A rolling decision method for wind farm participation in grid frequency support response

Abstract

This invention discloses a rolling decision-making method for multiple wind farms participating in the grid frequency support response, belonging to the field of power system stability and control. This method obtains the priority of each wind farm's participation in frequency regulation under each level of control parameters based on the benefits and costs of the graded power response of each wind farm. For faults with concentrated anticipated faults, a strategy table for wind farm group participation in frequency regulation is formulated based on the priority of each wind farm's participation. Then, it determines whether a anticipated fault has occurred in the grid. If a fault has occurred, the wind farm group frequency regulation response strategy table is queried, and each wind farm is controlled to participate in frequency regulation response according to the predetermined strategy. If no fault occurs, it determines whether the next rolling prediction cycle has been reached. If it has, the system frequency response function is updated; if not, the determination of whether a anticipated fault has occurred in the grid continues. This method achieves the grid frequency regulation target with minimal wind farm group frequency regulation costs.

Description

Technical Field

[0001] This invention belongs to the field of power system stability and control, and specifically relates to a rolling decision-making method for multiple wind farms participating in the grid frequency support response. Background Technology

[0002] Wind power is increasingly integrated into power systems, becoming a crucial power source. However, the output power of wind turbines is decoupled from the grid frequency, making them unable to respond to frequency changes. Therefore, when large-scale wind power is integrated into the grid, the overall system's frequency regulation capability is significantly weakened, potentially threatening the frequency stability of the power system. Many countries have established grid connection specifications regarding how wind power can provide frequency support for the power system. Some specifications allow wind power to participate in primary frequency regulation, while others require wind power to have a rapid frequency response capability.

[0003] When a power system suddenly experiences a significant power shortage, the frequency drops rapidly. Current frequency support methods available in the power system include power modulation via DC transmission lines, pumped storage, rapid shelving of interruptible loads, and the rapid frequency response of wind turbines. Except for wind power, the supporting power provided by other frequency support methods has a fixed value; however, the supporting power provided by wind power varies over time, depending on the frequency regulation control method and parameters, as well as the initial wind speed before frequency regulation begins. The frequency regulation capability and cost of wind turbines differ depending on the initial wind speed. Achieving the grid's frequency regulation target with minimal cost through reasonable decision-making is crucial for maintaining system frequency stability and protecting the economic interests of wind farms. However, there is currently no widely adopted decision-making method, especially for scenarios involving multiple wind farms participating in frequency regulation. Summary of the Invention

[0004] The purpose of this invention is to provide a rolling decision-making method for multiple wind farms participating in the frequency support response of the power grid. First, the total power support demand of the power system for the wind farm group can be determined. Then, the wind farms participating in the frequency regulation response within the group and the magnitude of their control parameters can be determined, thereby achieving the frequency regulation target of the power grid with the minimum frequency regulation cost of the wind farm group.

[0005] To achieve the above objectives, the rolling decision-making method for multiple wind farms participating in grid frequency support response of the present invention includes the following steps:

[0006] Step 1: Update the system frequency response function based on the starting combination of synchronous generators in the power grid and the changes in their power generation.

[0007] Step 2: Roll out the power output of each wind farm and obtain the wind speed of all units in each wind farm based on the wind speed combination model of each wind farm.

[0008] Step 3: Based on the anticipated fault set of the power grid, determine the power support index and corresponding system frequency change curves of the wind farm cluster under different anticipated faults;

[0009] Step 4: Input the expected frequency change curves under different anticipated faults into the simulation models of each wind farm using graded control parameters, and calculate the graded power response curves of each wind farm; each wind farm simulation model refers to a fast simulation model that only retains mechanical dynamics.

[0010] Step 5: Based on the benefits and costs of the graded power response of each wind farm, obtain the priority of each wind farm participating in frequency regulation under each level of control parameters;

[0011] Step 6: For the anticipated concentrated faults, formulate a strategy table for the wind farm group to participate in frequency regulation based on the priority of each wind farm's participation in frequency regulation.

