A wind-storage frequency coordinated control method based on the two-step wind power clustering method

Abstract

This application relates to the field of power system control technology, and in particular to a wind-storage frequency coordinated control method based on a two-step wind power grouping approach. Principal component analysis is used to extract dominant variables as primary grouping indices, and then wind energy utilization rate is introduced as a secondary grouping indices. This further refines the grouping based on the primary grouping, thereby optimizing the matching between the wind farm grouping results and the frequency regulation control strategy. This allows the aggregation model to more accurately represent the actual operating characteristics of the wind farm. After grouping the wind turbines based on the grouping results, differentiated control strategies are matched on a group-by-group basis. Different control strategies are selected based on the magnitude of equivalent rotational speed and equivalent electromagnetic power, thereby achieving accurate coordinated control of the wind turbine group. The aim is to solve the problem of how to improve the frequency response capability of wind farms.

Description

Technical Field

[0001] This application relates to the field of power system control technology, and in particular to a wind-storage frequency coordinated control method based on the two-step wind power grouping method. Background Technology

[0002] With the growth of new energy installed capacity, the new power system exhibits the "dual high" characteristics of high proportion of new energy and high proportion of power electronic equipment. This trend brings two challenges: First, the high proportion of new energy output has strong randomness and volatility, which can lead to wind turbines not being able to perform at their best during dispatch operations, resulting in a decrease in the overall power generation efficiency of the wind farm. Second, the high proportion of power electronic equipment restricts the system's key capabilities such as inertia and frequency regulation capability due to safety concerns. Although wind turbines can use rotor rotational kinetic energy to achieve rapid frequency regulation, there is still a secondary drop in frequency during the wind turbine speed recovery phase, resulting in the limitation of a single frequency regulation resource.

[0003] Among the relevant technical solutions, there are already power systems that use the complementarity of wind energy and energy storage for frequency regulation. However, the inventors discovered during the conception and implementation of this application that in large-scale wind farms, the speed of some wind turbines is too low to be suitable for rotor kinetic energy control. Furthermore, the frequency regulation effect of the wind farm control mode depends on the construction of its model. However, the control mode of traditional new wind farms has the problem of not matching the mathematical model, which makes it difficult for wind farms to have the same frequency response capability as conventional power plants.

[0004] Therefore, this application proposes a control method for a wind-storage system that can quickly respond to system frequency and is capable of frequency regulation and resource complementarity. On the one hand, it improves the matching degree between the control mode and the wind farm model, and on the other hand, it extends the continuous frequency regulation time of the wind farm and suppresses the secondary frequency drop, thereby ensuring the timely frequency response and stable operation of the power grid. Summary of the Invention

[0005] The main purpose of this application is to provide a wind-storage frequency coordination control method based on the two-step wind power clustering method, which aims to solve the problem of how to improve the frequency response capability of wind farms.

[0006] To achieve the above objectives, this application provides a wind-storage frequency coordination control method based on a two-step wind power clustering method, the method comprising:

[0007] S10, the wind turbine units are first clustered using the target wind turbine variables selected based on principal component analysis to obtain the primary wind power cluster. Then, the wind energy utilization rate is introduced to perform a second clustering on the primary wind power cluster to obtain the target wind power cluster result.

[0008] S20, based on the target wind power grouping results, the wind turbines are grouped, and the equivalent speed and equivalent electromagnetic power in the wind turbine group are calculated using the grouped wind turbines as the unit.

[0009] S30, based on the equivalent rotational speed and the equivalent electromagnetic power, determine the target control strategy for the doubly-fed asynchronous wind turbine in each wind turbine group, and perform frequency regulation based on the target control strategy corresponding to each wind turbine group.

[0010] Optionally, in step S30, the step of determining the target control strategy for the doubly-fed asynchronous wind turbine in each wind turbine group based on the equivalent rotational speed and the equivalent electromagnetic power includes:

[0011] When the equivalent rotational speed is within a first preset rotational speed range and the equivalent electromagnetic power is greater than a first preset power threshold, the target control strategy is determined to be a comprehensive inertia control strategy.

[0012] When the equivalent rotational speed is within a first preset rotational speed range and the equivalent electromagnetic power is less than or equal to a first preset power threshold, the target control strategy is determined to be a maximum power point tracking strategy.

[0013] When the equivalent speed is not within the first preset speed range and the equivalent speed is greater than the first preset speed threshold, the target control strategy is determined to be the variable pitch angle control strategy.

[0014] When the equivalent speed is not in the first preset speed range, and the equivalent speed is less than or equal to the first preset speed threshold, and the equivalent electromagnetic power is greater than the second preset power threshold, the target control strategy is determined to be the overspeed load reduction-variable pitch angle control strategy.

[0015] When the equivalent rotational speed is not within the first preset rotational speed range, and the equivalent rotational speed is less than or equal to the first preset rotational speed threshold, and the equivalent electromagnetic power is less than or equal to the second preset power threshold, the target control strategy is determined to be the overspeed load reduction-inertia control strategy.

[0016] Optionally, the integrated inertia control strategy includes: adjusting the virtual inertia control power ΔP1 and the droop control power ΔP2 according to the frequency difference;

[0017] The variable pitch angle control strategy includes: keeping the rotational speed of the doubly fed asynchronous wind turbine constant and increasing the pitch angle of the wind turbine.

[0018] The overspeed load reduction-pitch angle control strategy includes an overspeed load reduction strategy and the pitch angle control strategy, wherein the overspeed load reduction strategy includes increasing the rotational speed under the maximum power point tracking strategy;

[0019] The overspeed load reduction-inertia control strategy includes the overspeed load reduction strategy and the integrated inertia control strategy.

[0020] Optionally, the frequency-modulated active power variation ΔP in the integrated inertia control strategy sum Satisfy the following expression:

[0021]

[0022] In the formula, ΔP1 is the virtual inertia control power, ΔP2 is the droop control power, K1 is the virtual inertia parameter, and K2 is the droop parameter. For frequency difference, This represents the rate of change of frequency.

[0023] Optionally, both the virtual inertia parameter and the droop parameter are adjusted with the rate of change of frequency;

[0024] Among them, the adjusted virtual inertial parameters Satisfy the following expression:

[0025]

[0026] In the formula, the time constant T is proportional to the output power and frequency modulation exit time under virtual inertial control, and s is the Laplace operator;

[0027] The adjusted droop parameter K2,1 satisfies the following expression:

[0028] ;

[0029] ;

[0030] In the formula, t is the current simulation time, t re k is the time for the wind turbine to exit frequency regulation. c Parameters used to control the rate of change;

[0031] The frequency-modulated active power change ΔP of the integrated inertia control strategy after parameter adjustment sum satisfy:

[0032] .

