A three-segment frequency regulation control strategy for wind farms considering the time-delay effect of wake effect

Abstract

The present invention discloses a three-segment frequency regulation control strategy for a wind farm that takes into account the time-lag influence of the wake effect, and belongs to the field of frequency stability and control of a new power system. The control strategy combines the wake effect and time-lag effect that exist when the wind turbine participates in the frequency regulation of the power system, and designs three comprehensive inertia control strategies according to the different frequency curve characteristics and expected purposes in different time periods in the frequency regulation. In the first stage, in order to be able to quickly respond to load disturbances, improve the FN point and accelerate frequency stability, a comprehensive inertia control that considers the speed and frequency change is designed; in the second stage, considering the influence of the wake effect and time-lag effect on the frequency fluctuation, an adaptive virtual inertia control strategy based on the rotor kinetic energy is designed to effectively suppress the frequency fluctuation; in the third stage, a comprehensive inertia control strategy that changes with time is adopted. This dynamic adjustment is conducive to helping the system frequency quickly reach a steady-state value when the wake disturbance is reduced.

Description

Technical Field

[0001] The present invention relates to the field of frequency stability and control of novel power systems, and in particular to a three-segment frequency regulation control strategy for a wind farm taking into account the time lag influence of wake effect. Background Art

[0002] With the vigorous development of new energy, the large-scale access of wind power to the grid has also brought great challenges to the grid. Wind power generation equipment based on power electronic interfaces replaces traditional synchronous machines for large-scale grid connection. It is difficult to actively respond to changes in system frequency and presents near-zero inertia to the outside, resulting in a reduction in system inertia. When the load changes suddenly, the system frequency will fluctuate greatly. In order to enable the power system to operate safely and stably, wind turbines need to have frequency response capabilities and participate in grid frequency regulation. Comprehensive inertia control is a commonly used wind turbine frequency regulation control strategy, which includes two control methods: virtual inertia control and droop control. The frequency change rate and grid frequency deviation are introduced as input signals into the power control system of the unit. By quickly releasing the rotor kinetic energy, the wind turbine has the ability to quickly respond to frequency changes.

[0003] As the scale of wind farms expands, the wake effect between wind turbines and the time lag effect caused by the wake effect are becoming increasingly non-negligible. The wake effect is a phenomenon in which the wind speed received by the downstream wind turbine is reduced due to the energy absorbed by the wind turbine blades after the incoming wind passes through the upstream wind turbine. The wake effect causes the wind speed received by each wind turbine to be different from the rotor kinetic energy it possesses. Since there is a certain distance between the upstream and downstream wind turbines to ensure the normal operation of the wind turbines, there is a certain delay time for the wake wind speed of the upstream wind turbine to propagate from the upstream to the downstream wind turbine rotor. This is the time lag effect. The time lag time of wind turbines in different rows is also different. The time lag time of the upstream wind turbine wake is smaller than that of the downstream wind turbine wake. When the system is disturbed, the fan begins to release rotor kinetic energy under the action of comprehensive inertia. At this time, the fan speed decreases, which causes the upstream fan's wake wind speed to increase. After the time lag period, the increased wake wind of the upstream fan reaches the downstream fan, causing the downstream fan speed to increase, with more rotor kinetic energy to participate in the system frequency support, but this will also cause the frequency to fluctuate greatly with the arrival of wake fluctuations. Therefore, it is very important to study the segmented frequency support method based on the key time lag time point. Summary of the invention

[0004] The purpose of the present invention is to overcome the defects existing in the above-mentioned background technology. The present invention proposes a three-stage frequency regulation control strategy for wind farms that takes into account the time lag effect of the wake effect. The control strategy combines the wake effect and time lag effect existing when the wind turbine participates in the frequency regulation of the power system, and designs three comprehensive inertia control strategies according to the different frequency curve characteristics and expected purposes in different time periods in the frequency regulation. Different control targets in the three stages are achieved. In the first stage, the FN point is improved and it helps the power grid to restore stability as soon as possible in the early stage of the disturbance. In the second stage, the frequency fluctuation is effectively suppressed. In the third stage, the system frequency reaches the steady-state value as soon as possible.

[0005] The present invention adopts the following technical solutions to solve the above technical problems:

[0006] A three-segment frequency regulation control strategy for a wind farm taking into account the time lag of the wake effect comprises the following steps:

[0007] Step S1, assuming that the natural wind flows perpendicularly into the wind farm, and the wind turbines in the wind farm are arranged in a rectangular shape with equal distances between rows. In this case, the wind farm can be divided into rows with n wind turbines arranged in series; the time from the occurrence of the load disturbance until the first wave of wake fluctuations in the wind farm propagates to the downstream wind turbines is recorded as t 1 moment, which is the first stage; during this period, the wake fluctuation caused by the release of rotor kinetic energy of the upstream fan has not yet reached the downstream fan. The fan frequency regulation during this period does not need to consider the influence of the wake effect on the frequency regulation capability. Therefore, an adaptive comprehensive inertia control strategy is designed to improve the FN of the fan and make the frequency reach the steady-state value as quickly as possible. The FN is the lowest frequency point;