[0012] Step 7: Determine if the anticipated fault has occurred in the power grid. If a fault has occurred, proceed to step 8; otherwise, proceed to step 9.

[0013] Step 8: Query the frequency regulation response strategy table of the wind farm group and control each wind farm to participate in frequency regulation response according to the established strategy;

[0014] Step 9: Determine if the next rolling forecast cycle has been reached. If it has, return to step 1; otherwise, return to step 7.

[0015] Furthermore, the wind farm cluster includes four power support indicators under different anticipated faults, specifically the maximum value of active power change ΔP. MAX The time T for the change in active power to reach its maximum value M The time T for the change in active power to return to zero Z and the minimum allowable value ΔP for active power variation MIN ;

[0016] In step 3, the search range for these four power support indicators was obtained through offline calculation and online matching. The specific method is as follows:

[0017] Offline calculation of the frequency regulation response of a single wind turbine under different wind speeds and different system power deficits yields ΔP. MAX The four index values ​​at their maximum, ΔP MIN The four index values ​​at the minimum, T Z Maximum value and T Z Minimum value; where the step size for different wind speeds is 0.1 m / s, and the step size for different system power deficits is 0.01 pu, with a range of 0.05 pu to 0.2 pu;

[0018] During online matching, based on the wind speed prediction results for each turbine in the wind farm from step 2, for each turbine, the calculation results under the current anticipated fault are retrieved from the offline calculation results, and ΔP is calculated. MAX The four indicator values ​​at their maximum values ​​are connected to form an indicator curve L. MAX , will ΔP MIN The four minimum indicator values ​​are connected to form an indicator curve L. MIN ;

[0019] Then, the two index curves of each unit in the wind farm group are summed to obtain the two response index curves L of the wind farm group. MAX_Σ and L MIN_Σ Next, L MAX_Σ The maximum active power on the surface is used as the indicator ΔP MAX The upper limit of the value will determine the equivalent step support power ΔP of the wind farm. step As an indicator ΔP MAX The lower limit of the value;

[0020] Take L MIN_Σ The minimum active power on the surface is used as the index ΔP MIN The lower limit of the value, the index ΔP MIN The upper limit of the value is 0; take the index T from all wind turbine units. Z The maximum and minimum values ​​are respectively used as indicators T. Z The upper and lower limits of the possible values.

[0021] Furthermore, in step 3, the method for obtaining the power support index of the wind farm cluster under different anticipated faults is as follows:

[0022] First, the control objective f is to achieve the lowest possible system frequency. min1 With minimizing the power increase as the optimization objective, T is set. Z and T M The ratio is 3:1, and the index ΔP is determined using an optimization algorithm. MAX and T Z The value;

[0023] Then, fix the determined ΔP. MAX T Z and T M The value, expressed as the index ΔP MIN The lower limit is the initial value, and ΔP is increased successively. MIN Until the system frequency second drop amplitude Δf is met min2 Until the requirements are met.

[0024] Furthermore, in step 4, the graded control parameters of the wind farm refer to setting three sets of parameters of different sizes to obtain frequency regulation responses of different intensities and corresponding frequency regulation costs. The largest parameter is the maximum control parameter that can ensure the stable operation of the wind turbine under various anticipated faults, the smallest parameter is the parameter value that meets the requirements of the national standard for wind power frequency regulation response, and the intermediate parameter is the average of the largest and smallest parameters.

[0025] Furthermore, in step 5, the benefit of wind farm graded power response refers to the area on the power change curve of wind farm frequency regulation response where the power is higher than the steady-state power before frequency regulation; the cost of wind farm graded power response refers to the area on the power change curve of wind farm frequency regulation response where the power is lower than the steady-state power before frequency regulation.