[0033] Optionally, in the variable pitch angle control strategy, the wind turbine pitch angle satisfies the following expression:

[0034]

[0035] In the formula, This is a reference value for the propeller pitch angle. The increment of the propeller pitch angle that needs to be adjusted corresponds to the change in system frequency. f is the pitch angle droop coefficient, and f is the measurement frequency. N This is the rated frequency.

[0036] Optionally, in the overspeed load reduction strategy, the active power P output by the doubly-fed asynchronous wind turbine generator... del Satisfy the following expression:

[0037]

[0038] In the formula, P MPPT This refers to the active power output by the wind turbine in MPPT mode. The second-best tip speed ratio; C p,de The wind energy utilization coefficient under unloaded mode; Power point tracking factor in load reduction mode; d% is the load reduction rate. This refers to the generator rotor speed. The rotor speed of the wind turbine. air density, The optimal tip speed ratio is given by S, where S is the swept area and r is the radius of the fan blade.

[0039] Optionally, the target control strategy further includes:

[0040] Determine whether the equivalent frequency of the wind turbine group is less than a preset first frequency threshold and greater than a preset second frequency threshold, wherein the preset first frequency threshold is greater than the preset second frequency threshold;

[0041] If so, execute the integrated inertia control strategy.

[0042] During the execution of the comprehensive inertia control strategy, when it is determined that the wind turbine speed has recovered, the output power of the energy storage system is increased to compensate for the reduced output power of the wind turbine.

[0043] The total output ΔP of the energy storage system B satisfy:

[0044]

[0045] in:

[0046]

[0047]

[0048]

[0049]

[0050] In the formula, This refers to the rated active power of the wind turbine; The rated power of the energy storage The battery charge / discharge coefficient. To compensate for the virtual inertia coefficient of the secondary drop of the wind turbine in the improved energy storage system, It is the energy storage droop control coefficient corresponding to the failure of the wind turbine to meet the frequency performance requirements after it is removed from frequency regulation. Fan speed rate of change, This is the energy storage droop control coefficient. For frequency difference, t re k is the time for the wind turbine to exit frequency regulation. c Parameters for controlling the rate of change, The virtual inertia coefficient is used to compensate for the secondary drop of the wind turbine in the energy storage system before the improvement.

[0051] Optionally, in S10, the target wind turbine variables include at least one of rotor speed, active power output of the wind turbine, pitch angle, d-axis component of stator current, and d-axis component of rotor current.

[0052] Optionally, in step S10, the expression for the wind energy utilization rate is:

[0053]

[0054] In the formula, Let λ be the wind energy utilization rate, λ be the tip speed ratio, β be the blade pitch angle, and Λ be the intermediate variable for converting the relationship between λ and β.

[0055] This application has at least the following beneficial effects:

[0056] 1. From the perspective of "wind turbine cluster", a large-scale wind farm equivalent modeling method based on two-step clustering is proposed. By establishing a clustering index system with different dimensions, the differentiated control strategy matching of wind turbine clusters can be realized, which can significantly improve the frequency regulation capacity utilization rate of wind turbines.

[0057] 2. The initial grouping of wind farms uses principal component analysis (PCA) to extract dominant variables from multiple features as grouping indices; the secondary grouping takes into account the wind curtailment problem and introduces the wind energy utilization coefficient as a grouping index, making the grouping results more reasonable. The proposed aggregation and equivalent scheme can improve simulation efficiency while ensuring model accuracy.

[0058] 3. The proposed wind-storage multi-energy complementary frequency regulation method not only overcomes the limitations of a single frequency regulation resource, but also extends the continuous frequency regulation time of the wind farm through the timing coordination between resources, ensuring the stable operation of the power grid.

[0059] 4. A rapid response mechanism for wind and energy storage collaboration is established. Improved comprehensive inertial control of parameters facilitates overall optimization of frequency regulation performance. Simultaneously, to mitigate the negative impact of secondary frequency drops during turbine speed recovery, an energy storage support strategy and a turbine speed recovery mechanism are designed, effectively suppressing secondary frequency drops. Attached Figure Description

[0060] Figure 1 This is a flowchart illustrating the first embodiment of the wind-storage frequency coordination control method based on the two-step wind power clustering method of this application.

[0061] Figure 2 This is a comprehensive inertial control diagram with improved parameters according to the embodiments of this application;

[0062] Figure 3 This is a control framework diagram of the wind storage system involved in the embodiments of this application;

[0063] Figure 4 This is a model diagram of the equivalent collector circuit involved in the embodiments of this application;

[0064] Figure 5 This is a schematic diagram of the wind farm topology involved in the embodiments of this application;

[0065] Figure 6 This is a schematic diagram of the primary and secondary clustering phylogenetic diagrams involved in the embodiments of this application;

[0066] Figure 7 This is a schematic diagram of repeated measures variance analysis for calculating the sample size of the wind farm involved in the embodiments of this application;

[0067] Figure 8 This is a schematic diagram comparing the active power response curves of various wind farm models under stable operating conditions in the embodiments of this application;

[0068] Figure 9 This is a schematic diagram comparing the active power response curves of various wind farm models under random wind disturbance conditions in the embodiments of this application.

[0069] Figure 10 This is a simulation diagram illustrating the primary frequency regulation of the equivalent wind turbine unit with coordinated energy storage in an embodiment of this application.

[0070] Figure 11 This is a schematic diagram of the wind speed scenario at a certain moment and the equivalent wind speed distribution of each group of machines involved in the embodiments of this application.

[0071] Figure 12 This is a schematic diagram of the system frequency deviation curves before and after the collaborative optimization control involved in the embodiments of this application;

[0072] Figure 13 This is a schematic diagram of the active power response curve of the wind farm model involved in the embodiments of this application;

[0073] Figure 14 This is a schematic diagram of the system frequency deviation curves before and after the collaborative optimization control involved in the embodiments of this application.

[0074] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0075] To better understand the above technical solutions, exemplary embodiments of this disclosure will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of this disclosure to those skilled in the art.

[0076] First Embodiment

[0077] Reference Figure 1 This embodiment provides a wind-storage frequency coordination control method based on a two-step wind power clustering method, applicable to a wind-storage system including wind turbines and an energy storage system. The method includes the following steps:

[0078] S10, the wind turbine units are first clustered using the target wind turbine variables selected based on principal component analysis to obtain the primary wind power cluster. Then, the wind energy utilization rate is introduced to perform a second clustering on the primary wind power cluster to obtain the target wind power cluster result.

[0079] In this embodiment, the wind turbine first uses principal component analysis to extract the dominant variables as primary clustering indicators, optimizes the matching between the clustering results and the frequency regulation control strategy, and then introduces wind energy utilization rate as a secondary clustering indicator to further refine the clustering based on the primary clustering, so that the aggregation model can more accurately represent the actual operating characteristics of the wind farm.

