current moment is calculated using the wind energy formula.

[0119] The wind energy formula is:

[0120] ,

[0121] ,

[0122] ,

[0123] ,

[0124] in, for Wind turbine mechanical energy, air density, for Wind turbine The power coefficient, for Wind turbine The tip speed ratio, for Wind turbine The impeller angular velocity, for Wind turbine Effective tip speed ratio, for Wind turbine The pitch angle.

[0125] Based on the distance between each wind turbine and the upstream benchmark, a wake effect attenuation coefficient is set for each wind turbine; wherein, the wake effect attenuation coefficient is inversely proportional to the distance between the wind turbine and the upstream benchmark.

[0126] According to the wake effect decay coefficient of each wind turbine, the mechanical energy output by each wind turbine is corrected, and the corrected mechanical energy of each wind turbine at the current moment is calculated, and the correction formula is:

[0127]

[0128] According to the corrected mechanical energy of each wind turbine at the current moment, the actual value of the active power of each wind turbine at the current moment is calculated, and the calculation formula is:

[0129]

[0130] wherein, is the corrected mechanical energy of the wind turbine at the current moment, is the mechanical energy of the wind turbine at the current moment, is the wake effect decay coefficient of the wind turbine, is a complex variable, is the index of the wind turbine, is the time, is the time for converting the mechanical energy into the active power, , is the actual value of the active power of the wind turbine at the current moment. The present application simplifies the complex conversion process from mechanical energy to active power by using a first-order transfer function, effectively captures the key dynamic characteristics such as inertia of mechanical system and energy conversion delay through parameters such as complex variable and delay time, and converts the complex operation of solving a large number of differential equations into algebraic operation, greatly reducing the time-consuming calculation; considering that different wind turbines are located at different positions and exist wake effect, which will lead to the actual capture of wind energy of downstream units lower than the theoretical value, in order to improve the accuracy of the mechanical energy output by the wind turbine, based on the single-unit wake model, the wind speed considering the wake effect of each wind turbine at the current moment is calculated, so as to obtain the mechanical energy output by each wind turbine at the current moment; and by setting the distance between each wind turbine and the upstream reference, the wake effect decay coefficient of each wind turbine is set, the mechanical energy output by each wind turbine is corrected through the wake effect decay coefficient of each wind turbine, the energy loss caused by the wake is effectively compensated, the active power calculation result is more close to the actual operation condition, and the active power tracking performance of the wind farm is significantly improved.

[0131] The present application simplifies the complex conversion process from mechanical energy to active power by using a first-order transfer function, effectively captures the key dynamic characteristics such as inertia of mechanical system and energy conversion delay through parameters such as complex variable and delay time, and converts the complex operation of solving a large number of differential equations into algebraic operation, greatly reducing the time-consuming calculation; considering that different wind turbines are located at different positions and exist wake effect, which will lead to the actual capture of wind energy of downstream units lower than the theoretical value, in order to improve the accuracy of the mechanical energy output by the wind turbine, based on the single-unit wake model, the wind speed considering the wake effect of each wind turbine at the current moment is calculated, so as to obtain the mechanical energy output by each wind turbine at the current moment; and by setting the distance between each wind turbine and the upstream reference, the wake effect decay coefficient of each wind turbine is set, the mechanical energy output by each wind turbine is corrected through the wake effect decay coefficient of each wind turbine, the energy loss caused by the wake is effectively compensated, the active power calculation result is more close to the actual operation condition, and the active power tracking performance of the wind farm is significantly improved.

[0132] ​​​​​​​​In this embodiment, specifically, the method for solving the target active power allocation strategy for each prediction step is any one of linear programming, quadratic programming, or genetic algorithm.

[0133] This invention transforms the current reference value calculation model of the converter current loop into an incremental linear expression, obtaining the constraint conditions for the incremental active power reference value of the wind turbine. This fundamentally ensures that the converter current remains within a safe threshold during power regulation, effectively avoiding response hysteresis and equipment failure risks caused by current over-limit, and significantly improving power regulation efficiency. Simultaneously, based on a first-order delay function, it accurately models the dynamic characteristics of the wind turbine active power control loop, constructs a continuous state-space equation using diagonal matrix expansion, and combines this with discretization to form a complete discrete state-space equation. This achieves unified quantification and accurate prediction of the operating status of all units in the wind farm, significantly reducing power tracking errors caused by converter dynamic response delays. Furthermore, addressing the efficiency bottleneck of traditional active power calculation, it simplifies the conversion link from mechanical energy to active power using a first-order transfer function, transforming complex differential operations into efficient algebraic operations. Based on a single-unit wake model and attenuation coefficient correction strategy, it accurately compensates for energy losses caused by wake effects, ensuring that the power calculation results highly match actual operating conditions. By leveraging the synergistic effects of the aforementioned multi-dimensional innovative technologies, the active power allocation strategy for wind turbine units can be solved, simultaneously optimizing the active power response speed of multiple wind turbine units. This allows for better tracking of active power commands and enhances the power tracking performance of wind turbine units.