[0008] Step S2: From t 1 From time t, all wind turbines in the wind farm are in the wake fluctuation caused by frequency modulation. 2 This stage is the second stage; during this period, the frequency has been fluctuating greatly due to the time lag effect, so an adaptive virtual inertia control strategy based on rotor kinetic energy is designed, which not only takes into account the limitation of rotor speed, but also uses virtual inertia control to suppress frequency fluctuations and maintain frequency stability;

[0009] Step S3: The third stage is from t 2 From time t to the time t when the frequency reaches the steady-state value ess The purpose of this stage is to restore the frequency to a stable value as soon as possible. Therefore, a time-varying comprehensive inertia control strategy is designed. The virtual inertia coefficient is represented by a parabolic function and the droop coefficient is represented by a linear function. Both of them decrease with time. This method effectively reduces the inhibitory effect of the comprehensive inertia on the frequency in the frequency recovery stage, which is beneficial to improve the FN point and delay the frequency drop.

[0010] Furthermore, the specific steps of the adaptive comprehensive inertia control strategy are as follows:

[0011] Step S2-1, in the first stage 0~t 1 In the process, the system is subject to load disturbance. Under the action of the comprehensive inertia control strategy, the wind turbine begins to release the stored rotor kinetic energy, causing the wake wind speed of each wind turbine to begin to change. However, the wake of the upstream wind turbine has not reached the downstream wind turbine in this stage.

[0012] Step S2-2, in order to make the droop coefficient respond timely according to the current frequency drop of the power system, a droop coefficient set according to the frequency deviation is proposed, where the droop coefficient is smaller when the frequency deviation is smaller, and the droop coefficient is larger when the frequency deviation is larger;

[0013] Step S2-3, in the first stage, since the load disturbance frequency will reach the FN point in this stage, compared with the second and third stages, the first stage requires more rotor kinetic energy to provide frequency support. In order to prevent the rotor speed from exceeding the safe range due to the excessive droop coefficient during the frequency modulation process, a speed protection link needs to be added;

[0014] Step S2-4: If virtual inertia control with a fixed coefficient is adopted, although frequency support can be provided according to RoCoF when the frequency drops, it will hinder the frequency increase in the frequency recovery stage; therefore, a control strategy is designed to adjust the virtual inertia control coefficient according to the positive and negative value of RoCoF. When RoCoF is less than zero, virtual inertia control is adopted; when RoCoF is greater than zero, virtual negative inertia control is adopted, and RoCoF is the frequency change rate.

[0015] Furthermore, the specific steps of the adaptive virtual inertia control strategy based on rotor kinetic energy are as follows:

[0016] Step S3-1: In the second stage t 1 ~t 2 In the wind farm, the wake effect of the upstream wind turbine gradually affects the downstream wind turbine under the action of the time lag effect, and the operating conditions of the wind turbine also change frequently during this period;

[0017] In order to reduce the system frequency fluctuation in this stage, a virtual inertia control strategy considering the fan operating conditions is designed; the rotor speed directly reflects the operating state of the fan under non-operating conditions. The higher the speed, the more rotor kinetic energy, and the more system frequency support can be provided; therefore, the fan rotor speed ω and the virtual inertia coefficient K of the second stage are established. df2 The coupling relationship;

[0018] Step S3-2, when the wind speed of the downstream fan increases under the influence of the wake effect of the upstream fan, the fan speed increases and has a larger rotor kinetic energy. At this time, the virtual inertia control coefficient increases accordingly, releasing more rotor kinetic energy to provide timely frequency support for the system; when the wind speed of the downstream fan decreases under the influence of the wake of the upstream fan, the rotor kinetic energy that can be released by the fan decreases, and the virtual inertia control coefficient also decreases accordingly to prevent the fan from being cut off due to excessive release of rotor kinetic energy.

[0019] Furthermore, the specific steps of the time-varying integrated inertia control strategy are as follows:

[0020] Step S4-1: In the third stage t 2 ~t ess In the present invention, all wind turbines in the wind farm are under the influence of the wake effect, and the frequency fluctuation of the wind farm gradually decreases with time. In order to make the system frequency reach the steady-state value as soon as possible, a time-varying comprehensive inertia control strategy is designed.

[0021] Step S4-2, perform droop control. Droop control is to control the power output of the fan by frequency deviation. The larger the initial speed of the fan rotor, the more rotor kinetic energy is stored in the fan rotor, indicating that the fan is in a better operating condition. At this time, the larger the initial value and final value of the droop gain, the longer it takes to reach the final value. The initial value and final value of the droop coefficient of the fan with a smaller initial rotor speed are smaller than the corresponding value of the droop coefficient of the fan with a larger initial speed, and the time to converge to the final value is also shorter, which can effectively prevent the fan speed from being excessively released and falling below the lower limit value.

[0022] Step S4-3, perform inertia control, where RoCoF is used to control the power output of the wind turbine;

[0023] Step S4-4: design a virtual inertia coefficient represented by a parabolic function according to the inertia coefficient solution formula.