[0026] Furthermore, in step 5, the priority of each wind farm participating in frequency regulation under each level of control parameters is obtained by sorting them from largest to smallest based on the ratio of the benefits to costs of each wind farm's frequency regulation power response.

[0027] Furthermore, in step 6, the specific method for formulating the strategy table for wind farm clusters to participate in frequency regulation is as follows: for a certain fault in the expected fault concentration, based on the priority of each wind farm participating in frequency regulation obtained in step 5, a curve of the wind farm cluster that is higher than and closest to the total power support demand index of the wind farm cluster obtained in step 3 is obtained through a combination optimization method, along with several wind farms participating in the response and the control parameters of these wind farms.

[0028] Furthermore, in step 8, the specific method for querying the wind farm group frequency regulation response strategy table is as follows: based on the actual fault that occurs in the power grid, first obtain the power deficit of the power grid based on the measurement data, and then find a strategy in the strategy table formulated in step 6 that is higher than and closest to the actual power deficit, and obtain the result of whether each wind farm participates in frequency regulation and its frequency regulation control parameters.

[0029] Compared with the prior art, the present invention has the following significant advantages: The present invention can quickly make decisions on how wind farm clusters participate in grid frequency support. On the one hand, it is used to determine the total power support demand of the grid for wind farm clusters, and on the other hand, it determines the frequency regulation response parameters of each wind farm in the wind farm cluster. Thus, the grid's predetermined frequency regulation target can be achieved with the minimum frequency regulation cost of wind farm clusters. This plays an important role in promoting the advancement of wind power participation in grid frequency support technology. Attached Figure Description

[0030] Figure 1 It is a flowchart of a rolling decision-making method for multiple wind farms participating in grid frequency support response;

[0031] Figure 2 This is a schematic diagram of an IEEE-9 node system used to illustrate the implementation method;

[0032] Figure 3 This is a structural diagram of a fast simulation model used for frequency response simulation of wind farms;

[0033] Figure 4a The wind speed of the wind turbine was obtained based on wind farm output prediction data named WF-79.

[0034] Figure 4b The wind speed of the wind turbine was obtained based on the power output prediction data of a wind farm named WF-48.

[0035] Figure 5 This is a diagram of the decision indicators for wind farm output power;

[0036] Figure 6 It is an index curve when the power deficit is 0.1 and the minimum frequency control target is 49.5.

[0037] Figure 7 It is a comparison chart of the power output curve as an indicator curve and the frequency when wind power does not participate in frequency regulation;

[0038] Figure 8a This is a response curve of a wind farm named WF-79 under graded parameters;

[0039] Figure 8b This is the response curve of the wind farm named WF-48 under the graded parameters;

[0040] Figure 9 The relationship between the wind farm cluster response and indicators under the proposed strategy when the power deficit is 0.1 and the minimum frequency control target is 49.5;

[0041] Figure 10 This is a comparison chart of frequencies when the power deficit is 0.098 and the minimum frequency control target is 49.5, with wind power participating in frequency regulation according to the decision table and wind power not participating in frequency regulation. Detailed Implementation

[0042] The technical solutions in 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 embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0043] Various aspects of the invention are described with reference to the accompanying drawings, which illustrate numerous illustrative embodiments. The disclosed embodiments are not necessarily defined to include all aspects of the invention. It should be understood that the various concepts and embodiments described above, as well as those described below in more detail, can be implemented in any of many ways, because the concepts and embodiments disclosed herein are not limited to any particular implementation. Furthermore, some aspects of the invention disclosed may be used alone or in any suitable combination with other aspects of the invention disclosed.

[0044] The following is based on Figure 1 The flowchart of the method of the present invention shown below illustrates the specific implementation of the present invention:

[0045] Step 1: Update the system frequency response function based on the starting combination of synchronous generators in the power grid and the changes in their power generation.