[0134] To verify the effectiveness of the active power allocation method for wind farms of the present invention, this embodiment constructs an experimental scenario in which multiple wind turbines are connected to the power grid in parallel. A transmission line is laid between the power grid and the node of the first wind turbine, and transmission lines are also laid between the nodes of each wind turbine. Typical operating conditions are selected, and the operating condition response of a single wind turbine is monitored in detail under a given active power reference value.

[0135] like Figure 2 As shown, Figure 2 This diagram illustrates a comparison of the power response curves of wind turbines using the method of this invention and the conventional method after an initial active power command is given. The horizontal axis represents time (s), and the vertical axis represents active power (MW). The red line corresponds to the active power response curve of the method of this invention, and the blue line corresponds to the active power response curve of the conventional method. After an initial active power command of 2MW, the active power of the method of this invention rapidly increases to 2MW in the initial stage, while the response of the conventional method is significantly lagging. This verifies that, under steady-state power command scenarios, the response speed of the method of this invention is superior to that of the conventional method, enabling the wind turbine to reach the target power faster.

[0136] like Figure 3 As shown, Figure 3The wind turbine power response curve comparison diagram of the method of the present application and the traditional method under the initial given active power instruction and the step power instruction is shown in the figure. After the initial given active power instruction of 2MW, the step power instruction of 1MW is given at 10s, the method of the present application is quickly lifted at the step moment, and the target power can be quickly tracked, while the traditional method is slow in response, which shows that the present application can quickly adjust the output and has better dynamic response capability in the face of power instruction mutation, and is helpful to improve the adaptation efficiency of the unit to the working condition change.

[0137] As shown in Figure 4 , Figure 4 The active power reference value comparison diagram of the present application and the traditional method under different working conditions is shown in the figure, Figure 4 (a) in the figure is the active power reference value comparison diagram of the present application and the traditional method under the given active power instruction, Figure 4 (b) in the figure is the active power reference value comparison diagram of the present application and the traditional method under the given step power instruction.

[0138] As can be seen from Figure 4 , the active power reference value of the wind turbine using the method of the present application is temporarily greater than the given active power reference value in the above two cases, and the greater the active power reference value is, the faster the active power response speed of the wind turbine will be, so as to realize better tracking performance of the active power.

[0139] In summary, from the initial power maintenance to the step power adjustment, the method of the present application is superior to the traditional method in response speed and tracking accuracy, which effectively verifies the optimization effect of the present application on the active power distribution of the wind farm, and can help the efficient and stable operation of the unit.

[0140] The second embodiment provides a wind farm active power distribution system, which comprises:

[0141] A model construction module is configured to construct a current reference value calculation model of a current loop of a converter based on the error between the actual value and the reference value of the active power of the wind turbine, the current loop measurement current value of the corresponding converter of the wind turbine at the current moment, and the current saturation value of the PI controller of the corresponding converter of the wind turbine.

[0142] A constraint condition conversion module is configured to convert the current reference value calculation model of the current loop of the converter into a linear expression in the form of increment to obtain the constraint condition of the active power reference value increment of the wind turbine.

[0143] A target function construction module is configured to take the weighted sum of the deviation between the total active power instruction and the total active power of all wind turbines and the deviation between the actual value and the reference value of the active power of each wind turbine as the target function of the active power distribution strategy of the wind turbine.

[0144] The constraint condition construction module is configured to construct constraint conditions of the wind turbine active power distribution strategy based on wind farm output constraints, wind turbine output constraints, and constraint conditions of an active power reference value increment of each wind turbine.

[0145] The active power acquisition module is configured to substitute actual values of active power of all wind turbines at the current time into discrete space state equations of all wind turbines to obtain the active power prediction value of each prediction step.

