Primary frequency modulation method for offshore wind power cluster and shared energy storage

Abstract

The invention discloses a primary frequency modulation method for an offshore wind power cluster and shared energy storage, and the method comprises the steps: obtaining a safe operation rotating speed interval of a rotor of an offshore wind turbine, and determining a wind speed division threshold value based on the interval; then, the real-time access wind speed of each draught fan is obtained, the draught fans are divided into a plurality of draught fan sub-clusters, and a virtual reference draught fan is constructed for each sub-cluster; and calculating the relative frequency modulation capability factor of each fan according to the wind speed difference between each fan and the virtual reference fan. And based on the virtual reference fan and the shared energy storage device, establishing a primary frequency modulation mathematical model and an optimization objective function of the wind storage combined system, and solving by using a commercial solver to obtain an optimal active power output reference value of the virtual reference fan and the shared energy storage device. Active output reference values of all the fans are determined and issued, and coordinated primary frequency modulation of the wind power cluster and energy storage is achieved. The method can improve the frequency modulation response speed and precision, guarantee the safe operation of the fan, and enhance the supporting capability of the wind storage system in the primary frequency modulation of the power grid.

Description

Technical Field

[0001] This invention belongs to the field of frequency regulation technology for offshore wind power clusters, specifically relating to a primary frequency regulation method for offshore wind power clusters and shared energy storage. Background Technology

[0002] With the increasing maturity of new energy power generation technologies, offshore wind power is gradually gaining a larger share of the power system due to its advantages such as high operating efficiency, abundant resources, and no need for land occupation. However, the AC frequency of offshore wind farms is decoupled from the receiving-end grid frequency. When the system frequency changes abruptly, the wind turbines cannot actively support the frequency fluctuations of the receiving-end grid, leading to a decrease in the frequency stability of the power grid system. Therefore, exploring the frequency regulation capabilities of offshore wind power is a key research focus for improving the frequency stability of the power system. At the same time, as the scale of offshore wind farms gradually increases, considering the complex operating environment at sea, the rational control of numerous wind turbines is also a challenge for offshore wind power control strategies.

[0003] In the prior art, Chinese patent CN116054207A discloses a wind-storage integrated primary frequency regulation control method based on fuzzy control load shedding rate, relating to the field of wind-storage frequency regulation technology. This invention combines the current wind speed and frequency regulation safety margin as inputs to the fuzzy controller to obtain variable load shedding rates under different wind turbine and grid operating conditions; it combines the preset load shedding rate with wind turbine load shedding control criteria to allocate power for overspeed load shedding and pitch-variable load shedding of the wind turbine; finally, based on the supercapacitor's SOC state, considering grid frequency deviation and wind turbine output status, it determines the charging / discharging coefficient and self-recovery coefficient for backup frequency regulation and capacity replenishment.

[0004] However, this method has the following limitations: First, although it ensures the economic efficiency of wind farm operation and reasonably reserves the power for turbine load reduction through differentiated load reduction, it does not fully consider the impact of turbine load reduction methods, load reduction rates, and wind speeds on the frequency regulation capability of individual turbines within the wind farm. It fails to accurately quantify the frequency regulation capability of a single turbine, making it difficult to meet the frequency regulation control requirements of large-scale offshore wind farms. Second, while adaptive adjustment of the energy storage discharge coefficient based on the state of charge of the energy storage system can ensure optimal energy storage operation, energy storage is only used as a supplementary system to compensate for insufficient wind turbine frequency regulation power. It fails to combine the frequency regulation characteristics of both systems and fully leverage the frequency regulation advantages of the wind-storage system. Furthermore, this method lacks research on wind farm regulation issues, specifically how to ensure that numerous wind turbines fully utilize their frequency regulation capabilities during the frequency regulation process. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a primary frequency regulation method for offshore wind power clusters and shared energy storage.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] This invention provides a primary frequency regulation method for offshore wind power clusters and shared energy storage, comprising the following steps:

[0008] Obtain the safe operating speed range of the rotor of each wind turbine in the offshore wind turbine cluster, and determine the wind speed division threshold based on the safe operating speed range of the rotor;

[0009] The real-time access wind speed of each wind turbine in the offshore wind turbine cluster is obtained, and the offshore wind turbine cluster is divided into multiple wind turbine sub-clusters according to the wind speed division threshold.

[0010] A virtual reference wind turbine is constructed for each of the wind turbine sub-clusters, and the access wind speed of each virtual reference wind turbine is determined;

[0011] The relative frequency regulation capability factor of each wind turbine is calculated based on the access wind speed of each wind turbine in the wind turbine sub-cluster and the access wind speed of the corresponding virtual reference wind turbine.

[0012] Based on the constructed virtual reference wind turbines and shared energy storage devices, a primary frequency regulation mathematical model and optimization objective function for the wind-storage combined system are constructed.

[0013] The primary frequency regulation mathematical model and optimization objective function of the wind-storage combined system are solved using a commercial solver to obtain the optimal active power output reference values ​​of each virtual reference wind turbine and shared energy storage device.

[0014] Based on the optimal active power output reference value of each virtual reference wind turbine and the relative frequency regulation capability factor of each wind turbine, the active power output reference value of each wind turbine in the offshore wind turbine cluster is obtained and sent to each wind turbine.

[0015] Each wind turbine and the shared energy storage device are controlled to perform frequency regulation output according to the corresponding active power output reference value, thus completing the first frequency regulation process of the offshore wind power cluster and shared energy storage.

[0016] Furthermore, the wind speed threshold determined based on the rotor's safe operating speed range is calculated using the following formula:

[0017]

[0018] Where R is the radius of the wind turbine, λ opt For the optimal tip speed ratio, [ω min ,ω max ] represents the safe operating speed range of the wind turbine rotor, ω min For, ω max v1, v3, and v5 are the determined first, second, and third wind speed classification thresholds, respectively.

[0019] Further, obtaining the real-time access wind speed of each wind turbine in the offshore wind turbine cluster, and dividing the offshore wind turbine cluster into multiple wind turbine sub-clusters according to the wind speed division threshold, specifically including:

[0020] When the real-time access wind speed v of the wind turbine satisfies v1 < v < v3, the wind turbine is classified into the first wind turbine sub-cluster Ω1;

[0021] When the real-time access wind speed v of the wind turbine satisfies v3 < v y < v5, the wind turbine is classified into the second wind turbine sub-cluster Ω2;

[0022] When the real-time access wind speed v of the wind turbine satisfies v > v5, the wind turbine is classified into the third wind turbine sub-cluster Ω3;

[0023] Among them, the first wind turbine sub-cluster Ω1 adopts an overspeed power limit reduction strategy, the second wind turbine sub-cluster Ω2 adopts a variable pitch angle power limit reduction strategy, and the third wind turbine sub-cluster Ω3 adopts a combined reduction strategy of overspeed power limit and variable pitch angle power limit.

[0024] Further, constructing a virtual reference wind turbine for each of the wind turbine sub-clusters and determining the access wind speed of each of the virtual reference wind turbines, specifically including:

[0025] Constructing a virtual reference wind turbine for each of the wind turbine sub-clusters, including virtual reference wind turbine x, virtual reference wind turbine y, and virtual reference wind turbine z, corresponding to the first wind turbine sub-cluster Ω1, the second wind turbine sub-cluster Ω2, and the third wind turbine sub-cluster Ω3 respectively;

[0026] Obtaining the real-time access wind speed of each wind turbine in each wind turbine sub-cluster;

[0027] According to the statistical characteristic values of the real-time access wind speeds of all the wind turbines in each wind turbine sub-cluster, determining the access wind speeds of each of the virtual reference wind turbines, including v x 、v y 、v z , where the statistical characteristic values include weighted average or median.