[0024] Furthermore, in step S2-1, 1 The calculation formula is:

[0025]

[0026] In the formula, x 1,2 v is the distance from the first wind turbine to the second wind turbine in the windward direction of the wind farm; 1 is the initial wind speed of the first fan, which is equal to the incoming wind speed; at this stage, the frequency needs to be restored as quickly as possible;

[0027] The droop coefficient in step S2-2 is:

[0028]

[0029] In the formula, K pf1,max is the set maximum droop coefficient; Δf min is the minimum frequency deviation; Δf max is the maximum frequency deviation;

[0030] The relationship between the droop coefficient, the rotor speed per unit value, and the frequency deviation in step S2-3 is as follows:

[0031]

[0032] In the formula, is the current per unit speed of the fan rotor; is the lower limit per unit value of the fan rotor speed, which is 0.7pu; The upper limit per unit value of the fan rotor speed is 1.3 pu; is the per unit value of the optimal speed of the fan;

[0033] The virtual inertia coefficient K of the first stage in step S2-4 df1 The expression is:

[0034]

[0035] In the formula, K d is a constant greater than 0.

[0036] Furthermore, in step S3-1, 2 The calculation formula is:

[0037]

[0038] In the formula, t i-1,i is the time lag between the i-1th wind turbine and the i-th wind turbine in the wind farm; x i-1,i is the distance along the wind direction between the i-1th wind turbine and the i-th wind turbine in the wind farm; v i-1 is the initial wind speed of the i-1th wind turbine in the wind farm;

[0039] The rotor kinetic energy that can be released or absorbed by the fan in step S3-1 is:

[0040]

[0041] In the formula, H WT is the inertia constant of the fan; ω r is the current rotor speed of the fan; ω min is the minimum value of the fan rotor speed;

[0042] The virtual inertia control parameter of the second stage defined in step S3-2 is proportional to the rotor kinetic energy currently possessed by the wind turbine, and it can be obtained that:

[0043]

[0044] In the formula, ω max is the maximum value of the fan rotor speed; k is the virtual inertia control frequency modulation proportional coefficient in the fan frequency support stage.

[0045] Furthermore, the power output by the droop control in step S4-2 is expressed as:

[0046]

[0047] Where 1 / R is the droop gain; f nom is the rated power of the system; f sys is the current power of the system;

[0048] In order to ensure that the fan can recover stability as soon as possible, the droop coefficient is improved and the droop coefficient according to the frequency modulation time is obtained, which can be expressed by a linear function as follows:

[0049] K pf3 (t) = a droop t+b droop (9)

[0050] In the formula, a droop is the slope of the linear curve; b droop is the initial droop gain value; in order to prevent excessive release of rotor kinetic energy during frequency recovery, the droop coefficient should be gradually reduced over time, so a droop The coefficient is negative, but b droop The coefficient must be positive for the droop coefficient to be positive;

[0051] Furthermore, the power output by the virtual inertia control in step S4-3 is expressed as:

[0052]

[0053] Where K is the virtual inertia coefficient;

[0054] Formula (10) can be simplified using the per unit value of the speed:

[0055]

[0056] In the formula, The power per unit value added for virtual inertia control, is the per unit value of the current system frequency;

[0057] By integrating both sides of equation (11), we can get the additional electric energy of the fan as:

[0058]

[0059] In the formula, The per-unit value of the rotor kinetic energy released for virtual inertia control; is the per-unit value of the system steady-state frequency before the load disturbance occurs;

[0060] The per-unit power that the fan rotor can release is:

[0061]

[0062] In the formula, is the per-unit value of the rotor power that the fan can release; H is the current per unit speed of the fan rotor; WT is the inertia constant of the fan;

[0063] By integrating both sides of equation (13) with respect to time, the rotor kinetic energy that can be released by the fan is obtained as:

[0064]

[0065] In the formula, is the per unit value of the rotor kinetic energy that can be released by the fan; is the per unit value of the initial rotor speed of the fan before the load disturbance occurs in the system;

[0066] Based on the principle of energy conservation, assuming that the rotor kinetic energy released by the wind turbine is completely provided by virtual inertia control, we can get:

[0067]

[0068] Simplifying, we can get:

[0069]

[0070] Furthermore, the virtual inertia coefficient expression represented by the parabolic function in step S4-4 is as follows:

[0071] K df3 =a inertia t 2 +b inertia (17)

[0072]

[0073] In the formula, a inertia is the quadratic coefficient of the virtual inertia control parameter formula; b inertia is the constant term of the virtual inertia control parameter formula; t trip is the time when the designed virtual inertia control parameters begin to decrease; t final K is the time for the virtual inertia control parameter to reach the final value; initialis the initial value of the virtual inertia control coefficient; K final is the final value of the virtual inertia control coefficient;

[0074] Furthermore, the process of solving the initial value and final value of the virtual inertia control coefficient is as follows: Assuming that the system frequency is the rated value before the load disturbance occurs in the power grid, the initial value of the virtual inertia control coefficient is:

[0075]

[0076] In the formula, is the lower limit per unit value of the system frequency, which is 0.996 pu; H WT is the inertia constant of the fan; is the per unit value of the initial rotor speed of the fan before the load disturbance occurs in the system; is the lower limit per unit value of the fan rotor speed; therefore, the initial value of the virtual inertia control coefficient depends on the fan rotor speed before the load disturbance occurs in the system;

[0077] The final value of the virtual inertia control coefficient is:

[0078]

[0079] In the formula, is the upper limit per unit value of the system frequency; is the current per unit speed of the fan rotor;

[0080] K general It is expressed as:

[0081]

[0082] In the formula, K is the per unit value of the minimum operating rotor speed; general It represents the virtual inertia control coefficient when the rotor speed is minimum. The final virtual inertia control coefficient depends on the ratio of the stored rotor kinetic energy to the maximum amount of released rotor kinetic energy.