[0080] In some alternative implementations, the target wind turbine variables include at least one of rotor speed, active power output of the wind turbine, pitch angle, d-axis component of stator current, and d-axis component of rotor current. Furthermore, the primary wind power grouping is determined through the following steps:

[0081] (1) Calculation of initial values ​​of state variables: For large-scale wind turbine groups with frequency regulation control, the rotor speed ω, which is strongly correlated with active power, is selected as the equivalent value. r Active power output P of wind turbine e Pitch angle β, stator current d-axis component i sd d-axis component of rotor current i rd As the initial variable for the initial wind turbine partitioning, its calculation formula is:

[0082]

[0083] In the formula, ρ is the air density; r is the radius of the fan blades; S is the swept area; C p (λ,β) is the wind energy utilization coefficient, which is a nonlinear high-order function of the tip speed ratio λ and the pitch angle β.

[0084] (2) Data matrix construction: The original data is arranged in rows to form a matrix X: forming p state characterization parameters for n wind turbines in the wind farm. The original sample matrix X. For a p-dimensional random vector X = (X1, X2, ..., X... P ), where X j =(x j1 ,x j2 ,…,x jP );

[0085]

[0086] In the formula, p represents the fifth state characterization parameter.

[0087] (3) Data standardization: The matrix X is standardized to obtain a standardized matrix, thus eliminating the influence of dimensions;

[0088]

[0089] In the formula, To calculate the mean, The standard deviation can be expressed as:

[0090]

[0091] (4) Calculation of correlation coefficient matrix: Solve for the correlation coefficient matrix R of the standardized matrix X:

[0092]

[0093] (5) Eigenvalue and eigenvector extraction: Calculate the eigenvalues ​​and eigenvectors of R according to the following formula:

[0094]

[0095] In the formula, λ i Let a be the i-th eigenvalue. i Let be the eigenvector corresponding to the i-th eigenvalue, and λ1≥λ2≥…≥λ p ≥0 (R is a positive semidefinite matrix).

[0096]

[0097] (6) Contribution rate calculation: By calculating the variance contribution rate of each principal component and its cumulative value, we can evaluate its ability to explain the variance of the original data, thereby determining the number of principal components to be retained, and selecting the principal components that contribute the most to the interpretation of the data based on the loading matrix.

[0098]

[0099] In the formula, η i η is the variance contribution rate of the i-th principal component; cum The cumulative variance contribution rate of the first k principal components is typically required to be η. cum >85%; k is the number of principal components selected; m is the number of features of the original data, and m=p.

[0100] element l in load matrix L ij Let l represent the correlation coefficient between the i-th original variable and the j-th principal component. ij The larger the absolute value of the value, the stronger the correlation between the original variable and the principal component. The calculation formula is as follows:

[0101]

[0102] In the formula, V a Λ is the eigenvector matrix; λ It is a diagonal matrix, where the diagonal elements are eigenvalues ​​and the remaining elements are all zero; This represents taking the square root of each element of the eigenvalue moments.

[0103] In some alternative implementations, the expression for the wind energy utilization rate is:

[0104]

[0105] In the formula, Let λ be the wind energy utilization rate, β be the blade tip speed ratio, β be the blade pitch angle, and Λ be the intermediate variable relating λ and β. Furthermore, wind energy utilization rate can be used as a new indicator to reflect wind curtailment, thereby improving the rationality of clustering. The secondary grouping based on wind energy utilization rate can be performed using the following steps:

[0106] S20, based on the target wind power grouping results, the wind turbines are grouped, and the equivalent speed and equivalent electromagnetic power in the wind turbine group are calculated using the grouped wind turbines as the unit.

[0107] In this step, after obtaining the target wind turbine clustering results through two groupings, differentiated control strategy matching is implemented on the clustered wind turbine groups as units to improve the frequency regulation capacity utilization rate of the wind turbine groups. Differentiated control strategy matching refers to executing different control strategies based on different conditional criteria. In this embodiment, the parameters used as conditional criteria are the equivalent speed and equivalent electromagnetic power in the wind turbine group.

[0108] Equivalent speed refers to the virtual speed of each wind turbine in a wind turbine cluster when its rotor motion equation is equivalent to that of a single-unit system; and equivalent electromagnetic power refers to the weighted sum of the electromagnetic power of the generator of each wind turbine. The calculation of these two parameters can be achieved based on existing formulas, and will not be elaborated upon here.

[0109] S30, based on the equivalent rotational speed and the equivalent electromagnetic power, determine the target control strategy for the doubly-fed asynchronous wind turbine in each wind turbine group, and perform frequency regulation based on the target control strategy corresponding to each wind turbine group.

[0110] This step is the key step in this embodiment, and it will be described in detail below:

[0111] The target control strategies for doubly-fed asynchronous wind turbines include five strategies: integrated inertia control strategy, maximum power point tracking strategy, pitch angle control strategy, overspeed load reduction-pitch angle control strategy, and overspeed load reduction-inertia control strategy, among which:

[0112] When the equivalent rotational speed is within a first preset rotational speed range and the equivalent electromagnetic power is greater than a first preset power threshold, the target control strategy is determined to be a comprehensive inertia control strategy.

[0113] It should be further explained that virtual inertial control stops providing active power or even absorbs energy after the lowest frequency point, with a short response time; droop control has a slow response but can continuously increase active power output with Δf. Therefore, the two are often used in combination, a combination known as integrated inertial control. The frequency-modulated active power change ΔP in the integrated inertial control strategy... sum Satisfy the following expression:

[0114]

[0115] In the formula, ΔP1 is the virtual inertia control power, ΔP2 is the droop control power, K1 is the virtual inertia parameter, and K2 is the droop parameter. For frequency difference, This represents the rate of change of frequency.

[0116] It needs further explanation that DFIG's virtual inertial control introduces a frequency change rate-related term into the active power control reference command of the unit, making the originally "zero-inertia" wind turbine exhibit inertial characteristics similar to those of a synchronous generator unit, thus increasing its active power generation. for:

[0117]

[0118] Furthermore, droop control is essentially a differential control that can simulate the frequency regulation characteristics of a synchronous generator set, i.e., adjusting the intake air volume through Δf to increase active power. Similarly, droop control of a DFIG simulates the droop characteristics of a synchronous generator by adding a control module that releases rotor kinetic energy following the frequency difference.

[0119] Δf is the difference between the measured frequency f and the rated frequency f. N The deviation is calculated by multiplying the filtered frequency difference by the droop gain K. p This yields the additional active power ΔP generated when the wind turbine is under droop control. p :

[0120]

[0121] When the equivalent rotational speed is within a first preset rotational speed range and the equivalent electromagnetic power is less than or equal to a first preset power threshold, the target control strategy is determined to be a maximum power point tracking strategy.

[0122] It should be further explained that the Maximum Power Point Tracking (MPPT) strategy aims to maximize the output power of the wind turbine by adjusting the rotational speed or pitch angle. This part is existing technology and will not be elaborated here.