[0028] Further, calculating the relative frequency modulation ability factor of each wind turbine according to the access wind speed of each wind turbine in the wind turbine sub-cluster and the access wind speed of the corresponding virtual reference wind turbine, specifically including:

[0029] Calculating the load reduction rate of each wind turbine in each wind turbine sub-cluster;

[0030] Based on the load reduction rate of each wind turbine in each wind turbine sub-cluster, combined with the access wind speed of the corresponding virtual reference wind turbine, calculating the relative frequency modulation ability factor of each wind turbine.

[0031] Furthermore, the calculation of the load reduction rate of each wind turbine in each wind turbine sub-cluster specifically includes:

[0032] For some fans with wind speeds close to the minimum operating wind speed, the load reduction rate for each fan is set to zero, i.e., d i =0, does not participate in the first frequency modulation process;

[0033] For the first wind turbine subgroup Ω1 and the second wind turbine subgroup Ω2, the load reduction rate of each wind turbine is determined based on the linear relationship between the access wind speed and the load reduction rate of each wind turbine in subgroups Ω1 and Ω2. The formula is as follows:

[0034]

[0035] Where, d max The maximum load reduction rate of the wind turbine generator is given by Ω1, Ω2, and Ω3, which are the sets of wind turbines in the first, second, and third wind turbine sub-clusters, respectively. i Let v1 and v5 be the access wind speed of the i-th wind turbine, respectively, and let P(v1, v5) be the first and third wind speed thresholds. i ) i For fan i at wind speed v i Maximum operating power, P S P(v5) represents the total reserve power of the wind farm, and P(v5) represents the maximum operating power of the wind turbine corresponding to wind speed v5.

[0036] For the third wind turbine sub-cluster, the output of each wind turbine in the third wind turbine sub-cluster Ω3 is adjusted to the output power corresponding to the third wind speed threshold ν5, and the load reduction rate determined therefrom is calculated using the following formula:

[0037]

[0038] Obtain the load reduction rate of each wind turbine in each wind turbine sub-cluster.

[0039] Furthermore, the calculation of the relative frequency regulation capability factor of each wind turbine based on the load reduction rate of each wind turbine in each wind turbine sub-cluster, combined with the access wind speed of the corresponding virtual reference wind turbine, specifically includes:

[0040] For the first wind turbine sub-cluster Ω1, the formula for calculating the relative frequency regulation capability factor of each wind turbine is as follows:

[0041]

[0042] Among them, P im (t) and P xm (t) represents the electromagnetic power of wind turbine i in subgroup 1 and virtual reference wind turbine x, respectively. i0 With P x0These represent the steady-state operating power of wind turbine i and the virtual reference wind turbine x after load reduction, respectively; v x With P x The operating wind speed and rated output power of the virtual reference fan; J i With J x The moments of inertia of fan i and fan x are respectively, ω id ω imin ω xd With ω xmin t1 and t3 represent the steady-state operating speed and minimum speed of fan i and virtual reference fan x, respectively; t1 and t3 correspond to the start and end times of frequency modulation, respectively.

[0043] For the second wind turbine sub-cluster Ω2, the formula for calculating the relative frequency regulation capability factor of each wind turbine is as follows:

[0044]

[0045] Among them, P im (t) and P ym (t) represents the electromagnetic power of wind turbine i and virtual reference wind turbine y in the second wind turbine sub-cluster Ω2, respectively. i0 With P y0 These represent the steady-state operating power of wind turbine i and the virtual reference wind turbine y after load reduction, respectively; v y With P y The operating wind speed and rated output power of the virtual reference fan y; ω imax ω imin ω ymax With ω ymin These represent the maximum safe speed and minimum speed of fan i and virtual reference fan y, respectively;

[0046] For the third wind turbine sub-cluster Ω3, the formula for calculating the relative frequency regulation capability factor of each wind turbine is as follows:

[0047]

[0048] Where, d i With d z These represent the load reduction rates of wind turbine i and virtual reference wind turbine z in the third wind turbine sub-cluster Ω3, respectively; P(v z ) z For the virtual reference fan z at the operating wind speed v z The maximum output power at that point.

[0049] Furthermore, the primary frequency regulation mathematical model of the wind-storage combined system is as follows:

[0050]

[0051] u(k)=[ΔP x (k) ΔPy (k) ΔP z (k) g(k)] T

[0052] y(k)=[ΔP g (k),ΔP bess (k),∑ΔP wind (k),S(k),Δf(k)] T

[0053] r(k)=P load (k)

[0054] Among them, P V (k),P m (k) and ΔP g (k) represents the speed-regulating power, mechanical power, and electromagnetic power of a traditional thermal power unit at time k; ΔP bess S(k) and S(k) are the energy storage power regulation and state of charge at time k; and These represent the power adjustment values ​​of the three virtual reference wind turbines at time k, ∑ΔP wind (k) represents the total regulating power of the wind farm; Δf(k) represents the frequency change at time k, and ΔP x (k), ΔP y (k) and ΔP z g(k) represents the power regulation reference value of the three virtual reference wind turbines at time k; g(k) represents the reference value of the wind-storage coordination coefficient at time k.

[0055] Furthermore, the optimization objective function is:

[0056]

[0057] Where C is the optimization objective function, C β (k+i|k) is the prediction expression for the frequency regulation cost of the wind turbine pitch control at the current sampling time k to time k+i, C ωre (k+i|k) is the prediction expression for the frequency regulation cost of the wind turbine speed change at the current sampling time k to time k+i, C bess (k+i|k) is the prediction expression for the energy storage frequency regulation cost at the current sampling time k to the time k+i;

[0058] Among them, the cost C of the wind turbine pitch control process β The formula is:

[0059]

[0060] Where, Δβ z(k+i|k) is the prediction expression for the change in the z-pitch angle of the virtual reference wind turbine at the current sampling time k to the time k+i, β zd denoted as the steady-state pitch angle of the virtual reference wind turbine z after load reduction; c is the wear coefficient of the wind turbine pitch angle; and a1 is the cost coefficient of wind turbine pitch frequency regulation.

[0061] The cost C of the fan rotor adjustment process ωre The formula is:

[0062]

[0063] Where, Δω x (k+i|k) and Δω y (k+i|k) is the prediction expression for the adjustment of the virtual reference fan's x and y speeds at the current sampling time k to time k+i; ω xf With ω yf Let x be the critical speed of the virtual reference fan and y be y, and let a2 be the cost coefficient for speed regulation.

[0064] Furthermore, obtaining the active power output reference value of each wind turbine in the offshore wind turbine cluster based on the optimal active power output reference value of each virtual reference wind turbine and the relative frequency regulation capability factor of each wind turbine specifically includes:

[0065] For the first wind turbine sub-cluster Ω1, the reference value for the active power output of each wind turbine is:

[0066]

[0067] in, Let x be the reference value for the active power output of the i-th wind turbine. i Let ΔP be the relative frequency regulation capability factor of the i-th wind turbine. x (k) represents the optimal active power output reference value of the virtual reference wind turbine x at time k;

[0068] For the second wind turbine sub-cluster Ω2, the reference value for the active power output of each wind turbine is:

[0069]

[0070] Among them, y i Let ΔP be the relative frequency regulation capability factor of the i-th wind turbine. y (k) represents the optimal active power output reference value of the virtual reference wind turbine y at time k;

[0071] For the third wind turbine sub-cluster Ω3, the reference value for the active power output of each wind turbine is:

[0072]

[0073] Among them, zi Let ΔP be the relative frequency regulation capability factor of the i-th wind turbine. z (k) represents the optimal active power output reference value of the virtual reference wind turbine z at time k.