[0083] Compared with the prior art, the present invention adopts the above technical solution and has the following beneficial effects:

[0084] (1) The present invention proposes a three-stage frequency regulation control strategy for wind farms that takes into account the influence of wake effect and time lag. Considering the three-stage frequency regulation process with different characteristics of wind turbines participating in power system frequency regulation under the effects of wake and time lag, three different comprehensive inertia control strategies are designed to achieve different control objectives in the three stages. In the first stage, the FN point is improved and it helps the power grid to restore stability as soon as possible in the early stage of disturbance. In the second stage, frequency fluctuations are effectively suppressed. In the third stage, the system frequency reaches a steady-state value as soon as possible.

[0085] (2) Compared with other methods, the three-segment frequency regulation control strategy for wind farms proposed in the present invention, which takes into account the influence of wake effect and time lag, can accurately improve the efficiency and accuracy of frequency regulation according to the characteristics of different stages, taking into account the wake effect and time delay effect that affect the operating state of the wind turbine in the actual frequency regulation process, and enhancing the system's adaptability and response speed to disturbances. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 It is the grid frequency curve designed by the present invention taking into account the influence of wake time lag;

[0087] Figure 2 It is a schematic diagram of the droop control of the stage designed by the present invention;

[0088] Figure 3 It is a schematic diagram of virtual inertial control in stage 1 of the design of the present invention;

[0089] Figure 4 This is a schematic diagram of virtual inertial control in stage 2 of the present invention;

[0090] Figure 5 It is a schematic diagram of the third stage droop control designed by the present invention;

[0091] Figure 6 It is a schematic diagram of the third stage virtual inertia control designed by the present invention. DETAILED DESCRIPTION

[0092] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0093] A three-stage frequency regulation control strategy for wind farms taking into account the time lag of wake effect. The specific division of the three stages is as follows: Figure 1 As shown, the steps are as follows:

[0094] A. Assume that the natural wind flows perpendicularly into the wind farm, and the wind turbines in the wind farm are arranged in a rectangular shape with equal distances between each row. In this case, the wind farm can be divided into rows equal to n wind turbines arranged in series. From the occurrence of load disturbance until the first wave of wake fluctuations in the wind farm propagates to the downstream wind turbines, this time is recorded as t 1 At this moment, this stage is the first stage. During this period, the wake fluctuation caused by the release of rotor kinetic energy of the upstream fan has not yet reached the downstream fan. The fan frequency regulation during this period does not need to consider the influence of the wake effect on the frequency regulation capability. Therefore, an adaptive comprehensive inertia control strategy is designed to increase the FN point of the fan and make the frequency reach the steady-state value as quickly as possible;

[0095] B. From t 1 From time t, all wind turbines in the wind farm are in the wake fluctuation caused by frequency modulation. 2 This stage is the second stage. During this period, the frequency has been fluctuating greatly due to the time lag effect, so an adaptive virtual inertia control strategy based on rotor kinetic energy is designed, which not only takes into account the limitation of rotor speed, but also uses virtual inertia control to suppress frequency fluctuations and maintain frequency stability.

[0096] C. The third stage is from t 2 From time t to the time t when the frequency reaches the steady-state value ess The purpose of this stage is to restore the frequency to a stable value as soon as possible. Therefore, a time-varying integrated inertia control strategy is designed. The virtual inertia coefficient is represented by a parabolic function and the droop coefficient is represented by a linear function. Both of them gradually decrease with the increase of time. This method effectively reduces the inhibitory effect of the integrated inertia on the frequency during the frequency recovery stage, which is beneficial to improve the FN point and delay the frequency drop.

[0097] The above-mentioned three-segment frequency regulation control strategy for wind farms taking into account the time lag of wake effect, and the specific process of the adaptive comprehensive inertia control strategy proposed in step A are as follows:

[0098] A1. In the first stage 0~t 1 In the process, the system is subject to load disturbance. Under the action of the comprehensive inertia control strategy, the wind turbine begins to release the stored rotor kinetic energy, causing the wake wind speed of each wind turbine to begin to change. However, the wake of the upstream wind turbine has not yet reached the downstream wind turbine during this stage. 1 The calculation formula is:

[0099]

[0100] In the formula, x 1,2 v is the distance from the first wind turbine to the second wind turbine in the windward direction of the wind farm; 1 is the initial wind speed of the first fan, which is equal to the incoming wind speed.

[0101] In this stage, the frequency needs to be restored as quickly as possible, so an adaptive integrated inertia control strategy is designed.