[0046] This embodiment uses a simulation system for illustration. The structure of the simulation system is as follows: Figure 2 As shown. The simulation system includes two wind farms, WF-48 and WF-79, containing 48 and 79 wind turbines respectively. The frequency regulation control of the wind turbines uses conventional droop control. The frequency response function (transfer function) of this simulation system is:

[0047]

[0048] This frequency response function only includes the frequency regulation effect of synchronous generators and power loads, and does not include the frequency regulation response of wind power. Based on the above frequency response function and the simplified mechanical dynamic model of the wind turbine, a model can be established as follows: Figure 3 The image shows a fast simulation model used for simulating the frequency response of a wind farm.

[0049] Step 2: Roll out the power output of each wind farm and obtain the wind speed of all units in each wind farm based on the wind speed combination model of each wind farm.

[0050] Assuming the predicted power output of wind farm WF-48 is 16MW and the predicted power output of wind farm WF-79 is 79MW, based on the wind speed combination model of these two wind farms, we can obtain the following... Figure 4a , Figure 4b The wind speeds of all turbines in the wind farm shown.

[0051] Step 3: Based on the anticipated fault set of the power grid, determine the power support indicators and corresponding system frequency variation curves of the wind farm cluster under different anticipated faults. The anticipated fault set involves scanning all faults in the power grid and retaining the fault components that have a significant impact on the power grid frequency.

[0052] Furthermore, the power support index for wind farm clusters under different anticipated faults includes four values, specifically the maximum value of active power change ΔP. MAX The time T for the change in active power to reach its maximum value M The time T for the change in active power to return to zero Z and the minimum allowable value ΔP for active power variation MIN .

[0053] The graph formed by the four indicators is as follows Figure 5 As shown, the indicators are determined based on the frequency regulation target. The expected frequency regulation target can be achieved when the active power output curve of the wind farm is higher than the graph enclosed by the indicators.

[0054] Furthermore, the search range of the four power support indicators for the wind farm cluster under different anticipated faults was obtained through offline calculation and online matching. Specifically, the frequency regulation response of a single wind turbine was calculated offline under different wind speeds (variation step size 0.1 m / s) and different system power deficits (variation step size 0.01 pu, variation range 0.05 pu to 0.2 pu), to obtain ΔP. MAX The four index values ​​at their maximum, ΔP MIN The four index values ​​at the minimum, T Z Maximum value and T Z Minimum value; During online matching, based on the wind speed prediction results for each turbine in the wind farm from step 2, for each turbine, the calculation results under the current anticipated fault are obtained from the offline calculation results, and ΔP is calculated. MAX The four indicator values ​​at their maximum values ​​are connected to form an indicator curve L. MAX , will ΔP MIN The four minimum indicator values ​​are connected to form an indicator curve L. MIN Then, the two index curves of each unit in the wind farm group are summed to obtain the two response index curves L of the wind farm group. MAX_Σ and L MIN_Σ Next, L MAX_Σ The maximum active power on the index ΔP MAX The upper limit of the value will determine the equivalent step support power ΔP of the wind farm. step For the indicator ΔP MAX The lower limit of the value of L; MIN_Σ The minimum active power on the index ΔP MIN The lower limit of the value, the index ΔP MIN The upper limit of the value is 0; take the index T from all wind turbine units. Z The maximum and minimum values ​​are respectively used as indicators T. Z The upper and lower limits of the possible values.

[0055] Based on the above method, the value ranges of the four indicators are shown in Table 1:

[0056] Table 1 shows the range of decision-making indicators when the power deficit is 0.1 pu.