[0074] Compared with the prior art, the present invention has the following advantages:

[0075] (1) Considering the differences in the operating conditions of individual wind turbines under the wake effect, this invention first sets a suitable wind turbine load reduction strategy based on wind speed in order to reasonably reserve the load reduction power of the wind turbines. Due to the different load reduction methods, the active reserve energy of the wind turbines is in different forms: the active resource form of overspeed load reduction is the kinetic energy that the rotor can release or absorb, and the active resource form of pitch load reduction is the mechanical energy during the pitch angle change process. Therefore, the different active components of the wind turbines under overspeed load reduction, pitch load reduction, and overspeed-pitch load reduction methods were analyzed respectively. Under the three load reduction strategies, the frequency regulation capability of individual wind turbines was quantified by comprehensively considering the wind turbine load reduction rate, operating wind speed, and active components, and a method that can accurately evaluate the frequency regulation capability of wind farms was defined.

[0076] (2) To address the challenge of regulating a large number of wind turbines operating under diverse conditions within a large offshore wind farm, the proposed method divides the wind farm into three groups based on load shedding. A virtual reference turbine is introduced into each group, and its regulated power is used as a control variable. This reduces the number of optimization variables for the controller compared to single-turbine regulation. Next, the ratio of the frequency regulation capability between the wind farm turbines and the virtual reference turbine is defined as the relative frequency regulation capability factor, establishing a connection between the two. This allows for offline calculation to determine the optimal power regulation amount for the actual wind farm turbines. By reducing the high-dimensional optimization space of the controller, the method ensures that each wind farm turbine can fully utilize its frequency regulation capability, solves the problem of primary frequency regulation modeling in large offshore wind farms, and improves both the efficiency and effectiveness of frequency regulation.

[0077] (3) In the mathematical modeling process, this invention fully considers the frequency regulation advantages of energy storage system and wind farm. Considering that the energy storage charging and discharging rate is relatively fast, it can suppress the frequency fluctuation in the early stage of load change in a timely manner. A small time response constant is set for it. However, considering that the frequency regulation cost of energy storage is relatively high, a large weight coefficient is set for energy storage cost in the objective function. Similarly, the frequency regulation speed of wind turbine is relatively slow and the frequency regulation cost is relatively low. This embodiment sets a large time constant and a small cost coefficient, so that the wind storage system can give full play to its frequency regulation advantages and reduce the cost of frequency regulation process. Attached Figure Description

[0078] Figure 1 This is a flowchart of the frequency modulation method according to an embodiment of the present invention;

[0079] Figure 2This is a flowchart illustrating the frequency response of a wind-storage integrated system based on the MPC algorithm, according to an embodiment of the present invention.

[0080] Figure 3 This is a graph showing the variation of each power component in an embodiment of the present invention;

[0081] Figure 4 This is a graph showing the power variation of each component in an embodiment of the present invention;

[0082] Figure 5 This is a schematic diagram of the primary frequency regulation model of the wind-storage combined system according to an embodiment of the present invention;

[0083] Figure 6 This is a topology diagram of an offshore wind farm and energy storage system according to an embodiment of the present invention;

[0084] Figure 7 This is a schematic diagram illustrating load changes according to an embodiment of the present invention;

[0085] Figure 8 This is a comparison diagram of system frequency changes under control strategies 1 and 2 in this embodiment of the invention;

[0086] Figure 9 This is a comparison diagram of the wind turbine operating characteristics under strategy 1 and strategy 2 in this embodiment of the invention;

[0087] Figure 10 This is a graph showing the variation of energy storage and wind turbine output power curves according to an embodiment of the present invention;

[0088] Figure 11 This is a graph showing the change in SOC of energy storage according to an embodiment of the present invention;

[0089] Figure 12 Frequency variation diagrams under different strategies in embodiments of the present invention;

[0090] Figure 13 This is a graph showing the change in fan speed under different strategies in an embodiment of the present invention. Detailed Implementation

[0091] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0092] Example 1:

[0093] This embodiment provides a primary frequency regulation method for offshore wind power clusters and shared energy storage, such as Figure 1 , Figure 2As shown in the figure, it includes the following steps:

[0094] Step S1: Obtain the rotor safe operating speed range of each wind turbine in the offshore wind turbine cluster, and determine the wind speed division threshold based on the rotor safe operating speed range, specifically including:

[0095]

[0096] where R is the radius of the wind turbine, λ opt is the optimal tip speed ratio, [ω min , ω max is the rotor safe operating speed range of the wind turbine, ω min is, ω max is, v1, v3, and v5 are the determined first wind speed division threshold, second wind speed division threshold, and third wind speed division threshold respectively. According to research, the wake effect in offshore wind farms is significant. The wake can extend more than 70 km, and the average amplitude of the wind speed decay of downstream wind turbines can reach 12.17%, and the highest decay amplitude can reach 19.3%. Therefore, the determination of the above thresholds fully considers the wind speed differences caused by the wake.

[0097] Step S2: Obtain the real-time access wind speed of each wind turbine in the offshore wind turbine cluster, and divide the offshore wind turbine cluster into multiple wind turbine sub-clusters according to the wind speed division threshold, specifically including:

[0098] When the real-time access wind speed v of the wind turbine satisfies v1 < v < v3, the wind turbine is classified into the first wind turbine sub-cluster Ω1;

[0099] When the real-time access wind speed v of the wind turbine satisfies v3 < v y < v5, the wind turbine is classified into the second wind turbine sub-cluster Ω2;

[0100] When the real-time access wind speed v of the wind turbine satisfies v > v5, the wind turbine is classified into the third wind turbine sub-cluster Ω3;

[0101] Among them, the first wind turbine sub-cluster Ω1 adopts an overspeed power limit load reduction strategy and preferentially releases rotor kinetic energy to participate in frequency modulation; the second wind turbine sub-cluster Ω2 adopts a variable pitch angle power limit load reduction strategy to achieve frequency modulation by adjusting the pitch angle; the third wind turbine sub-cluster Ω3 adopts a combined load reduction strategy of overspeed power limit and variable pitch angle power limit, and comprehensively uses rotor kinetic energy and pitch adjustment to provide primary frequency modulation reserve.

[0102] Considering the wake effect and wind condition variations in actual offshore wind farms, the frequency regulation capabilities of wind turbines vary significantly. It is crucial to ensure that regulation commands match the turbines' frequency regulation capabilities to avoid overload and wasted frequency regulation resources. Treating the output of all turbines directly as independent decision variables can easily exacerbate frequency response lag due to communication delays at sea. Therefore, this embodiment adopts a hierarchical control strategy: within each turbine sub-cluster, a virtual reference turbine (x, y, z) is designated as the dominant unit, directly participating in the controller's optimization of regulation commands; other turbines adjust their output based on their relative frequency regulation capability factors with the corresponding virtual reference turbine, achieving overall coordinated control. This method reduces the number of independent control variables for all units in the original wind farm to three dominant control variables, significantly improving the real-time performance and coordination of frequency regulation control.