[0102] A2. In order to make the droop coefficient respond timely to the current frequency drop of the power system, a droop coefficient set according to the frequency deviation is proposed. When the frequency deviation is small, the droop coefficient is small, and when the frequency deviation is large, the droop coefficient is large. The proposed droop coefficient is:

[0103]

[0104] In the formula, K pf1,max is the set maximum droop coefficient; Δf min is the minimum frequency deviation; Δf max is the maximum frequency deviation.

[0105] A3. In the first stage, since the load disturbance frequency will reach the FN point in this stage, compared with the second and third stages, the first stage requires more rotor kinetic energy to provide frequency support. In order to prevent the rotor speed from exceeding the safe range due to the excessive droop coefficient during the frequency modulation process, a speed protection link needs to be added.

[0106] The relationship between the droop coefficient, rotor speed per unit value, and frequency deviation is as follows:

[0107]

[0108] In the formula, is the current per unit speed of the fan rotor; is the lower limit per unit value of the fan rotor speed, which is 0.7pu; is the upper limit per unit value of the fan rotor speed, which is 1.3pu; is the per unit value of the optimal speed of the fan. min ≤|Δf|≤Δf max The droop coefficient graph of the interval is as follows Figure 2 shown.

[0109] A4. If virtual inertia control with a fixed coefficient is used, although it can provide frequency support according to RoCoF when the frequency drops, it will hinder the frequency increase during the frequency recovery phase. Therefore, a control strategy is designed to adjust the virtual inertia control coefficient according to the positive and negative value of RoCoF. When RoCoF is less than zero, virtual inertia control is used; when RoCoF is greater than zero, virtual negative inertia control is used. Figure 3 As shown, its expression is:

[0110]

[0111] In the formula, K d is a constant greater than 0.

[0112] The aforementioned three-segment frequency regulation control strategy for a wind farm taking into account the time lag of the wake effect, and the process of an adaptive virtual inertia control strategy taking into account the wind turbine speed proposed in step B are specifically as follows:

[0113] B1. In the second stage 1 ~t 2In the wind farm, the wake effect of the upstream wind turbine gradually affects the downstream wind turbine under the action of the time lag effect, and the operating conditions of the wind turbine also change frequently during this period. 2 The calculation formula is:

[0114]

[0115] In the formula, t i-1,i is the time lag between the i-1th wind turbine and the i-th wind turbine in the wind farm; x i-1,i is the distance along the wind direction between the i-1th wind turbine and the i-th wind turbine in the wind farm; v i-1 is the initial wind speed of the i-1th wind turbine in the wind farm.

[0116] In order to reduce the system frequency fluctuation in this stage, a virtual inertia control strategy considering the wind turbine operating conditions is designed.

[0117] The rotor speed directly reflects the operating state of the fan when it is not in operation. The higher the speed, the more rotor kinetic energy it has and the more system frequency support it can provide. Therefore, the fan rotor speed ω and the second-stage virtual inertia coefficient K are established. df2 coupling relationship.

[0118] The rotor kinetic energy that a fan can release or absorb is:

[0119]

[0120] In the formula, H WT is the inertia constant of the fan; ω r is the current rotor speed of the fan; ω min is the minimum value of the fan rotor speed.

[0121] B2. Define the virtual inertia control parameters of the second stage to be proportional to the rotor kinetic energy currently possessed by the wind turbine, such as Figure 4 As shown, the following formula can be obtained:

[0122]

[0123] In the formula, ω max is the maximum value of the fan rotor speed; k is the virtual inertia control frequency modulation proportional coefficient in the fan frequency support stage.

[0124] When the wind speed of the downstream fan increases under the influence of the wake effect of the upstream fan, the fan speed increases and has a larger rotor kinetic energy. At this time, the virtual inertia control coefficient increases accordingly, releasing more rotor kinetic energy to provide timely frequency support for the system; when the wind speed of the downstream fan decreases under the influence of the wake of the upstream fan, the rotor kinetic energy that can be released by the fan decreases, and the virtual inertia control coefficient also decreases accordingly, preventing the fan from being cut off due to excessive release of rotor kinetic energy.

[0125] In the above-mentioned three-segment frequency regulation control strategy for a wind farm taking into account the time lag of the wake effect, the specific steps of proposing a comprehensive inertia control strategy that varies with time in step C are as follows:

[0126] C1. In the third stage 2 ~t ess In the present invention, all wind turbines in the wind farm are under the influence of the wake effect, and the frequency fluctuation of the wind farm gradually decreases with time. In order to make the system frequency reach the steady-state value as soon as possible, a time-varying comprehensive inertia control strategy is designed.

[0127] C2. Droop control controls the power output of the fan by frequency deviation. The power output of droop control is expressed as:

[0128]

[0129] Where 1 / R is the droop gain; f nom is the rated power of the system; f sys is the current power of the system.

[0130] In order to ensure that the fan can recover stability as soon as possible, the droop coefficient is improved and the droop coefficient according to the frequency modulation time is obtained, such as Figure 5 As shown, it can be expressed as a linear function:

[0131] K pf3 (t) = a droop t+b droop (9)

[0132] In the formula, a droop is the slope of the linear curve; b droop is the initial droop gain value. In order to prevent excessive release of rotor kinetic energy during frequency recovery, the droop coefficient should be gradually reduced over time, so a droop The coefficient is negative, but b droop The coefficient must be positive for the droop coefficient to be positive.