[0123] When the equivalent speed is not within the first preset speed range and the equivalent speed is greater than the first preset speed threshold, the target control strategy is determined to be the variable pitch angle control strategy.

[0124] It should be further explained that the variable pitch angle control strategy, while maintaining a constant DFIG (Doubly Fed Induction Generator) speed, increases the turbine pitch angle to allow the turbine to operate in a suboptimal state, reserving some active power for backup. This strategy is suitable for constant speed and constant power regions. When system disturbances occur, the pitch angle controller can decrease the pitch angle... To release reserved power, in some alternative implementations, the actual pitch angle for:

[0125]

[0126] In the formula, This is a reference value for the propeller pitch angle; This is the pitch angle droop coefficient. The pitch angle that needs to be adjusted to correspond to changes in system frequency.

[0127] When the equivalent speed is not in the first preset speed range, and the equivalent speed is less than or equal to the first preset speed threshold, and the equivalent electromagnetic power is greater than the second preset power threshold, the target control strategy is determined to be the overspeed load reduction-variable pitch angle control strategy.

[0128] The overspeed load reduction-pitch angle control strategy includes an overspeed load reduction strategy and a pitch angle control strategy, wherein the overspeed load reduction strategy includes increasing the rotational speed under the maximum power point tracking strategy.

[0129] Under the overspeed load shedding strategy, when the DFIG operates in MPPT mode, increasing or decreasing the rotor speed can reduce its active power output. However, decreasing the speed will reduce the unit's operational stability. Therefore, overspeed load shedding control that increases the rotor speed can be selected. In some optional implementations, the active power P output by the doubly-fed asynchronous wind turbine generator... del Satisfy the following expression:

[0130]

[0131] In the formula, P MPPT This refers to the active power output by the wind turbine in MPPT mode. The second-best tip speed ratio; C p,de The wind energy utilization coefficient under unloaded mode; Power point tracking factor in load reduction mode; d% is the load reduction rate. This refers to the generator rotor speed. The rotor speed of the wind turbine. air density, The optimal tip speed ratio is given by S, where S is the swept area and r is the radius of the fan blade.

[0132] When the equivalent rotational speed is not within the first preset rotational speed range, and the equivalent rotational speed is less than or equal to the first preset rotational speed threshold, and the equivalent electromagnetic power is less than or equal to the second preset power threshold, the target control strategy is determined to be the overspeed load reduction-inertia control strategy.

[0133] It should be further explained that the overspeed load reduction-inertia control strategy includes the overspeed load reduction strategy and the integrated inertia control strategy, that is, the two control strategies are combined for control.

[0134] In the technical solution provided in this embodiment, principal component analysis is used to extract dominant variables as primary clustering indicators. Then, wind energy utilization rate is introduced as a secondary clustering indicator. The clustering is further refined based on the primary clustering, thereby optimizing the matching between the wind farm clustering results and the frequency regulation control strategy. This makes the aggregation model more accurately represent the actual operating characteristics of the wind farm. After the wind turbines are clustered based on the clustering results, differentiated control strategies are matched on a per-group basis. Different control strategies are selected based on the magnitude of the equivalent speed and equivalent electromagnetic power, thereby achieving accurate coordinated control of the wind turbine group.

[0135] Second Embodiment

[0136] Based on the first embodiment, the virtual inertia control in the integrated inertial control is dynamically adjusted with the frequency change rate dΔf / dt. dΔf / dt reaches its maximum value at time t0 during the initial frequency change, and gradually decreases as the unit's frequency regulation mechanism responds. nadir1 At time t0, when the frequency drops to its lowest point, dΔf / dt = 0; after the frequency enters the recovery phase, dΔf / dt > 0; in steady state, dΔf / dt returns to zero. t0 is the initial frequency change time, t... nadir1 This represents the time corresponding to the lowest frequency point.

[0137] For virtual inertia control, to smooth the step change in energy at time t0, a first-order inertial element can be introduced before parameter K1 to slow down the energy release rate; t nadir1 At a certain moment, the control begins to absorb energy from the system. To stabilize the system frequency, K1 needs to be adjusted after dΔf / dt<0 transitions to dΔf / dt>0. Setting K1 directly to a negative value at this point would further exacerbate frequency fluctuations. Therefore, setting K1 to zero when dΔf / dt>0 is the optimal choice. The new virtual inertia control parameter K... 1,1 for:

[0138]

[0139] The virtual inertia control power after parameter improvement is:

[0140]

[0141] Therefore, parameter K 1,1 After detecting dΔf / dt>0, ΔP1=0, then K 1,1 =0 same virtual inertia control exits one frequency modulation. K 1,1 The time constant T value is proportional to the output power and frequency modulation exit time under virtual inertial control.

[0142] The results were:

[0143]

[0144] In the formula, the time constant T is proportional to the output power and frequency modulation exit time under virtual inertial control, and s is the Laplace operator.

[0145] On the other hand, after the speed reduction phase ends, the system enters the speed recovery phase. To avoid excessive energy demand during speed recovery that could trigger a new power surge, the frequency regulation power of the wind turbine needs to be limited during the frequency recovery phase. A droop control strategy combined with energy storage can be used to mitigate the impact of the secondary frequency drop. During frequency recovery, virtual inertia control gradually weakens to zero, while droop control becomes dominant. However, according to the droop control principle, a frequency deviation of Δf will lead to continuous power output, causing the rotor to remain at a low speed for an extended period, threatening the safety of the wind turbine. Therefore, when the frequency stabilizes, the wind turbine will exit frequency regulation, but this may cause a sharp decrease in the system's frequency regulation power, further exacerbating the secondary frequency drop. re This refers to the time when the fan stops frequency regulation.

[0146] When the system frequency passes t nadir1 Subsequently, to avoid a secondary drop in frequency, the output of the droop control needs to be gradually reduced. A Logistic function can be introduced, and a smooth descent can be achieved by dynamically adjusting the parameter K2. Its mathematical expression is as follows:

[0147]

[0148] In the formula, a is the initial value, taken as a=1; b is the final value, taken as b=0; k c k is a parameter used to control the rate of change. c <0 indicates a downward trend. t c The center point (inflection point) is where the rate of change in wind turbine power output is fastest, and the increase in system power deficit is also fastest. Frequency drops generally begin near the inflection point. Based on the frequency regulation time requirements of large wind farms, the function is set to t. c =t re The center point is +7.5s, which is the time t when the wind turbine is scheduled to exit frequency regulation. re By adding a certain margin to the original function, the rate of descent of the function is slowed down, ensuring a smoother frequency modulation process. This yields the transformed function. :

[0149]