[0103] Step S3: Construct a virtual reference wind turbine for each wind turbine sub-cluster and determine the access wind speed for each virtual reference wind turbine, specifically including:

[0104] A virtual reference wind turbine is constructed for each wind turbine sub-cluster, including virtual reference wind turbine x, virtual reference wind turbine y, and virtual reference wind turbine z, which correspond to the first wind turbine sub-cluster Ω1, the second wind turbine sub-cluster Ω2, and the third wind turbine sub-cluster Ω3, respectively.

[0105] Obtain the real-time access wind speed of each wind turbine within each wind turbine sub-cluster;

[0106] Based on the statistical characteristics of the real-time access wind speed of all wind turbines within each wind turbine sub-cluster, the access wind speed of each virtual reference wind turbine is determined, including v. x v y v z The statistical characteristics include the weighted average or median.

[0107] Step S4: Based on the access wind speed of each wind turbine in the wind turbine sub-cluster and the access wind speed of the corresponding virtual reference wind turbine, calculate the relative frequency regulation capability factor of each wind turbine, specifically including:

[0108] Calculate the load reduction rate of each wind turbine in each wind turbine sub-cluster, specifically including:

[0109] For the first wind turbine sub-cluster Ω1, the load reduction rate of each wind turbine in the first wind turbine sub-cluster Ω1 is set to zero d. i =0, where d i Let be the load reduction rate of the i-th fan. Since the wind speed is lower than the first wind speed threshold, the fan is in poor operating condition. If the load reduction continues, the fan will stop. Therefore, the load reduction rate of each fan in this sub-cluster is set to zero.

[0110] For the second wind turbine sub-cluster Ω2, the load reduction rate of each wind turbine is determined based on the linear relationship between the access wind speed and the load reduction rate of each wind turbine in the second wind turbine sub-cluster Ω2. The formula is as follows:

[0111]

[0112] And through total power balance constraints:

[0113]

[0114] Where, d max Ω1, Ω2, and Ω3 represent the turbine sets of the first, second, and third turbine sub-clusters, respectively. i Let ν1 and ν5 be the access wind speed of the i-th wind turbine, respectively, and let P(ν1, ν5) be the first and third wind speed thresholds. i ) i For fan i at wind speed ν i Maximum operating power, P S P(v5) represents the total reserve power of the wind farm, and P(v5) represents the maximum operating power of the wind turbine corresponding to wind speed v5.

[0115] For the third wind turbine sub-cluster, the output of each turbine in the third wind turbine sub-cluster is adjusted to the output power corresponding to the third wind speed threshold v5, and the load reduction rate determined therefrom is calculated using the following formula:

[0116]

[0117] Obtain the load reduction rate of each wind turbine in each wind turbine sub-cluster. When the wind speed exceeds the third wind speed threshold, the wind turbine is close to its rated power or is in an over-generation state. At this time, the output of each wind turbine needs to be adjusted to the output power corresponding to the wind speed. Based on the load reduction rate and combined with the operating wind speed of the virtual reference wind turbine, calculate the relative frequency regulation capability factor of each wind turbine: By comparing the load reduction potential and kinetic energy release capability of a single wind turbine with its virtual reference wind turbine, the relative frequency regulation capability factor is obtained and used for subsequent frequency regulation power allocation so that the frequency regulation power undertaken by each wind turbine matches its own capability.

[0118] Based on the load reduction rate of each wind turbine in each wind turbine sub-cluster, and combined with the access wind speed of the corresponding virtual reference wind turbine, the relative frequency regulation capability factor of each wind turbine is calculated, specifically including:

[0119] For the first wind turbine sub-cluster Ω1, during the process of adjusting the active power of the wind turbines to support the grid frequency, the sources of power change include: rotor kinetic energy component and wind turbine load shedding reserve component, such as... Figure 3 As shown in the diagram. Here, t1 is the start time of frequency regulation, t2 is the end time of rotor kinetic energy release, and t3 is the steady-state time of frequency regulation. During the change in supporting power, the rotor kinetic energy response is fast but the supporting process is short; the supporting process of the wind turbine releasing reserve power is longer. Therefore, the frequency regulation capability x of the wind turbine in this unloaded state... i It is determined jointly by these two parts:

[0120] Although the kinetic energy stored in the rotor can respond quickly to a decrease in frequency, the rotor speed of the wind turbine is limited by safety constraints. imin Due to the constraints, if the rotor speed of multiple wind turbines in a wind power cluster drops below the safe speed, it will cause a sharp drop in the overall power generation of the wind farm. Therefore, this embodiment will use ω imin This serves as the minimum threshold for the kinetic energy release capability of a quantified wind turbine rotor. The formula for calculating the relative frequency regulation capability factor of each wind turbine is as follows:

[0121]

[0122]

[0123] Among them, P im (t) and p xm (t) represents the electromagnetic power of wind turbine i in subgroup 1 and virtual reference wind turbine x, respectively. i0 With P x0 These represent the steady-state operating power of wind turbine i and the virtual reference wind turbine x after load reduction, respectively; v x With P x The operating wind speed and rated output power of the virtual reference fan; J i With J x The moments of inertia of fan i and fan x are respectively, ω id ω imin ω xd With ω xmin t1 and t3 represent the steady-state operating speed and minimum speed of fan i and virtual reference fan x, respectively; t1 and t3 correspond to the start and end times of frequency modulation, respectively.

[0124] For the second wind turbine sub-cluster Ω2, in the medium wind speed range, there is still some room for load reduction in rotational speed, and the steady-state rotational speed after load reduction is generally close to the maximum rotational speed. The load reduction capacity at this point is determined by three parts: the rotor kinetic energy component, the wind turbine output power component, and the pitch angle mechanical power component, such as... Figure 4 As shown, under this load reduction strategy, the pitch angle and the wind energy captured by the turbine together cause changes in the turbine's mechanical power. Therefore, the relative frequency regulation capability factor y... si It can be represented as:

[0125]

[0126] Since this embodiment only considers the maximum frequency regulation potential of the wind turbine, the above formula is simplified to obtain the formula for calculating the relative frequency regulation capability factor of each wind turbine:

[0127]

[0128] Among them, P im (t) and Pym (t) represents the electromagnetic power of wind turbine i and virtual reference wind turbine y in the second wind turbine sub-cluster Ω2, respectively. i0 With P y0 These represent the steady-state operating power of wind turbine i and the virtual reference wind turbine y after load reduction, respectively; ν y With P y The operating wind speed and rated output power of the virtual reference fan y; ω imax ω imin ω ymax With ω ymin These represent the maximum safe speed and minimum speed of fan i and virtual reference fan y, respectively;

[0129] For the third wind turbine sub-cluster Ω3, compared to overspeed load reduction, the variable pitch angle control method has a wider range of applications. However, the pitch angle change process causes wear on mechanical components, and the adjustment speed is relatively slow. Under this load reduction strategy, the wind turbine frequency regulation capability is determined only by the load reduction rate, therefore the relative frequency regulation capability factor z... i The expression:

[0130]

[0131] Where, d i With d z These represent the load reduction rates of wind turbine i and virtual reference wind turbine z in the third wind turbine sub-cluster Ω3, respectively; P(v z ) z For the virtual reference fan z at the operating wind speed v z Maximum output power;

[0132] d i With d z The load shedding rates of the wind turbines in the wind farm and the virtual reference wind turbine are respectively, and the initial pitch angle β of turbine i is given. id The formula is:

[0133] v i C P (λ opt ,β id =v5C P (λ opt ,0)

[0134] At the same time, since the pitch angle varies over a large range, the maximum value of the pitch angle should be constrained:

[0135]

[0136] Among them, P imin β is the minimum limit value of the output power of fan i. imin With β imax This defines the limit of the wind turbine pitch angle adjustment range.