[0057] index Range of values index Range of values <![CDATA[ΔP MAX (pu)]]> [0.009,0.244] <![CDATA[T Z (s)]]> [2.272,11.312] <![CDATA[ΔP MIN (pu)]]> [-0.291,0] <![CDATA[T M (s)]]> [0.757,3.771]

[0058] Furthermore, the power support index of the wind farm cluster under different anticipated faults is obtained through a step-by-step determination method. Specifically, the method is as follows: First, the control target f is to achieve the minimum system frequency. min1 With minimizing the power increase as the optimization objective, T is set. Z and T M The ratio is 3:1, and the index ΔP is determined using an optimization algorithm. MAX and T Z The value of ΔP is then fixed; MAX T Z and T M The value, expressed as the index ΔP MIN The lower limit is the initial value, and ΔP is increased successively. MIN Until the system frequency second drop amplitude Δf is met min2 Until the requirements are met.

[0059] Under a pre-conceived fault causing a power deficit of 0.1 pu, the wind farm cluster power support index and the corresponding system frequency variation curves are as follows: Figure 6 and Figure 7 As shown. The frequency modulation target achieved by this indicator is that the minimum first frequency drop is no less than 49.50Hz, and the second frequency drop is less than 0.03Hz.

[0060] Step 4: Input the expected frequency variation curves under different anticipated faults into the simulation models of each wind farm using graded control parameters, and calculate the results. Figure 8a , Figure 8b The diagram shows the graded power response curves for each wind farm. The simulation models for each wind farm refer to rapid simulation models that only retain mechanical dynamics.

[0061] Furthermore, the graded control parameters of a wind farm refer to setting three sets of parameters of different sizes to obtain frequency regulation responses of different intensities and corresponding frequency regulation costs. The largest parameter is the maximum control parameter that can ensure the stable operation of the wind turbine under various anticipated faults, the smallest parameter is the parameter value that meets the requirements of the national standard for wind power frequency regulation response, and the intermediate parameter is the average of the largest and smallest parameters.

[0062] In this embodiment, the three sets of parameters are set as shown in Table 2:

[0063] Table 2 Values ​​of Level 3 Control Parameters

[0064]

[0065] Step 5: Based on the benefits and costs of the graded power response of each wind farm, obtain the priority of each wind farm participating in frequency regulation under each level of control parameters.

[0066] Furthermore, the benefit of graded power response of wind farms refers to the area on the power change curve of the wind farm frequency regulation response where the power is higher than the steady-state power before frequency regulation; the cost of graded power response of wind farms refers to the area on the power change curve of the wind farm frequency regulation response where the power is lower than the steady-state power before frequency regulation for the first 50 seconds.

[0067] Furthermore, the priority of each wind farm participating in frequency regulation under each level of control parameters is obtained by sorting them from largest to smallest based on the ratio of the benefits to costs of each wind farm's frequency regulation power response.

[0068] The benefit-to-cost ratios of the two wind farms under various parameters, as well as the ranking of response priorities, are shown in Tables 3 and 4:

[0069] Table 3: Ratio of Benefits to Costs for Each Level of Wind Farm

[0070] Response level Wind farm WF-48 Wind farm WF-79 Level 1 0.4198 0.3666 Level 2 0.5983 0.3679 Level 3 0.2677 0.3496

[0071] Table 4 Priority Ranking Table

[0072]

[0073] Step 6: For the anticipated concentrated faults, formulate a strategy table for the wind farm group to participate in frequency regulation based on the priority of each wind farm's participation in frequency regulation.

[0074] Furthermore, the specific method for formulating the strategy table for wind farm clusters to participate in frequency regulation is as follows: For a certain fault in the expected fault concentration, based on the priority of each wind farm participating in frequency regulation obtained in step 5, a curve of the wind farm cluster that is higher than and closest to the total power support demand index of the wind farm cluster obtained in step 3 is obtained through a combination optimization method, along with several wind farms participating in the response and the control parameters of these wind farms.

[0075] Taking a anticipated fault causing a power deficit of 0.1 pu as an example, the corresponding power support index of the wind farm group is as follows: Figure 6 As shown in Table 5, the strategy for wind farm cluster participation in frequency regulation is formulated based on the priority of each wind farm's participation in frequency regulation. The relationship between wind farm cluster response and indicators is shown in Table 5. Figure 9 As shown in the figure, the response of the wind farm cluster is higher than the index graph, which can meet the frequency regulation requirements.