[0150] Assume t c =17.5s, as k c The increase in absolute value, The steeper the descent trend, the more necessary it is to choose a reasonable k. cThe value, while ensuring sufficient frequency support during the primary frequency regulation of the fan, aims to minimize the rate of droop control reduction. This embodiment... Take k c = -0.6, at this point the function 10 seconds before the graphic, It basically remains near the initial value, because the lowest frequency point generally occurs within 5 seconds of the initial frequency change, during which time the DFIG output does not decrease; 10s~25s, Rapid descent; approximately 30 seconds This is equivalent to DFIG exiting FM. Because... There are no extreme points throughout the process. This smooth transition characteristic effectively alleviates the problem of secondary frequency drop caused by the sudden decrease in DFIG droop output, thereby reducing the degree of frequency drop. The new droop control parameter K... 2,1 The droop control power is shown below:

[0151]

[0152]

[0153] The integrated inertial control of DFIG after parameter improvement is as follows:

[0154] In the initial stage of frequency regulation, the improved virtual inertial control and droop control work together; during the frequency recovery phase, the improved virtual inertial control withdraws from frequency regulation to avoid frequency fluctuations caused by absorbing energy from the grid; after the frequency passes the lowest point, the droop control outputs power according to... The change pattern automatically exits frequency modulation, reducing the risk of secondary frequency drops. Therefore, the frequency modulation active power change ΔP of the DFIG... sum for:

[0155]

[0156] Integrated inertial control with improved parameters, such as Figure 2 As shown in the diagram, traditional methods experience a significant power drop during the initial speed recovery phase, which can cause a substantial impact on the system frequency. The two-stage control scheme proposed in this embodiment, by adding a speed descent phase, avoids the rapid increase in speed during the initial recovery phase, prolongs the acceleration process time, significantly reduces the instantaneous power drop, and lowers the instantaneous impact on the system.

[0157] After simplification, the adjusted droop parameter K2,1 satisfies the following expression:

[0158] ;

[0159] ;

[0160] In the formula, t is the current simulation time, tre k is the time for the wind turbine to exit frequency regulation. c Parameters used to control the rate of change;

[0161] Frequency-modulated active power variation ΔP of the integrated inertia control strategy after parameter adjustment sum satisfy:

[0162]

[0163] Third Embodiment

[0164] Based on any of the above embodiments, this embodiment introduces an energy storage system as a flexibility adjustment unit, and employs integrated inertial control with improved parameters to assist wind farm frequency regulation. Specifically:

[0165] Step S100: Determine whether the equivalent frequency of the wind turbine group is less than a preset first frequency threshold and greater than a preset second frequency threshold, wherein the preset first frequency threshold is greater than the preset second frequency threshold.

[0166] In this step, the equivalent frequency refers to the virtual frequency when the wind turbine group is equivalent to a single-unit system.

[0167] In some alternative implementations, the expression for determining the recovery of the fan speed is satisfied as follows:

[0168]

[0169] In the formula, This represents the fan speed at the initial time t. This represents the fan speed at time t+T.

[0170] Step S200: If yes, execute the integrated inertia control strategy;

[0171] Step S300: During the execution of the comprehensive inertia control strategy, when it is determined that the wind turbine speed recovers, the output power of the energy storage system is increased to compensate for the reduced output power of the wind turbine.

[0172] In this step, as the power output under DFIG droop control gradually decreases, the DFIG speed... Upon recovery, the energy storage system needs to promptly replenish the power output lost during the DFIG's frequency regulation withdrawal process to prevent a secondary frequency drop. This means the energy storage system's output must increase accordingly from zero to compensate for the reduced output of the wind turbine and improve the steady-state frequency of the system's primary frequency regulation. (Based on the wind turbine shown...) The trend of coefficient changes, the virtual inertia coefficient of energy storage compensating for the secondary drop of the wind turbine. and contribution for:

[0173]

[0174] If the frequency still does not meet the frequency performance requirements after the DFIG is discontinued, then the droop control coefficient of the energy storage system... and contribution for:

[0175]

[0176] Total output ΔP of energy storage-assisted DFIG frequency regulation B for:

[0177]

[0178] In the above formula, This refers to the rated active power of the wind turbine; The rated power of the energy storage The battery charge / discharge coefficient. To compensate for the virtual inertia coefficient of the secondary drop of the wind turbine in the improved energy storage system, It is the energy storage droop control coefficient corresponding to the failure of the wind turbine to meet the frequency performance requirements after it is removed from frequency regulation. Fan speed rate of change, This is the energy storage droop control coefficient. For frequency difference, t re k is the time for the wind turbine to exit frequency regulation. c Parameters for controlling the rate of change, The virtual inertia coefficient is used to compensate for the secondary drop of the wind turbine in the energy storage system before the improvement.

[0179] Fourth embodiment

[0180] Based on any of the foregoing embodiments, this embodiment also provides a wind storage system, the control framework of which is as follows: Figure 3 As shown, the system includes wind turbine generators and an energy storage battery system. The DFIG is connected to the grid via dual PWM converters. The wind farm employs a multi-machine equivalent method based on the concept of cluster aggregation, with each wind turbine group matched with a corresponding control strategy. Through the complementarity of control strategies among the turbine groups, the frequency regulation potential of the wind turbines themselves is fully explored. The energy storage battery is composed of lead-carbon batteries and uses integrated inertial control. Energy exchange and storage are completed through a bidirectional DC / DC converter, and then the energy is connected to the AC bus via DC / AC.

[0181] Furthermore, this embodiment also provides a mathematical equivalent model for establishing a system composed of wind turbine generators and energy storage battery systems within a wind-storage system, referred to as a wind turbine cluster model. The wind turbine cluster model includes an equivalent wind turbine generator model and an equivalent power collection line model. The equivalent wind turbine generator model includes an equivalent wind turbine model, an equivalent transmission system model, and an equivalent generator and transformer model. The equations for the equivalent wind turbine model are as follows:

[0182]

[0183] In the formula, C peq C is the wind energy utilization coefficient of the equivalent unit. ph , where h and N are the wind energy utilization coefficients of the h-th wind turbine, and N is the number of wind turbines in the turbine group, 1≤h≤N.

[0184]

[0185] In the formula, P meq P is the sum of the mechanical power output of N wind turbines within the same turbine cluster; mh The mechanical power output of the h-th wind turbine unit; V is the air density; S is the swept area of ​​the wind turbine; V h V represents the effective input wind speed of the h-th wind turbine unit. eq The wind speed is equivalent to that of a wind turbine. Electromagnetic power P e_eq It equals the mechanical power multiplied by the generator efficiency. To simplify the calculation, the generator efficiency can be taken as 1.

[0186]

[0187] In the formula, V eq This refers to the equivalent wind speed of the same turbine group.

[0188]

[0189] In the formula, ω eq The equivalent speed of the same engine group; ω h Let be the rotor speed of the h-th fan in the same group.