[0137] Step S5: Based on the constructed virtual reference wind turbines and shared energy storage devices, construct the primary frequency regulation mathematical model and optimization objective function of the wind-storage joint system;

[0138] The primary frequency regulation mathematical model of a wind-storage integrated system, such as Figure 5 As shown, in this embodiment, the AC power grid is considered as a traditional synchronous generator set with primary frequency regulation capability. Let H be the inertial response constant of the traditional power grid. Since offshore wind power clusters do not have frequency regulation capability, when the wind power penetration rate connected to the grid is m, let the grid response constant be H. sea ,but:

[0139]

[0140] Among them, S G S bess With S wind These are the capacities of thermal power units, energy storage, and wind turbine clusters, respectively. According to... Figure 5 A mathematical model for the primary frequency modulation of the entire system can be established:

[0141]

[0142] The standard form of the discrete equations based on model predictive control algorithms can be expressed as:

[0143]

[0144] u(k)=[ΔP x (k) ΔP y (k) ΔP z (k) g(k)] T

[0145] y(k)=[ΔP g (k),ΔP bess (k),∑ΔP wind (k),S(k),Δf(k)] T

[0146] r(k)=P load (k)

[0147] Among them, P V (k),P m (k) and ΔP g (k) represents the speed-regulating power, mechanical power, and electromagnetic power of a traditional thermal power unit at time k; ΔP bess S(k) and S(k) are the energy storage power regulation and state of charge at time k; and These represent the power adjustment values ​​of the three virtual reference wind turbines at time k, ∑ΔP wind(k) represents the total regulating power of the wind farm; Δf(k) represents the frequency change at time k, and ΔP x (k), ΔP y (k) and ΔP z g(k) represents the power regulation reference value of the three virtual reference wind turbines at time k; g(k) represents the reference value of the wind-storage coordination coefficient at time k.

[0148] Within the prediction time domain, considering the combined regulation of wind turbines and energy storage, this embodiment primarily focuses on achieving optimal frequency regulation while minimizing process losses, thus optimizing the system's control variables. Therefore, the objective function is considered from the following three aspects, and the optimized objective function is as follows:

[0149]

[0150] Where C is the optimization objective function, C β (k+i|k) is the prediction expression for the frequency regulation cost of the wind turbine pitch control at the current sampling time k to time k+i, C ωre (k+i|k) is the prediction expression for the frequency regulation cost of the wind turbine speed change at the current sampling time k to time k+i, C bess (k+i|k) is the prediction expression for the energy storage frequency regulation cost at the current sampling time k to the time k+i;

[0151] Among them, the cost C of the wind turbine pitch control process β The formula is:

[0152]

[0153] Where, Δβ z (k+i|k) is the prediction expression for the change in the z-pitch angle of the virtual reference wind turbine at the current sampling time k to the time k+i, β zd Let c be the steady-state pitch angle of the virtual reference wind turbine z after load reduction; c is the wind turbine pitch angle wear coefficient; and a1 is the cost coefficient for wind turbine pitch control frequency regulation. When adjusting the wind turbine output power through pitch control, additional mechanical torque is required to change the wind turbine pitch angle, thus incurring wear costs. Let c be the wind turbine wear coefficient, and Δβ be the pitch angle change. z The cost of unit wear is described by a quadratic function of (k+i|k); and, wind curtailment will occur during the pitch angle change. Let the wind curtailment cost coefficient be a1, and the power loss cost is described by a quadratic function of the pitch angle adjustment degree.

[0154] During the process of the wind turbine rotor releasing kinetic energy, while meeting the frequency regulation requirements of the wind turbine, it is also necessary to ensure that the power deficit required for energy storage replenishment when the wind turbine rotor speed recovers is minimized. A weighting coefficient a2 is set to represent the cost C of the wind turbine rotor adjustment process. ωre The formula is:

[0155]

[0156] Where, Δω x (k+i|k) and Δω y (k+i|k) is the prediction expression for the adjustment of the virtual reference fan's x and y speeds at the current sampling time k to time k+i; ω xf With ω yf Let x be the critical speed of the virtual reference fan and y be y, and let a2 be the cost coefficient for speed regulation.

[0157] During energy storage operation, high-power operation will reduce the lifespan of the energy storage; at the same time, a large range of state of charge deviations will also cause some damage to the battery.

[0158]

[0159] Where a3 and a4 are the cost coefficients for energy storage power charging and discharging and state of charge offset, respectively.

[29] S ref This is the reference value for the state of charge of energy storage, and is taken as a constant.

[0160] The constraints of the primary frequency regulation mathematical model of the wind-storage integrated system include:

[0161] Wind turbine power variation constraints: Since wind turbines participate in the system frequency regulation process by reserving active power, there is an upper limit constraint on the output power of the wind turbines. At the same time, the operating power of the wind turbines cannot be too low.

[0162] P min ≤P x0 +ΔP x (k+i|k)≤P max

[0163] Among them, P min This is the minimum operating power of the wind turbine, typically a fixed value; P x0 P is the steady-state power of the virtual reference wind turbine x. max The maximum output power of the wind turbine is related to the operating wind speed. Similarly, the power constraints for the virtual reference wind turbine at y and z are:

[0164]

[0165] Constraints on pitch angle variation: The pitch angle is controlled by a hydraulic servo system, therefore the minimum pitch angle is generally 0. When the virtual reference wind turbine y undergoes pitch adjustment and load reduction, the constraint on its variation is as follows:

[0166] -(β zd -β min )≤Δβ z (k+i|k)≤(βmax -β zd )

[0167] Where: β min With β max These represent the minimum and maximum pitch angles.

[0168] Constraints on energy storage operation: During the coordinated adjustment of wind turbines and energy storage, the output power of energy storage is related to the operating status of the wind turbines. Due to the small capacity of energy storage and the high operating cost of energy storage, the wind-storage coordination coefficient is subject to the following constraints:

[0169] 0 < g(k+i|k) < 0.8

[0170] Energy storage state of charge constraints:

[0171] S min ≤S(k+i|k)≤S max

[0172] In the formula: S min S is the limit value under charged state. max This represents the upper limit of the state of charge.

[0173] Fan speed constraint: To ensure safe and stable operation of the fan, the fan speed is constrained.

[0174]

[0175] Step S6: Use a commercial solver to solve the primary frequency regulation mathematical model and optimization objective function of the wind-storage combined system to obtain the optimal active power output reference values ​​of each virtual reference wind turbine and shared energy storage device.

[0176] like Figure 2 The diagram shows the flow of the control strategy proposed in this embodiment. The prediction model is solved using the quadratic programming solver built into MATLAB. The control variables in the result include a control sequence [Δu(k), Δu(k+1), ..., Δu(k+N)] over a period of time. P )], where N P For the control time domain, based on the characteristics of the model predictive control algorithm, only the first parameter Δu(k) in the control sequence is taken, which is ΔP. x (k), ΔP y (k), ΔP z g(k) and g(k) are used as reference values ​​for the optimal adjustment of the three virtual reference wind turbines and the energy storage system at time k. They are then substituted back into the prediction equation to update the state variables, thus completing the optimization solution for one time.