[0076] Table 5. Guiding control parameters for wind farm clusters when the power deficit is 0.1.

[0077] wind farm Response level Control parameter values WF-48 2 50 WF-79 2 50

[0078] Step 7: Determine if the expected fault has occurred in the power grid. If a fault has occurred, proceed to step 8; if no fault has occurred, proceed to step 9.

[0079] Step 8: Query the frequency regulation response strategy table of the wind farm group and control each wind farm to participate in frequency regulation response according to the established strategy.

[0080] Furthermore, the specific method for querying the frequency regulation response strategy table of the wind farm group is as follows: First, based on the actual fault that occurred in the power grid, the power deficit of the power grid is obtained based on the measurement data. Then, in the strategy table formulated in step 6, a strategy that is higher than and closest to the actual power deficit is found, and the results of whether each wind farm participates in frequency regulation and its frequency regulation control parameters are obtained.

[0081] Assuming a anticipated fault causing a power deficit of 0.098 pu occurs, according to the strategy table lookup principle, the response strategy for a anticipated fault with a power deficit of 0.1 pu should be adopted, as shown in Table 5. The frequency response results of the two wind farms before and after participating in frequency regulation according to the strategy in Table 5 are as follows: Figure 10 As shown in the figure, the minimum frequency drop was effectively increased to above 49.50Hz, and the second frequency drop was less than 0.03Hz, which can accurately achieve the expected support effect on the system frequency.

[0082] Step 9: Determine if the next rolling forecast cycle has been reached. If it has, return to step 1; otherwise, return to step 7.

[0083] The rolling prediction cycle is related to the wind farm output prediction speed, prediction distance, and response strategy production cycle under the anticipated fault set. The settings will vary for different systems and users. In this embodiment, the rolling prediction cycle is 20 seconds. The rolling execution is as described above and will not be repeated.

[0084] While the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the invention. Those skilled in the art can make various modifications and refinements without departing from the spirit and scope of the invention. Therefore, the scope of protection of the present invention shall be determined by the claims.

Claims

1. A rolling decision-making method for multiple wind farms participating in the grid frequency support response, characterized by: Includes the following steps: Step 1: Update the system frequency response function based on the starting combination of synchronous generators in the power grid and the changes in their power generation. Step 2: Roll out the power output of each wind farm and obtain the wind speed of all units in each wind farm based on the wind speed combination model of each wind farm. Step 3: Scan all faults in the power grid, retaining those that have a significant impact on the grid frequency to form a set of anticipated faults. Based on this set of anticipated faults, determine the power support indicators and corresponding system frequency change curves for the wind farm cluster under different anticipated faults. in: The wind farm cluster includes four power support indicators under different anticipated faults, specifically the maximum value of active power change. The time it takes for the active power change to reach its maximum value Time for active power change to return to zero Minimum allowable value for active power variation ; In this step, the search range for the four power support indicators was obtained through offline calculation and online matching. Specifically: Offline calculation of the frequency regulation response of a single wind turbine under different wind speeds and different system power deficits yields... The four index values ​​at their maximum The four indicator values ​​at the minimum hour Maximum sum Minimum value; where the step size for different wind speeds is 0.1 m / s, and the step size for different system power deficits is 0.01 pu, with a range of 0.05 pu to 0.2 pu; During online matching, based on the wind speed prediction results for each turbine in the wind farm from step 2, the calculation results under the current anticipated fault are retrieved from the offline calculation results for each turbine. The four indicator values ​​at their maximum values ​​are connected to form an indicator curve. L MAX ,Will The four minimum indicator values ​​are connected to form an indicator curve. L MIN ; Then, the two index curves of each unit in the wind farm group are summed to obtain the two response index curves L of the wind farm group. MAX_Σ and L MIN_Σ Next, L MAX_Σ The maximum active power on the surface is used as an indicator. The upper limit of the value will determine the equivalent step support power of the wind farm. As an indicator The lower limit of the value; Take L MIN_Σ The minimum active power on the surface is used as an indicator. The lower limit of the value of the indicator The upper limit of the value is 0; take the index from all wind turbine units. The maximum and minimum values ​​are used as indicators. The upper and lower limits of the possible values; Step 4: Input the expected frequency change curves under different anticipated faults into the simulation models of each wind farm using graded control parameters, and calculate the graded power response curves of each wind farm. Step 5: Based on the benefits and costs of the graded power response of each wind farm, obtain the priority of each wind farm participating in frequency regulation under each level of control parameters; Step 6: For the anticipated concentrated faults, formulate a strategy table for the wind farm group to participate in frequency regulation based on the priority of each wind farm's participation in frequency regulation. Step 7: Determine if the anticipated fault has occurred in the power grid. If a fault has occurred, proceed to step 8; otherwise, proceed to step 9. Step 8: Query the frequency regulation response strategy table of the wind farm group and control each wind farm to participate in frequency regulation response according to the established strategy; Step 9: Determine if the next rolling forecast cycle has been reached. If it has, return to step 1; otherwise, return to step 7.