[0190]

[0191] In the formula, This refers to the equivalent pitch angle of the same aircraft group; Let h be the pitch angle of the h-th wind turbine in the same group.

[0192] The equivalent transmission system model equations are as follows:

[0193]

[0194] In the formula, K eq and D eq These are the equivalent shaft stiffness coefficient and damping coefficient of the unit, respectively.

[0195] The equations for the equivalent generator and transformer model are as follows:

[0196]

[0197] In the formula, S Geq P Geq X s_eq R s_eq These are the rated capacity, active power, stator reactance, and stator impedance of the equivalent generator in the generator group, respectively. , , , These are the rated capacity, active power, stator impedance, and stator resistance of each generator in the same group.

[0198]

[0199] In the formula, S Teq and Z Teq These are the apparent power of the equivalent generator unit's terminal transformer and the impedance of the terminal transformer, respectively. and These are the apparent power and impedance of the turbine terminal transformer for wind turbine unit h, respectively.

[0200] In addition, the wind turbine cluster model includes an equivalent wind turbine generator model and an equivalent power collection line model, the equivalent power collection line model as follows: Figure 4 As shown, the equation is as follows:

[0201]

[0202] In the formula, This refers to the power loss between the outlet of the first fan in the first row and the grid connection point. , , , For complex power, voltage, current, and impedance: Voltage difference between the outlet of the first fan in the first row and the grid connection point: Subscript middle, For action For example, , =1~2; the superscript * indicates conjugate.

[0203]

[0204] In the formula, This refers to the power loss between the outlet of the second fan in the second row and the grid connection point.

[0205]

[0206] In the formula, The sum of the collector network losses corresponding to the equivalent machine after the aggregation of the A-th group of DFIGs before equivalence is assumed to be the sum of the DFIGs in the first row and the second row and the second column.

[0207]

[0208] In the formula, Z eq-A Based on the principle that the losses of the collector network are equal before and after the equivalence, the collector network impedance corresponding to the equivalent machine after the aggregation of the Ath group of DFIGs is determined.

[0209] When the voltage difference within the wind farm is ignored, the equivalent charging capacitor C eq-A This is the sum of the charging capacitances of all cables in group A before the equivalent value is reached:

[0210] .

[0211] Verification Example 1

[0212] Based on any of the above embodiments, this embodiment uses simulation to verify the effectiveness of the proposed large-scale wind turbine aggregation and equivalence method based on the two-step wind power clustering method:

[0213] A grid-connected system consisting of 50 1.5MW DFIG units was built based on the RTLAB platform. Its topology is as follows: Figure 5 As shown. The system raises the generator terminal voltage to 35kV through a 0.69kV / 35kV transformer, and then connects to the grid via three branches to the 35kV / 230kV wind farm step-up substation: Branch A contains 18 units, Branch B contains 15 units, and Branch C contains 17 units.

[0214] (1) Extraction of wind turbine component group indicators and establishment of aggregated equivalent models:

[0215] The wind speed at a certain moment in the operating data of the doubly-fed wind farm was selected as the initial input wind speed for each wind turbine in the detailed model of the wind farm. The parameters of each unit are shown in Table 1 and Table 2 respectively:

[0216] Table 1: Wind speed parameters of each wind turbine

[0217]

[0218] Table 2: Basic Simulation Parameters of Main Equipment and Circuits

[0219]

[0220] The initial values ​​of the DFIG cluster indexes are obtained based on the calculation of the initial values ​​of the state variables, as shown in Table 3.

[0221] Table 3: Characteristic Quantities of Each Wind Turbine

[0222]

[0223] Principal component analysis was performed on the data using SPSS software.

[0224] Table 4: KMO and Bartlett's Sphericity Tests

[0225]

[0226] First, the suitability of the data was assessed using R-type factor analysis. As shown in Table 4, the KMO value was 0.611 (≥0.5), indicating that the data was suitable for factor analysis; the Bartlett's test of sphericity had a p-value of 0.000 (<0.05), rejecting the null hypothesis and proving that there was a significant correlation between the variables, making the data suitable for principal component analysis.

[0227] Factors were extracted based on the principle that the eigenvalues ​​were greater than 1. Table 5 shows that there are two factors with eigenvalues ​​greater than 1 in the variable correlation matrix, namely 3.514 and 1.136, with a cumulative contribution rate of 92.994%, and two principal components were finally extracted.

[0228] Table 5: Cumulative Variance from Factor Analysis

[0229]

[0230] Therefore, extracting the first two factors can reflect 92.994% of the original variable's variance, satisfying the principle that the cumulative variance contribution rate is greater than 85%. Based on this, the first two principal components, ωr and Pe, can well cover most of the operating parameter information in the original DFIG data and can be used as the standard for cluster partitioning. To further clarify the variable characteristics in the principal components, principal component analysis was performed on these two factors, and the final result is the rotational loading matrix shown in Table 6.

[0231] Table 6: Rotated Factor Loadings

[0232]

[0233] As shown in Table 6, after considering redundancy and correlation, among the variables characterizing the operating state of a doubly-fed wind turbine, the rotor speed ω is the most important. r Grid-connected active power output P e It can more completely reflect the dynamic characteristics of the doubly fed fan during frequency regulation.

[0234] To delve deeper into the intrinsic relationships between the data, this embodiment employs R-type clustering analysis to process the clustering indicators extracted by principal component analysis. The results are as follows: Figure 6 As shown.

[0235] (3) Increase the sample size calculation for wind turbine units:

[0236] To compare the similarity of three clustering methods (aggregate equivalence method, K-means clustering, and single-machine multiplication method) with the output power of detailed models and to determine the required sample size, this paper uses the F-distribution of G-Power software for sample size estimation based on the characteristics of multiple complex models. By setting repeated measures ANOVA, this study can directly compare the similarity of the three clustering methods in the same wind farm, thereby improving the accuracy of the analysis. In the sample size calculation, this paper sets the significance level (the probability of committing a Type I error) to α1=0.05, the statistical power (the probability of the conclusion being correct) to Power(1-β1)=0.8, and uses the moderate effect size to f=0.25.

[0237] Table 7: Sample Size Calculation for G-Power

[0238]

[0239] The calculation results are as follows Figure 7 As shown in Table 7, a total sample size of at least 28 wind turbines is required. Insufficient sample size may lead to the following problems: (1) Differences may not be clearly reflected: the similarity differences between the three clustering methods and the output power of the detailed model may be difficult to detect; (2) Resource optimization may be limited: a small sample size may result in the research results failing to provide effective support for wind farm model optimization; (3) Insufficient stability of results: the research results may be affected by extreme values ​​or random fluctuations, affecting the reliability of the research conclusions. Therefore, ensuring a sufficient sample size is the key to the design of this study, which can effectively improve statistical power, control the error rate, and provide a scientific basis for wind farm model optimization.