[0133] The larger the initial speed of the fan rotor, the more rotor kinetic energy is stored in the fan rotor, indicating that the fan is in a better operating condition. At this time, the larger the initial and final values ​​of the droop gain are, the longer it takes to reach the final value. final,1 The initial and final values ​​of the droop coefficient of the fan with a smaller initial rotor speed are both smaller than the corresponding values ​​of the droop coefficient of the fan with a larger initial speed. The time t to converge to the final value is final,2 It is also small, which can effectively prevent the fan speed from being excessively released and falling below the lower limit.

[0134] C3. Inertia control is to control the power output of the wind turbine by RoCoF. The power output of virtual inertia control is expressed as:

[0135]

[0136] Where K is the virtual inertia coefficient.

[0137] Formula (10) can be simplified using the per unit value of the speed:

[0138]

[0139] In the formula, The power per unit value added for virtual inertia control, It is the per unit value of the current system frequency.

[0140] By integrating both sides of equation (11), we can get the additional electric energy of the fan as:

[0141]

[0142] In the formula, The per-unit value of the rotor kinetic energy released for virtual inertia control; It is the per-unit value of the system steady-state frequency before the load disturbance occurs.

[0143] The per-unit power that the fan rotor can release is:

[0144]

[0145] In the formula, is the per-unit value of the rotor power that the fan can release; H is the current per unit speed of the fan rotor; WT is the inertia constant of the fan.

[0146] By integrating both sides of equation (13) with respect to time, the rotor kinetic energy that can be released by the fan is obtained as:

[0147]

[0148] In the formula, is the per unit value of the rotor kinetic energy that can be released by the fan; It is the per-unit value of the initial rotor speed of the fan before the system is disturbed by load.

[0149] Based on the principle of energy conservation, assuming that the rotor kinetic energy released by the wind turbine is completely provided by virtual inertia control, we can get:

[0150]

[0151] Simplifying, we can get:

[0152]

[0153] C4. Based on the obtained formula for solving the inertia coefficient, a virtual inertia coefficient represented by a parabolic function is designed, such as Figure 6 As shown, the expression is as follows:

[0154] K df3 =a inertia t 2 +b inertia (17)

[0155]

[0156] In the formula, a inertia is the quadratic coefficient of the virtual inertia control parameter formula; b inertia is the constant term of the virtual inertia control parameter formula; t trip is the time when the designed virtual inertia control parameters begin to decrease; t final K is the time for the virtual inertia control parameter to reach the final value; initial is the initial value of the virtual inertia control coefficient; K final is the final value of the virtual inertia control coefficient.

[0157] Assuming that the system frequency is the rated value before the load disturbance occurs in the power grid, the initial value of the virtual inertia control coefficient is:

[0158]

[0159] In the formula, is the lower limit per unit value of the system frequency, which is 0.996 pu. WT is the inertia constant of the fan; is the per unit value of the initial rotor speed of the fan before the load disturbance occurs in the system; is the lower limit per unit value of the fan rotor speed; therefore, the initial value of the virtual inertia control coefficient depends on the rotor speed of the fan before the load disturbance occurs in the system.

[0160] The final value of the virtual inertia control coefficient is:

[0161]

[0162] In the formula, is the upper limit per unit value of the system frequency; It is the current per unit speed of the fan rotor.

[0163] K general It is expressed as:

[0164]

[0165] In the formula, K is the per unit value of the minimum operating rotor speed. general It represents the virtual inertia control coefficient when the rotor speed is minimum. The final virtual inertia control coefficient depends on the ratio of the stored rotor kinetic energy to the maximum amount of released rotor kinetic energy.

[0166] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A three-segment frequency regulation control strategy for a wind farm taking into account the time lag of the wake effect, characterized in that: The control strategy includes the following steps: Step S1, assuming that the natural incoming wind blows perpendicularly into the wind farm, and the wind turbines in the wind farm are arranged in a rectangular shape, with equal distances between each row. At this time, the wind farm can be arranged in rows equal to n wind turbines arranged in series; from the occurrence of the load disturbance until the first wave of wake fluctuations in the wind farm propagates to the downstream wind turbines, this time is recorded as time t1, and this stage is the first stage; during this period of time, the wake fluctuations caused by the release of rotor kinetic energy of the upstream wind turbine have not yet reached the downstream wind turbine, and the wind turbine frequency regulation during this period does not need to consider the influence of the wake effect on the frequency regulation capability, so an adaptive comprehensive inertia control strategy is designed to improve the FN of the wind turbine and make the frequency reach the steady-state value as quickly as possible, and the FN is the lowest frequency point; Step S2, from time t1 to time t2 when all wind turbines in the wind farm are in wake fluctuation caused by frequency modulation, this stage is the second stage; during this period of time, the frequency has been fluctuating greatly due to the time lag effect, so an adaptive virtual inertia control strategy based on rotor kinetic energy is designed, which not only takes into account the limitation of rotor speed, but also adopts virtual inertia control to suppress frequency fluctuation and maintain frequency stability; Step S3: The third stage is from time t2 to time t2 when the frequency reaches the steady-state value. ess The purpose of this stage is to restore the frequency to a stable value as soon as possible. Therefore, a time-varying comprehensive inertia control strategy is designed. The virtual inertia coefficient is represented by a parabolic function and the droop coefficient is represented by a linear function. Both of them decrease with time. This method effectively reduces the inhibitory effect of the comprehensive inertia on the frequency in the frequency recovery stage, which is beneficial to improve the FN point and delay the frequency drop.