2. The rolling decision-making method according to claim 1, characterized in that: In step 3, the method for obtaining the power support index of the wind farm cluster under different anticipated faults is as follows: First, the goal is to achieve the minimum system frequency control. With the goal of minimizing power generation, the following settings are made: and The ratio is 3:1, and the index is determined using an optimization algorithm. and The value; Then, fix the already determined , and The value, in terms of indicators The lower limit is the initial value, which is increased successively. Until the system frequency second drop amplitude is met Until the requirements are met.

3. The rolling decision-making method according to claim 1, characterized in that: step In section 4, the graded control parameters of the wind farm refer to setting three sets of parameters of different sizes to obtain frequency regulation responses of different intensities and corresponding frequency regulation costs. The largest parameter is the maximum control parameter that can ensure the stable operation of the wind turbine under various anticipated faults. The smallest parameter is the parameter value that meets the requirements of the national standard for wind power frequency regulation response. The intermediate parameter is the average value of the largest and smallest parameters.

4. The rolling decision-making method according to claim 1, characterized in that: In step 5, the benefit of wind farm graded power response refers to the area on the power change curve of wind farm frequency regulation response where the power is higher than the steady-state power before frequency regulation; the cost of wind farm graded power response refers to the area on the power change curve of wind farm frequency regulation response where the power is lower than the steady-state power before frequency regulation.

5. The rolling decision-making method according to claim 1, characterized in that: In step 5, the priority of each wind farm participating in frequency regulation under each level of control parameters is obtained by sorting them from largest to smallest based on the ratio of the benefits to costs of each wind farm's frequency regulation power response.

6. The rolling decision-making method according to claim 1, characterized in that: In step 6, the specific method for formulating the strategy table for wind farm group participation in frequency regulation is as follows: For a certain fault in the expected fault concentration, based on the priority of each wind farm participating in frequency regulation obtained in step 5, a curve of the wind farm group that is higher than and closest to the total power support demand index of the wind farm group obtained in step 3 is obtained through a combination optimization method, along with several wind farms participating in the response and the control parameters of these wind farms.

7. The rolling decision-making method according to claim 1, characterized in that: In step 8, the specific method for querying the frequency regulation response strategy table of the wind farm group is as follows: based on the actual fault that occurs in the power grid, first obtain the power deficit of the power grid based on the measurement data, and then find the strategy in the strategy table formulated in step 6 that is higher than and closest to the actual power deficit, and obtain the result of whether each wind farm participates in frequency regulation and its frequency regulation control parameters.

Images