[0240] (2) Comparison of the external response of four equivalent models of wind turbines under different operating conditions:

[0241] To verify the effectiveness of the algorithm, based on the above clustering results, the active power curves at the grid connection under different operating conditions were compared and analyzed, including the output results of the single-machine multiplication method, the traditional K-means clustering algorithm, the clustering algorithm in this paper, and the detailed model.

[0242] 1) Comparative analysis under steady-state operating conditions

[0243] Based on the wind speeds of each wind turbine in Table 1, the active power output characteristic curves at their busbars under four equivalent models can be obtained, such as... As shown, during the 25-45 second period after the system reaches steady state, a comparative analysis of the output power of each model and the absolute error of the detailed model reveals that the proposed improved model has an error of only 0.0788, while the errors of the traditional model and the single-machine model reach 0.4328 and 2.5782, respectively. The calculation results show that, under steady-state operating conditions, the improved model achieves an accuracy improvement of 81.79% and 96.94% compared to the traditional model and the single-machine model, respectively.

[0244] 2) Comparative analysis under random wind disturbance conditions

[0245] like Figure 9 As shown, at t=37s, the system is subjected to random wind disturbance, with the wind speed exhibiting a dynamic characteristic of first increasing and then decreasing, and the disturbance duration is less than 2 seconds. For this transient process, the active power output response characteristics at the busbar of four equivalent models were compared and analyzed. Error analysis shows that the absolute errors in output power of the proposed improved model, the traditional model, and the single-machine model relative to the detailed model are 0.1231, 0.3943, and 2.2416, respectively. That is, under random wind disturbance conditions, the accuracy of the proposed improved model is improved by 68.78% and 94.51% compared to the traditional model and the single-machine model, respectively.

[0246] Verification Example 2

[0247] Based on any of the above embodiments, this embodiment uses simulation to verify the effectiveness of the proposed wind-storage frequency coordination control method based on the two-step wind power clustering method:

[0248] Assuming a wind turbine-energy storage battery system, such as Figure 10 As shown, the total installed capacity of the wind farm is 75MW, of which G1 and G2 are both thermal power units with a rated capacity of 100MW each; the energy storage system has an installed capacity of 4MW and a rated power of 0.8MW. The equivalent wind speeds of the four wind turbine groups are as follows: Figure 11 As shown. load1, load2, load3, and load4 are the loads of the system, which are 117MW, 37.5MW, 26.5MW, and 95MW respectively. load5 is set as a 25MW disturbance load that is connected at 50s to verify the effectiveness of the proposed strategy.

[0249] (1) Comparison of system frequency responses under different control strategies:

[0250] Taking a sudden load increase as an example, the control strategy is shown in Table 8, and the simulation results are as follows: Figure 12 As shown in Table 9, the frequency change rate RoCoF is defined as the average rate of change from the start of the frequency drop (initial frequency set to the nominal value of 50Hz) to the lowest frequency point.

[0251] Table 8: Control Strategies

[0252]

[0253] Table 9: Comparison of System Frequency Regulation Data under Different Control Strategies

[0254]

[0255] like Figure 12 As shown, the frequency modulation effect varies significantly under different control strategies:

[0256] (1) In Strategy 2, the wind farm uses the control strategy described in this paper to participate in frequency regulation (combining power reserve control and rotor kinetic energy control). Through the complementarity of the control strategy among the turbine groups, the frequency drop rate is slowed down and the water hammer effect is initially compensated, so that the minimum frequency is increased to 49.7918Hz (the minimum frequency deviation is reduced by 43.74% compared with Strategy 1), RoCoF is improved to -0.07516Hz / s (RoCoF is increased by 29.85% compared with Strategy 1), and the steady-state value is significantly improved to 49.9514Hz (the steady-state frequency deviation is reduced by 32.50% compared with Strategy 1). However, since Strategy 2 does not consider the participation of energy storage in frequency regulation, there is a lack of active power output during the wind turbine speed recovery stage, the frequency drops twice, and the frequency regulation curve is not smooth enough.

[0257] (2) In Strategy 3, both wind and storage adopt the control strategy described in this paper (i.e. Figure 3 The composite frequency regulation strategy shown implements a synergistic compensation mechanism for the water hammer effect: on the one hand, the wind farm provides rapid power support through rotor kinetic energy control; on the other hand, the energy storage system precisely fills the power gap. The combined effect of these two mechanisms significantly reduces the impact on grid equipment. This strategy achieves comprehensive improvements in key frequency regulation indicators: the minimum frequency is increased to 49.8069Hz (the minimum frequency deviation is reduced by 47.83% / 7.25% compared to strategy one / two), RoCoF is improved to -0.07479Hz / s (RoCoF is improved by 30.20% / 0.49% compared to strategy one / two), and the steady-state value is increased to 49.9533Hz (the steady-state frequency deviation is reduced by 35.14% / 3.91% compared to strategy one / two). Furthermore, the active power compensation from energy storage during the wind turbine speed recovery phase improves the minimum deviation of the second frequency drop by 16.46%. In summary, compared with other strategies, Strategy 4 organically integrates the advantages of wind and energy storage frequency regulation resources. It not only overcomes the limitations of a single frequency regulation resource, but also makes the effect of wind and energy storage coordinated frequency regulation not limited to the physical output period of energy storage. The overall control optimization of the system has a positive impact on the frequency response across the entire time scale.

[0258] Among them, the wind turbine clusters in Strategy 2 and Strategy 3 are based on Figure 11 The equivalent wind speed distribution diagrams for each turbine group shown are based on... Figure 6Based on the grouping results, an appropriate control strategy is selected to fully utilize its frequency regulation resources to support grid frequency regulation. Specifically, the equivalent speed of the first generator group satisfies 0.99ω. C ≤ω eq ≤1.01ω C It is located in the constant speed region, but the equivalent electromagnetic power does not satisfy P. e_eq >0.9P e_max Therefore, overspeed unloading is adopted to reserve power margin, and virtual inertia support is provided through integrated inertia control; the equivalent speed of the second group of machines satisfies ω B <ω eq <0.99ω C Located in the MPPT region, and with equivalent electromagnetic power satisfying P e_eq >0.55P e_max The third unit employs integrated inertia control for frequency modulation; the equivalent rotational speed of the third unit meets the requirement of 0.99 ω. C ≤ω eq ≤1.01ω C It is located in the constant speed region, and the equivalent electromagnetic power satisfies P e_eq >0.9P e_max The fourth turbine group employs variable pitch and overspeed load reduction to release reserve power for frequency regulation; the equivalent speed of the fourth turbine group does not meet the requirement of 0.99 ω. C ≤ω eq ≤1.01ω C Located in the constant power region, variable pitch control is used to release reserve power for frequency regulation. The selection of this strategy combines the operating characteristics of each turbine group with the grid's frequency regulation requirements, maximizing the wind turbines' participation in frequency regulation. The active power response curves of the four equivalent turbine groups in the improved model under strategy four are shown below. Figure 13 As shown. Total energy storage output ΔP B like As shown.