[0146] The solution module is configured to substitute the wind farm data at the current time and the active power prediction value of each prediction step into the objective function of the wind turbine active power distribution strategy to obtain the target active power distribution strategy of each prediction step.

[0147] Those skilled in the art will understand that embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROMs, optical storage devices, etc.) having computer-usable program code embodied therein.

[0148] The present application is described with reference to flowcharts and / or block diagrams of methods, apparatus (systems) and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, as well as combinations of flows and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general purpose computer, a special purpose computer, an embedded processor, or other programmable data processing apparatus to produce a machine, so that the instructions, which are executed via the processor of the computer or other programmable data processing apparatus, generate an apparatus that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in a flow or multiple flows and / or blocks Figure 1 The functions specified in a flow or multiple flows and / or blocks

[0149] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction apparatus that implements the functions specified in the flowcharts and / or block diagrams. Figure 1 The functions specified in a flow or multiple flows and / or blocks Figure 1 The functions specified in a flow or multiple flows and / or blocks

[0150] These computer program instructions can also be loaded into a computer or other programmable data processing devices, so that a series of operational steps are performed on the computer or other programmable data processing devices to generate a computer implemented process, so that the instructions executed on the computer or other programmable data processing devices provide steps for implementing the function specified in the flowchart Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0151] Obviously, the above embodiments are only examples for clearly illustrating the present application, and are not intended to limit the embodiments. Based on the above description, other different forms of changes or variations can be made by those skilled in the art. Here, all the embodiments are not required to be exhausted, and the obvious changes or variations derived therefrom are still within the protection scope of the present application.

Claims

1. A method of active power distribution in a wind farm, characterized in that, The method comprises the following steps: a current reference value calculation model of a current loop of a converter is constructed based on an error between an actual value and a reference value of active power of a wind turbine, a current loop measured current value of a converter corresponding to the wind turbine at a current moment, and a current saturation value of a PI controller of the converter corresponding to the wind turbine; the current reference value calculation model of the current loop of the converter is converted into a linear expression in an incremental form to obtain a constraint condition of an active power reference value increment of the wind turbine; a deviation between a total sum of active power of all wind turbines and a total active power instruction and a weighted sum of deviations between actual values and reference values of active power of each wind turbine are taken as a target function of a wind turbine active power distribution strategy; constraint conditions of the wind turbine active power distribution strategy are constructed based on wind farm output constraints, wind turbine output constraints, and the constraint condition of the active power reference value increment of each wind turbine; actual values of active power of all wind turbines at a current moment are substituted into discrete space state equations of all wind turbines to obtain active power prediction values at each prediction step; wind farm data at the current moment and the active power prediction values at each prediction step are substituted into the target function of the wind turbine active power distribution strategy to obtain a target active power distribution strategy at each prediction step.

2. A method of active power distribution in a wind farm according to claim 1, characterized in that, The wind farm data at the current moment comprises actual values of active power of each wind turbine, a total active power instruction, and current loop measured current values of a converter corresponding to each wind turbine.

3. A method of active power distribution in a wind farm according to claim 2, characterized in that, An actual value of active power of each wind turbine at each moment is obtained by the following steps: a first wind turbine in a wind farm that is affected by a natural flow is taken as an upstream reference, and wind speeds of each wind turbine at a current moment are calculated based on a single-machine wake model and considering a wake effect; mechanical energy output by each wind turbine at the current moment is calculated through a wind energy formula based on the wind speeds of each wind turbine at the current moment and considering the wake effect; a wake effect attenuation coefficient of each wind turbine is set according to a distance between each wind turbine and the upstream reference; the wake effect attenuation coefficient is inversely proportional to the distance between the wind turbine and the upstream reference; the mechanical energy output by each wind turbine is corrected according to the wake effect attenuation coefficient of each wind turbine, and a corrected mechanical energy of each wind turbine at the current moment is calculated; a correction formula is as follows: ， an actual value of active power of each wind turbine at the current moment is calculated according to the corrected mechanical energy of each wind turbine at the current moment; a calculation formula is as follows: ， wherein, is the wind turbine at time the corrected mechanical energy, wherein, the wind turbine at time the mechanical energy of the wind turbine at time is the wake effect decay coefficient of the wind turbine, is a complex variable, is the wind turbine index, is the time, is the time for converting the mechanical energy to active power dynamic response conversion, is the active power actual value of the wind turbine at time .