[0177] Step S7: Based on the optimal active power output reference value of each virtual reference wind turbine and the relative frequency regulation capability factor of each wind turbine, obtain the active power output reference value of each wind turbine in the offshore wind turbine cluster, and distribute it to each wind turbine. Specifically, this includes:

[0178] For the first wind turbine sub-cluster Ω1, the reference value for the active power output of each wind turbine is:

[0179]

[0180] in, Let x be the reference value for the active power output of the i-th wind turbine. i Let ΔP be the relative frequency regulation capability factor of the i-th wind turbine. x (k) represents the optimal active power output reference value of the virtual reference wind turbine x at time k;

[0181] For the second wind turbine sub-cluster Ω2, the reference value for the active power output of each wind turbine is:

[0182]

[0183] Among them, y i Let ΔP be the relative frequency regulation capability factor of the i-th wind turbine. y (k) represents the optimal active power output reference value of the virtual reference wind turbine y at time k;

[0184] For the third wind turbine sub-cluster Ω3, the reference value for the active power output of each wind turbine is:

[0185]

[0186] Among them, z i Let ΔP be the relative frequency regulation capability factor of the i-th wind turbine. z (k) represents the optimal active power output reference value of the virtual reference wind turbine z at time k.

[0187] Step S8: Control each wind turbine and shared energy storage device to perform frequency regulation output according to the corresponding active power output reference value, and complete the first frequency regulation process of offshore wind power cluster and shared energy storage.

[0188] Example 2:

[0189] This embodiment uses the Matlab simulation platform to build a simulation model based on actual data from the Rudong wind farm, verifying the effectiveness of the proposed strategy. Figure 6 As shown, this embodiment includes three offshore wind farms. Wind farms 1 and 3 have 60 wind turbines arranged in 10 rows and 6 columns, each with a rated power of 5MW and a rated wind speed of v. d=12.5m / s; Wind farm 2 has 50 wind turbines evenly distributed in 10 rows and 5 columns, each with a rated capacity of 4MW and a rated wind speed of v. d =10.5 m / s; Configure a shared energy storage power station based on lithium battery technology, with an energy storage capacity of 10.5 MW·h; The steady-state output power of the two synchronous generators are 250 MW and 200 MW respectively. The parameter settings are shown in Table 1:

[0190] Table 1

[0191]

[0192]

[0193] 1. Verification of the effectiveness of frequency modulation strategy based on MPC control

[0194] With the large-scale integration of new energy sources into the grid and the increasing diversification of load types, loads can generate continuous small disturbances under any circumstances, which can significantly impact system stability. To verify that the wind turbine strategy in this embodiment can effectively address this problem, this embodiment compares and analyzes strategies (1) and (2): (1) the strategy of this invention; (2) the wind turbines in the wind farm adopt the same load reduction instruction (d% = 5.5%). Under a certain wind speed, random loads are generated as follows... Figure 7 The range shown is the load change sequence between -20 and 20MW, and it is set to change once every 50 seconds.

[0195] At a certain moment, assuming that the cut-in wind speed of the first row of wind turbines in the three wind farms is the same under ideal conditions, and setting v0 = 13.8 m / s, the distribution of the initial wind speed of each wind farm is calculated by the wake.

[0196] As can be seen from the wind speed distribution, wind turbines in the same row within a wind farm experience the same wind speed. Since wind speed is a key indicator for classifying wind turbine load reduction strategies in this embodiment, wind turbines in the same row can be considered to operate under the same conditions. Simultaneously, due to the wake effect, the 8th row wind turbines in wind farms 1 and 3, and the 6th row wind turbine in wind farm 2, are already below the minimum operating wind speed, and therefore, these turbines no longer have frequency regulation capabilities. Next, the wind speed ranges are divided: wind turbines in rows 4-7 of wind farms 1 and 3, and wind turbines in rows 4-6 of wind farm 2, employ load reduction method 1; wind turbines in rows 2 and 3 of wind farms 1-3 employ load reduction method 2; and wind turbines in row 1 of wind farms 1-3 employ load reduction method 3. In summary, the number of wind turbines under different load reduction strategies can be determined:

[0197] When the load fluctuates continuously, the frequency variation graph is as follows: Figure 8 As shown, the power and rotational speed of offshore wind power clusters are compared, for example... Figure 9 As shown.

[0198] pass Figure 8 It can be seen that the strategy proposed in this embodiment is superior to strategy 2 in controlling frequency fluctuations during continuous load fluctuations. Strategy 2 achieves a maximum frequency fluctuation of 0.23Hz, while strategy 1 ensures a maximum frequency fluctuation of 0.17Hz. From the overall frequency regulation effect, it can be seen that the final frequency deviation of the strategy in this embodiment is between -0.083 and 0.064Hz, while strategy 2 has a slower adjustment speed and maintains a final frequency deviation between -0.113 and 0.0873Hz. Within 0–250s, when load fluctuations are relatively small, both strategies 1 and 2 can effectively regulate frequency fluctuations. When the system experiences large load fluctuations, i.e., within 250–300s, strategy 1 can maintain a stable final frequency deviation of -0.083Hz, while strategy 2, due to insufficient frequency regulation capability of the wind farm, can only adjust the frequency deviation to -0.113Hz. This is because when Strategy 2 uses the same load reduction command, the strong frequency regulation capabilities of wind turbines in each wind speed range are not utilized, while turbines operating close to shutdown are not considered for protection. Although reserve power is guaranteed, the frequency regulation capabilities of the wind farm are not fully utilized, leading to significant frequency fluctuations in the system. This verifies that the strategy in this embodiment has a better frequency regulation effect and improves the stability of the system frequency under random load disturbances.

[0199] To more intuitively highlight the advantages of the strategy in this embodiment, the operating characteristic curve of the wind turbine during the frequency regulation period was selected, such as... Figure 9 As shown.

[0200] 2. Analysis of internal power optimization and allocation in the wind-storage system

[0201] This embodiment not only considers improving frequency fluctuations, but also the economical allocation of frequency support for wind turbines and energy storage. The output power variation process of energy storage and wind turbines is as follows: Figure 10 As shown, the energy storage SOC state change process is as follows: Figure 11 As shown.

[0202] Depend on Figure 10 As shown, within 50–100s, 300–400s, and 450–500s, due to small load changes, the frequency regulation requirements can be met solely by the thermal power units, and the wind and energy storage system does not generate power. Within 150–300s and 400–450s, when load fluctuations are larger, the wind and energy storage system responds quickly to frequency fluctuations. Furthermore, when the load fluctuates significantly, the wind and energy storage system can rapidly increase active power. However, as the thermal power units gradually replenish the power, the overall output of the wind and energy storage system begins to decline and stabilize. After the frequency stabilizes for a period, the control system restarts the fan speed recovery. After the fan power is set to a certain downward value, the energy storage system begins to replenish power to support the fan speed recovery, while maintaining the total output of the wind and energy storage system without significant fluctuations. Figure 10It can be seen that, overall, under the control strategy proposed in this embodiment, the state of charge value of the energy storage remains within a good range despite continuous frequency fluctuations.

[0203] Specifically, taking a simulation period of 150-200s, the response time constant of the energy storage converter is smaller than that of the wind turbine pitch angle and speed control, thus the energy storage response is faster. Under the influence of the energy storage frequency regulation cost constant, the state of charge offset cost, and the wind turbine frequency regulation cost constant, the wind turbine frequency regulation cost is lower. As the total output of wind and energy storage tends to stabilize, the wind turbine regulation power will gradually increase, while the energy storage output will begin to decrease. Therefore, it is verified that the strategy proposed in this embodiment can optimize the coordination between wind and energy storage.