[0259] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0260] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations 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, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0261] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0262] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0263] It should be noted that any reference signs placed between parentheses in the claims should not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. This application can be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In a unit claim enumerating several means, several of these means may be embodied by the same item of hardware. The use of the words first, second, and third, etc., does not indicate any order. These words can be interpreted as names.

[0264] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0265] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A wind-storage frequency coordinated control method based on a two-step wind power clustering method, characterized in that, Applied to wind-storage systems, the method includes the following steps: S10, the wind turbine units are first clustered using the target wind turbine variables selected based on principal component analysis to obtain the primary wind power cluster. Then, the wind energy utilization rate is introduced to perform a second clustering on the primary wind power cluster to obtain the target wind power cluster result. S20, based on the target wind power grouping results, the wind turbines are grouped, and the equivalent speed and equivalent electromagnetic power in the wind turbine group are calculated using the grouped wind turbines as the unit. S30, based on the equivalent rotational speed and the equivalent electromagnetic power, determine the target control strategy for the doubly fed asynchronous wind turbine generator in each wind turbine group, so as to perform frequency regulation based on the target control strategy corresponding to each wind turbine group; In step S30, the step of determining the target control strategy for the doubly-fed asynchronous wind turbine in each wind turbine group based on the equivalent rotational speed and the equivalent electromagnetic power includes: When the equivalent rotational speed is within a first preset rotational speed range and the equivalent electromagnetic power is greater than a first preset power threshold, the target control strategy is determined to be a comprehensive inertia control strategy. When the equivalent rotational speed is within a first preset rotational speed range and the equivalent electromagnetic power is less than or equal to a first preset power threshold, the target control strategy is determined to be a maximum power point tracking strategy. When the equivalent speed is not within the first preset speed range and the equivalent speed is greater than the first preset speed threshold, the target control strategy is determined to be the variable pitch angle control strategy. When the equivalent speed is not in the first preset speed range, and the equivalent speed is less than or equal to the first preset speed threshold, and the equivalent electromagnetic power is greater than the second preset power threshold, the target control strategy is determined to be the overspeed load reduction-variable pitch angle control strategy. When the equivalent speed is not in the first preset speed range, and the equivalent speed is less than or equal to the first preset speed threshold, and the equivalent electromagnetic power is less than or equal to the second preset power threshold, the target control strategy is determined to be the overspeed load reduction-inertia control strategy. The expression for the wind energy utilization rate is: ； In the formula, Let λ be the wind energy utilization rate, λ be the tip speed ratio, β be the blade pitch angle, and Λ be the intermediate variable for converting the relationship between λ and β.

2. The method as described in claim 1, characterized in that, The integrated inertia control strategy includes: adjusting the virtual inertia control power ΔP1 and the droop control power ΔP2 according to the frequency difference; The variable pitch angle control strategy includes: keeping the rotational speed of the doubly fed asynchronous wind turbine constant and increasing the pitch angle of the wind turbine. The overspeed load reduction-pitch angle control strategy includes an overspeed load reduction strategy and the pitch angle control strategy, wherein the overspeed load reduction strategy includes increasing the rotational speed under the maximum power point tracking strategy; The overspeed load reduction-inertia control strategy includes the overspeed load reduction strategy and the integrated inertia control strategy.

3. The method as described in claim 2, characterized in that, The frequency-modulated active power change ΔP in the integrated inertia control strategy sum Satisfy the following expression: ； In the formula, ΔP1 is the virtual inertia control power, ΔP2 is the droop control power, K1 is the virtual inertia parameter, and K2 is the droop parameter. For frequency difference, This represents the rate of change of frequency.

4. The method as described in claim 3, characterized in that, Both the virtual inertial parameter and the droop parameter are adjusted according to the rate of change of frequency; Among them, the adjusted virtual inertial parameters Satisfy the following expression: ； In the formula, the time constant T is proportional to the output power and frequency modulation exit time under virtual inertial control, and s is the Laplace operator; Among them, the adjusted droop parameter K 2,1 Satisfy the following expression: ； ； In the formula, t is the current simulation time, t re k is the time for the wind turbine to exit frequency regulation. c Parameters used to control the rate of change; The frequency-modulated active power change ΔP of the integrated inertia control strategy after parameter adjustment sum satisfy: 。 5. The method as described in claim 2, characterized in that, In the variable pitch angle control strategy, the wind turbine pitch angle satisfies the following expression: ； In the formula, This is a reference value for the propeller pitch angle. The increment of the propeller pitch angle that needs to be adjusted corresponds to the change in system frequency. f is the pitch angle droop coefficient, and f is the measurement frequency. N This is the rated frequency.

6. The method as described in claim 2, characterized in that, In the overspeed load reduction strategy, the active power P output by the doubly-fed asynchronous wind turbine generator... del Satisfy the following expression: ； In the formula, P MPPT This refers to the active power output by the wind turbine in MPPT mode. The second-best tip speed ratio; C p,de The wind energy utilization coefficient under unloaded mode; Power point tracking factor in load reduction mode; d% is the load reduction rate. This refers to the generator rotor speed. The rotor speed of the wind turbine. air density, The optimal tip speed ratio is given by S, where S is the swept area and r is the blade radius.

7. The method according to any one of claims 2 to 6, characterized in that, The target control strategy also includes: Determine whether the equivalent frequency of the wind turbine group is less than a preset first frequency threshold and greater than a preset second frequency threshold, wherein the preset first frequency threshold is greater than the preset second frequency threshold; If so, execute the integrated inertia control strategy; During the execution of the comprehensive inertia control strategy, when it is determined that the wind turbine speed has recovered, the output power of the energy storage system is increased to compensate for the reduced output power of the wind turbine. The total output ΔP of the energy storage system B satisfy: ； in: ； ； ； ； In the formula, This refers to the rated active power of the wind turbine; The rated power of the energy storage The battery charge / discharge coefficient. To compensate for the virtual inertia coefficient of the secondary drop of the wind turbine in the improved energy storage system, It is the energy storage droop control coefficient corresponding to the failure of the wind turbine to meet the frequency performance requirements after it is removed from frequency regulation. Fan speed rate of change, This is the energy storage droop control coefficient. For frequency difference, t re k is the time for the wind turbine to exit frequency regulation. c Parameters for controlling the rate of change, The virtual inertia coefficient is used to compensate for the secondary drop of the wind turbine in the energy storage system before the improvement.

8. The method as described in claim 1, characterized in that, In S10, the target wind turbine variables include at least one of rotor speed, active power output of wind turbine, pitch angle, d-axis component of stator current, and d-axis component of rotor current.

Images