2. A three-segment frequency modulation control strategy for a wind farm taking into account the time lag of the wake effect according to claim 1, characterized in that: The specific steps of the adaptive integrated inertia control strategy are as follows: Step S2-1, in the first stage 0-t1, the system is subject to load disturbance, and the wind turbine begins to release the stored rotor kinetic energy under the action of the comprehensive inertia control strategy, causing the wake wind speed of each wind turbine to begin to change, but the wake of the upstream wind turbine has not reached the downstream wind turbine during this stage; Step S2-2, in order to make the droop coefficient respond timely according to the current frequency drop of the power system, a droop coefficient set according to the frequency deviation is proposed, where the droop coefficient is smaller when the frequency deviation is smaller, and the droop coefficient is larger when the frequency deviation is larger; Step S2-3, in the first stage, since the load disturbance frequency will reach the FN point in this stage, compared with the second and third stages, the first stage requires more rotor kinetic energy to provide frequency support. In order to prevent the rotor speed from exceeding the safe range due to the excessive droop coefficient during the frequency modulation process, a speed protection link needs to be added; Step S2-4: If virtual inertia control with a fixed coefficient is adopted, although frequency support can be provided according to RoCoF when the frequency drops, it will hinder the frequency increase in the frequency recovery stage; therefore, a control strategy is designed to adjust the virtual inertia control coefficient according to the positive and negative value of RoCoF. When RoCoF is less than zero, virtual inertia control is adopted; when RoCoF is greater than zero, virtual negative inertia control is adopted, and RoCoF is the frequency change rate.

3. A three-segment frequency modulation control strategy for a wind farm taking into account the time lag of the wake effect according to claim 1, characterized in that: The specific steps of the adaptive virtual inertia control strategy based on rotor kinetic energy are as follows: Step S3-1, in the second stage t1-t2, the wake effect of the upstream wind turbine in the wind farm gradually affects the downstream wind turbine under the action of the time lag effect, and the operating conditions of the wind turbine also change frequently during this period; In order to reduce the system frequency fluctuation in this stage, a virtual inertia control strategy considering the wind turbine operating conditions is designed; The rotor speed directly reflects the operating state of the fan when it is not in operation. The higher the speed, the more rotor kinetic energy it has and the more system frequency support it can provide. Therefore, the fan rotor speed ω and the second-stage virtual inertia coefficient K are established. df2 The coupling relationship; Step S3-2, when the wind speed of the downstream fan increases under the influence of the wake effect of the upstream fan, the fan speed increases and has a larger rotor kinetic energy. At this time, the virtual inertia control coefficient increases accordingly, releasing more rotor kinetic energy to provide timely frequency support for the system; when the wind speed of the downstream fan decreases under the influence of the wake of the upstream fan, the rotor kinetic energy that can be released by the fan decreases, and the virtual inertia control coefficient also decreases accordingly to prevent the fan from being cut off due to excessive release of rotor kinetic energy.

4. A three-segment frequency modulation control strategy for a wind farm taking into account the time lag of the wake effect according to claim 1, characterized in that: The specific steps of the time-varying integrated inertia control strategy are as follows: Step S4-1: In the third stage t2-t ess In the present invention, all wind turbines in the wind farm are under the influence of the wake effect, and the frequency fluctuation of the wind farm gradually decreases with time. In order to make the system frequency reach the steady-state value as soon as possible, a time-varying comprehensive inertia control strategy is designed. Step S4-2, perform droop control. Droop control is to control the power output of the fan by frequency deviation. The larger the initial speed of the fan rotor, the more rotor kinetic energy is stored in the fan rotor, indicating that the fan is in a better operating condition. At this time, the larger the initial value and final value of the droop gain are, the longer it takes to reach the final value. The initial value and final value of the droop coefficient of the fan with a smaller initial rotor speed are both smaller than the corresponding value of the droop coefficient of the fan with a larger initial speed, and the time to converge to the final value is also shorter, which can effectively prevent the fan speed from being excessively released and falling below the lower limit; Step S4-3, perform inertia control, where RoCoF is used to control the power output of the wind turbine; Step S4-4: design a virtual inertia coefficient represented by a parabolic function according to the inertia coefficient solution formula.

5. A three-segment frequency modulation control strategy for a wind farm taking into account the time lag of the wake effect according to claim 2, characterized in that: The calculation formula of t1 in step S2-1 is: In the formula, x 1,2 is the distance from the first wind turbine to the second wind turbine in the windward direction of the wind farm; v1 is the initial wind speed of the first wind turbine, which is equal to the incoming wind speed; at this stage, the frequency needs to be restored as soon as possible; The droop coefficient in step S2-2 is: In the formula, K pf1,max is the set maximum droop coefficient; Δf min is the minimum frequency deviation; Δf max is the maximum frequency deviation; The relationship between the droop coefficient, the rotor speed per unit value, and the frequency deviation in step S2-3 is as follows: In the formula, is the current per unit speed of the fan rotor; is the lower limit per unit value of the fan rotor speed, which is 0.7pu; The upper limit per unit value of the fan rotor speed is 1.3 pu; is the per unit value of the optimal speed of the fan; The virtual inertia coefficient K of the first stage in step S2-4 df1 The expression is: In the formula, K d is a constant greater than 0.