4. A method of active power distribution in a wind farm according to claim 1, characterized in that, the current reference value calculation model of the current loop of the converter is as follows: ， ； wherein, is a current value measured by a current loop of a corresponding converter of the wind turbine at a current time, is a proportional controller parameter of the PI controller, is an active power reference value of the wind turbine, is an active power value of the wind turbine, is an integral controller parameter of the PI controller, is a complex variable, is a current reference value of the current loop of the converter, is a current saturation value of the PI controller of the corresponding converter of the wind turbine.

5. A method of active power distribution in a wind farm according to claim 4, characterized in that, the current reference value calculation model of the current loop of the converter is converted into a linear expression in an incremental form; a formula is as follows: ， According to the constant relationship between the integral element and the linear expression, the integral element is converted into , and the converted formula is: ​ ， Based on the converted formula, The constraint condition of the active power reference increment of the wind turbine is obtained as follows: ， wherein, is a wind turbine active power reference value increment, is a wind turbine active power value increment, is a current saturation value of a PI controller of a converter corresponding to the wind turbine, is a current loop measured current value of the converter corresponding to the wind turbine at a current time, is a proportional controller parameter of the PI controller, is an integral controller parameter of the PI controller, is an integral equivalent coefficient, is a complex variable.

6. A method of allocating active power in a wind farm according to claim 1, characterized in that, the target function of the wind turbine active power distribution strategy is as follows: ， ， ， wherein, is a target function of the wind turbine active power distribution strategy, represents a target function of optimizing the deviation between the total sum of all wind turbine active power and the total active power instruction, represents a target function of optimizing the deviation between the actual value of each wind turbine active power and the active power reference value, represents minimizing, is a weight of , is a weight of , is the total prediction step number, is the prediction step index, is the total number of wind turbines, is the wind turbine index, is the active power value of the wind turbine at the prediction step , is the actual value of the active power of the wind turbine at the current time, is the total active power instruction, is the norm square, is the active power reference value of the wind turbine at the prediction step .

7. A method of allocating active power in a wind farm according to claim 1, characterized in that, the wind farm output constraints are constructed based on the total sum of active power of all wind turbines being equal to the total active power instruction; the wind turbine output constraints are constructed based on the active power reference value of each wind turbine being greater than or equal to 0 and being less than or equal to an available capacity of the wind turbine.

8. A method of allocating active power in a wind farm according to claim 1, characterized in that, The discrete space state equations of all wind turbines are obtained by the following steps: a first-order delay function is used to describe dynamic characteristics of an active power control loop of a wind turbine. The dynamic characteristics of the wind turbine active power control loop are converted into a linear expression in incremental form to obtain a state space equation of the wind turbine active power loop; Based on the state space equation of the wind turbine active power loop, a continuous state space equation of all wind turbines is constructed by extending a diagonal matrix; The continuous state space equation of all wind turbines is discretized to obtain a discrete space state equation of all wind turbines.

9. A method of allocating active power in a wind farm according to claim 1, characterized in that, The method for solving the target active power distribution strategy of each prediction step is any one of linear programming, quadratic programming, and genetic algorithm.

10. A wind farm active power distribution system, characterized by The method comprises the following steps: A model construction module is configured to construct a current reference value calculation model of a converter current loop based on an error between an actual value and a reference value of wind turbine active power, a current loop measured current value of a converter corresponding to the wind turbine at a current time, and a current saturation value of a PI controller of the converter; A constraint condition conversion module is configured to convert the current reference value calculation model of the converter current loop into a linear expression in incremental form to obtain a constraint condition of the wind turbine active power reference value increment; A target function construction module is configured to take a weighted sum of deviations between a total active power instruction and a total sum of active power of all wind turbines and between an actual value and a reference value of active power of each wind turbine as a target function of the wind turbine active power distribution strategy; A constraint condition construction module is configured to construct a constraint condition of the wind turbine active power distribution strategy based on a wind farm output constraint, a wind turbine output constraint, and a constraint condition of the active power reference value increment of each wind turbine; An active power acquisition module is configured to substitute actual values of active power of all wind turbines at a current time into the discrete space state equation of all wind turbines to obtain active power prediction values at each prediction step; A solving module is configured to substitute wind farm data at the current time and the active power prediction values at each prediction step into the target function of the wind turbine active power distribution strategy to obtain the target active power distribution strategy at each prediction step.

Images