[0204] 3. Validation of the relative frequency modulation capability difference coefficient

[0205] Assuming an initial wind speed v0 = 13.8 m / s, this embodiment will simulate and verify the following three different wind turbine control strategies: (1) the strategy of this embodiment; (2) the frequency regulation capability of the wind farm under the overspeed load reduction strategy; and (3) without considering the difference in wind turbine frequency regulation capability. Assuming that the load is stable for the first 140 seconds of system operation, a sudden increase of 40 MW in load is set, and the frequency simulation results are as follows. Figure 12 As shown, the speed change process is as follows: Figure 13 As shown.

[0206] Table 2

[0207]

[0208] As shown in Table 2, the strategy proposed in this embodiment has the best effect in regulating frequency fluctuations. Although Strategy 2 improves the frequency regulation deviation, the rotor speed of the wind turbines in the wind farm at medium and high wind speeds is relatively high, resulting in a smaller frequency regulation capability for these turbines. Furthermore, when quantifying the frequency regulation capability of low-wind-speed turbines, only the rotor kinetic energy component is considered, leading to a weaker overall frequency regulation capability of the wind farm. The frequency is adjusted to 49.91Hz at 11 seconds after the disturbance occurs, which is a larger frequency deviation than the strategy in this embodiment. Since the high-wind-speed turbines in this embodiment reserve frequency regulation capability through pitch reduction and consider the diversity of active power resources of medium and low-wind-speed turbines, the frequency regulation capability of turbines under various operating wind conditions is accurately quantified. Therefore, the frequency is stabilized to 49.957Hz at 7 seconds. Although Strategy 3 can regulate the frequency to a certain extent, it fails to consider the differences in frequency regulation capability of individual turbines. During the regulation process, some turbines operate with limited power and undergo protective shutdowns, resulting in long-term frequency fluctuations, and the frequency stabilizes down by 0.067Hz.

[0209] Take the fan speeds of strategies 1 and 2 under the same low wind speed, such as Figure 13(a) Strategy 2 failed to comprehensively consider the role of active power components, resulting in a significant increase in the wind turbine speed to absorb energy, causing the wind turbine speed to rise to 0.977 pu, which poses a risk of the wind turbine speed exceeding the safe speed and shutting down. It also verifies that the active power resource utilization rate under the strategy of this embodiment is greater than that under strategy 2. At the same time, the rotor of strategy 2 drops rapidly in the first 3 seconds, while the strategy of this embodiment takes into account the change of the active power component of the wind turbine output during this process, and the speed change is relatively stable.

[0210] Selecting strategy 1 and strategy 3, both of which involve the speed change curves of high-speed wind turbines, such as... Figure 13 (b) A wind turbine shutdown occurred. Due to receiving the same regulation command, some wind turbines operating at higher wind speeds experienced speeds exceeding their maximum speed limits and were thus protected and forced to exit the frequency regulation process. At 147 seconds, this caused a secondary frequency drop in the grid, and a significant decrease in the wind farm's frequency regulation capability, posing a serious threat to the safe operation of the wind farm. Furthermore, because the high wind speeds were already close to their limits, the failure to complete the frequency regulation task within 6 seconds after the frequency fluctuation was the most significant delay in system frequency recovery.

[0211] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0212] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A primary frequency regulation method for offshore wind power clusters and shared energy storage, characterized in that, It includes the following steps: Obtain the rotor safe operating speed ranges of each wind turbine in the offshore wind turbine cluster, and determine the wind speed division threshold based on the rotor safe operating speed ranges; Obtain the real-time access wind speeds of each wind turbine in the offshore wind turbine cluster, and divide the offshore wind turbine cluster into multiple wind turbine sub-clusters according to the wind speed division threshold; Construct a virtual reference wind turbine for each of the wind turbine sub-clusters, and determine the access wind speeds of the virtual reference wind turbines; Calculate the relative frequency modulation ability factors of each wind turbine according to the access wind speeds of the wind turbines in the wind turbine sub-cluster and the access wind speeds of the corresponding virtual reference wind turbines; Based on the constructed virtual reference wind turbines and the shared energy storage device, construct a primary frequency modulation mathematical model and an optimization objective function for the wind-storage combined system; Use a commercial solver to solve the primary frequency modulation mathematical model and the optimization objective function of the wind-storage combined system to obtain the optimal active power output reference values of the virtual reference wind turbines and the shared energy storage device; Obtain the active power output reference values of each wind turbine in the offshore wind turbine cluster according to the optimal active power output reference values of the virtual reference wind turbines and the relative frequency modulation ability factors of each wind turbine, and send them to each wind turbine; Control each wind turbine and the shared energy storage device to execute the frequency modulation output according to the corresponding active power output reference values to complete the primary frequency modulation process of the offshore wind power cluster and the shared energy storage.

2. The primary frequency regulation method for offshore wind power clusters and shared energy storage according to claim 1, characterized in that, The formula for determining the wind speed division threshold based on the rotor safe operating speed range is: Where R is the radius of the wind turbine, λ opt For the optimal tip speed ratio, [ω min ,ω max ] represents the safe operating speed range of the wind turbine rotor, ω min For, ω max v1, v3, and v5 are the determined first, second, and third wind speed classification thresholds, respectively.

3. A primary frequency regulation method for offshore wind power clusters and shared energy storage according to claim 1 or 2, characterized in that, The obtaining of the real-time access wind speeds of each wind turbine in the offshore wind turbine cluster and the dividing of the offshore wind turbine cluster into multiple wind turbine sub-clusters according to the wind speed division threshold specifically include: When the real-time access wind speed v of the wind turbine satisfies v1 < v < v3, include this wind turbine in the first wind turbine sub-cluster Ω1; When the real-time access wind speed V of the wind turbine satisfies V3 < V y < V5, the wind turbine is classified into the second wind turbine sub-cluster Ω2; When the real-time access wind speed V of the wind turbine satisfies V > V5, include this wind turbine in the third wind turbine sub-cluster Ω3; Among them, the first wind turbine sub-cluster Ω1 adopts an overspeed power limit and load reduction strategy, the second wind turbine sub-cluster Ω2 adopts a pitch angle power limit and load reduction strategy, and the third wind turbine sub-cluster Ω3 adopts a combined overspeed power limit and pitch angle power limit and load reduction strategy.

4. The primary frequency regulation method for offshore wind power clusters and shared energy storage according to claim 1, characterized in that, The constructing of a virtual reference wind turbine for each of the wind turbine sub-clusters and the determination of the access wind speeds of the virtual reference wind turbines specifically include: Construct a virtual reference wind turbine for each of the wind turbine sub-clusters, including virtual reference wind turbine x, virtual reference wind turbine y, and virtual reference wind turbine z, corresponding to the first wind turbine sub-cluster Ω1, the second wind turbine sub-cluster Ω2, and the third wind turbine sub-cluster Ω3 respectively; Obtain the real-time access wind speeds of each wind turbine in each wind turbine sub-cluster; Based on the statistical characteristics of the real-time access wind speed of all wind turbines within each wind turbine sub-cluster, the access wind speed of each virtual reference wind turbine is determined, including V. x V y V z The statistical characteristic values ​​mentioned therein include the weighted average or median.