6. A three-segment frequency regulation control strategy for a wind farm taking into account the time lag of the wake effect according to claim 3, characterized in that: The calculation formula of t2 in step S3-1 is: Where, t i-1,i is the time lag between the i-1th wind turbine and the i-th wind turbine in the wind farm; x i-1,i is the distance along the wind direction between the i-1th wind turbine and the i-th wind turbine in the wind farm; v i-1 is the initial wind speed of the i-1th wind turbine in the wind farm; The rotor kinetic energy that can be released or absorbed by the fan in step S3-1 is: In the formula, H WT is the inertia constant of the fan; ω r is the current rotor speed of the fan; ω min is the minimum value of the fan rotor speed; The virtual inertia control parameter of the second stage defined in step S3-2 is proportional to the rotor kinetic energy currently possessed by the wind turbine, and it can be obtained that: In the formula, ω max is the maximum value of the fan rotor speed; k is the virtual inertia control frequency modulation proportional coefficient in the fan frequency support stage.

7. A three-segment frequency modulation control strategy for a wind farm taking into account the time lag of the wake effect according to claim 4, characterized in that: The power output by the droop control in step S4-2 is expressed as: Where 1 / R is the droop gain; f nom is the rated power of the system; f sys is the current power of the system; In order to ensure that the fan can recover stability as soon as possible, the droop coefficient is improved and the droop coefficient according to the frequency modulation time is obtained, which can be expressed by a linear function as follows: K pf3 (t)=a droop t+b droop (9) In the formula, a droop is the slope of the linear curve; b droop is the initial droop gain value; in order to prevent excessive release of rotor kinetic energy during frequency recovery, the droop coefficient should be gradually reduced over time, so a droop The coefficient is negative, but b droop The coefficient must be positive for the droop coefficient to be positive.

8. A three-segment frequency modulation control strategy for a wind farm taking into account the time lag of the wake effect according to claim 4, characterized in that: The power output by the virtual inertia control in step S4-3 is expressed as: Where K is the virtual inertia coefficient; Formula (10) can be simplified using the per unit value of the speed: In the formula, The power per unit value added for virtual inertia control, is the per unit value of the current system frequency; By integrating both sides of equation (11), we can get the additional electric energy of the fan as: In the formula, The per-unit value of the rotor kinetic energy released for virtual inertia control; is the per-unit value of the system steady-state frequency before the load disturbance occurs; The per-unit power that the fan rotor can release is: In the formula, is the per-unit value of the rotor power that the fan can release; H is the current per unit speed of the fan rotor; WT is the inertia constant of the fan; By integrating both sides of equation (13) with respect to time, the rotor kinetic energy that can be released by the fan is obtained as: In the formula, is the per unit value of the rotor kinetic energy that can be released by the fan; is the per unit value of the initial rotor speed of the fan before the load disturbance occurs in the system; Based on the principle of energy conservation, assuming that the rotor kinetic energy released by the wind turbine is completely provided by virtual inertia control, we can get: Simplifying, we can get:

9. A three-segment frequency modulation control strategy for a wind farm taking into account the time lag of the wake effect according to claim 4, characterized in that: The virtual inertia coefficient expression represented by the parabolic function in step S4-4 is as follows: K df3 =a inertia t 2 +b inertia (17) In the formula, a inertia is the quadratic coefficient of the virtual inertia control parameter formula; b inertia is the constant term of the virtual inertia control parameter formula; t trip is the time when the designed virtual inertia control parameters begin to decrease; t final K is the time for the virtual inertia control parameter to reach the final value; initial is the initial value of the virtual inertia control coefficient; K final is the final value of the virtual inertia control coefficient.

10. A three-segment frequency regulation control strategy for a wind farm taking into account the time lag of the wake effect according to claim 9, characterized in that: The process of solving the initial value and final value of the virtual inertia control coefficient is as follows: Assuming that the system frequency is the rated value before the load disturbance occurs in the power grid, the initial value of the virtual inertia control coefficient is: In the formula, is the lower limit per unit value of the system frequency, which is 0.996 pu; H WT is the inertia constant of the fan; is the per unit value of the initial rotor speed of the fan before the load disturbance occurs in the system; is the lower limit per unit value of the fan rotor speed; therefore, the initial value of the virtual inertia control coefficient depends on the fan rotor speed before the load disturbance occurs in the system; The final value of the virtual inertia control coefficient is: In the formula, is the upper limit per unit value of the system frequency; is the current per unit speed of the fan rotor; K general It is expressed as: In the formula, K is the per unit value of the minimum operating rotor speed; general It represents the virtual inertia control coefficient when the rotor speed is minimum. The final virtual inertia control coefficient depends on the ratio of the stored rotor kinetic energy to the maximum amount of released rotor kinetic energy.

Images