5. The primary frequency regulation method for offshore wind power clusters and shared energy storage according to claim 1, characterized in that, The calculating of the relative frequency modulation ability factors of each wind turbine according to the access wind speeds of the wind turbines in the wind turbine sub-cluster and the access wind speeds of the corresponding virtual reference wind turbines specifically include: Calculate the load reduction rates of each wind turbine in each wind turbine sub-cluster; Based on the load reduction rates of each wind turbine in each wind turbine sub-cluster, combined with the access wind speeds of the corresponding virtual reference wind turbines, calculate the relative frequency modulation ability factors of each wind turbine.

6. The primary frequency regulation method for offshore wind power clusters and shared energy storage according to claim 5, characterized in that, The calculating of the load reduction rates of each wind turbine in each wind turbine sub-cluster specifically includes: For some fans with wind speeds close to the minimum operating wind speed, the load reduction rate for each fan is set to zero, i.e., d i =0, does not participate in the first frequency modulation process; For the first wind turbine subgroup Ω1 and the second wind turbine subgroup Ω2, the load reduction rate of each wind turbine is determined based on the linear relationship between the access wind speed and the load reduction rate of each wind turbine in subgroups Ω1 and Ω2. The formula is as follows: Where, d max The maximum load reduction rate of the wind turbine generator is given by Ω1, Ω2, and Ω3, which are the sets of wind turbines in the first, second, and third wind turbine sub-clusters, respectively. i Let v1 and v5 be the access wind speed of the i-th wind turbine, respectively, and let P(v1, v5) be the first and third wind speed thresholds. i ) i For fan i at wind speed v i Maximum operating power, P S P(v5) represents the total reserve power of the wind farm, and P(v5) represents the maximum operating power of the wind turbine corresponding to wind speed v5. For the third wind turbine sub-cluster, the output of each wind turbine in the third wind turbine sub-cluster Ω3 is adjusted to the output power corresponding to the third wind speed threshold ν5, and the load reduction rate determined therefrom is calculated using the following formula: Obtain the load reduction rate of each wind turbine in each wind turbine sub-cluster.

7. The primary frequency regulation method for offshore wind power clusters and shared energy storage according to claim 5, characterized in that, The calculation of the relative frequency regulation capability factor for each wind turbine, based on the load reduction rate of each turbine in each wind turbine sub-cluster and the access wind speed of the corresponding virtual reference turbine, specifically includes: For the first wind turbine sub-cluster Ω1, the formula for calculating the relative frequency regulation capability factor of each wind turbine is as follows: Among them, P im (t) and p xm (t) represents the electromagnetic power of wind turbine i in subgroup 1 and virtual reference wind turbine x, respectively. i0 With P x0 These represent the steady-state operating power of wind turbine i and the virtual reference wind turbine x after load reduction, respectively; v x With P x The operating wind speed and rated output power of the virtual reference fan; J i With J x The moments of inertia of fan i and fan x are respectively, ω id ω imin ω xd With ω xmin t1 and t3 represent the steady-state operating speed and minimum speed of fan i and virtual reference fan x, respectively; t1 and t3 correspond to the start and end times of frequency modulation, respectively. For the second wind turbine sub-cluster Ω2, the formula for calculating the relative frequency regulation capability factor of each wind turbine is as follows: Among them, P im (t) and P ym (t) represents the electromagnetic power of wind turbine i and virtual reference wind turbine y in the second wind turbine sub-cluster Ω2, respectively. i0 With P y0 These represent the steady-state operating power of wind turbine i and the virtual reference wind turbine y after load reduction, respectively; v y With P y The operating wind speed and rated output power of the virtual reference fan y; ω imax ω imin ω ymax With ω ymin These represent the maximum safe speed and minimum speed of fan i and virtual reference fan y, respectively; For the third wind turbine sub-cluster Ω3, the formula for calculating the relative frequency regulation capability factor of each wind turbine is as follows: Where, d i With d z These represent the load reduction rates of wind turbine i and virtual reference wind turbine z in the third wind turbine sub-cluster Ω3, respectively; P(v z ) z For the virtual reference fan z at the operating wind speed v z The maximum output power at that point.

8. The primary frequency regulation method for offshore wind power clusters and shared energy storage according to claim 1, characterized in that, The primary frequency regulation mathematical model of the wind-storage combined system is as follows: u(k)=[ΔP x (k)ΔP y (k)ΔP z (k)g(k)] T y(k)=[ΔP g (k),ΔP bess (k),∑ΔP wind (k),S(k),Δf(k)] T r(k)=P load (k) Among them, P V (k),P m (k) and ΔP g (k) represents the speed-regulating power, mechanical power, and electromagnetic power of a traditional thermal power unit at time k; ΔP bess S(k) and S(k) are the energy storage power regulation and state of charge at time k; and These represent the power adjustment values ​​of the three virtual reference wind turbines at time k, ∑ΔP wind (k) represents the total regulating power of the wind farm; Δf(k) represents the frequency change at time k, and ΔP x (k), ΔP y (k) and ΔP z g(k) represents the power regulation reference value of the three virtual reference wind turbines at time k; g(k) represents the reference value of the wind-storage coordination coefficient at time k.

9. The primary frequency regulation method for offshore wind power clusters and shared energy storage according to claim 1, characterized in that, The optimization objective function is: Where C is the optimization objective function, C β (k+i|k) is the prediction expression for the frequency regulation cost of the wind turbine pitch control at the current sampling time k to time k+i, C ωre (k+i|k) is the prediction expression for the frequency regulation cost of the wind turbine speed change at the current sampling time k to time k+i, C bess (k+i|k) is the prediction expression for the energy storage frequency regulation cost at the current sampling time k to the time k+i; Among them, the cost C of the wind turbine pitch adjustment process β The formula is: Where, Δβ z (k+i|k) is the prediction expression for the change in the z-pitch angle of the virtual reference wind turbine at the current sampling time k to the time k+i, β zd is the steady-state pitch angle of the virtual reference wind turbine z after load reduction; c is the wear coefficient of the wind turbine pitch angle; a1 is the cost coefficient of wind turbine pitch frequency regulation. The cost C of the fan rotor adjustment process ωre The formula is: Where, Δω x (k+i|k) and Δω y (k+i|k) is the prediction expression for the adjustment of the virtual reference fan's x and y speeds at the current sampling time k to time k+i; ω xf With ω yf Let x be the critical speed of the virtual reference fan and y be y, and let a2 be the cost coefficient for speed regulation.

10. The primary frequency regulation method for offshore wind power clusters and shared energy storage according to claim 1, characterized in that, The step of obtaining the active power output reference value of each wind turbine in the offshore wind turbine cluster based on the optimal active power output reference value of each virtual reference wind turbine and the relative frequency regulation capability factor of each wind turbine specifically includes: For the first wind turbine sub-cluster Ω1, the reference value for the active power output of each wind turbine is: in, Let x be the reference value for the active power output of the i-th wind turbine. i Let ΔP be the relative frequency regulation capability factor of the i-th wind turbine. x (k) represents the optimal active power output reference value of the virtual reference wind turbine x at time k; For the second wind turbine sub-cluster Ω2, the reference value for the active power output of each wind turbine is: Among them, y i Let ΔP be the relative frequency regulation capability factor of the i-th wind turbine. y (k) represents the optimal active power output reference value of the virtual reference wind turbine y at time k; For the third wind turbine sub-cluster Ω3, the reference value for the active power output of each wind turbine is: Among them, z i Let ΔP be the relative frequency regulation capability factor of the i-th wind turbine. z (k) represents the optimal active power output reference value of the virtual reference wind turbine z at time k.

Images