Primary frequency regulation method for wind storage combined system based on multivariable fuzzy logic control

Abstract

This invention provides a primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control, comprising: establishing a wind-storage integrated system model, dividing wind speed regions, and establishing a multivariable fuzzy logic control method. The primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control provided by this invention can realize power allocation for wind power and energy storage participating in primary frequency regulation, fully leverage the frequency regulation advantages of DFIG and energy storage, and improve the frequency stability of power systems with high wind power penetration.

Description

Technical Field

[0001] This invention relates to the field of wind farm technology, and in particular to a primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control. Background Technology

[0002] The inherent volatility and intermittency of new energy sources, such as wind power, significantly impact the safe and stable operation of the power grid, resulting in insufficient active support for the grid, primarily manifested in its influence on grid frequency. However, wind turbines cannot spontaneously achieve inertial response and primary frequency regulation; additional control measures such as rotor overspeed control, pitch angle control, and virtual inertial integrated control are required, or energy storage should be incorporated into the wind farm to form a wind-storage combined power generation system to establish a coupling relationship. If wind turbines participate in frequency regulation alone in a wind-storage system, it can easily cause a secondary drop or rise in grid frequency during frequency recovery. If energy storage participates in frequency regulation alone, its required capacity is too large and fails to fully utilize the wind turbine's ability to participate in power system frequency regulation. Therefore, the frequency regulation performance of wind power and energy storage should be fully utilized to form a wind-storage combined power generation system to participate in the primary frequency regulation of the power system. Further research is needed on how wind turbines and energy storage can collaboratively participate in frequency regulation. Existing research has not considered the different frequency regulation capabilities of wind turbines under different wind speed scenarios, has not fully utilized the frequency regulation potential of wind turbines and energy storage, and has not formulated corresponding wind-storage frequency regulation control strategies based on different wind conditions. Therefore, it is essential to design a primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control. Summary of the Invention

[0003] The purpose of this invention is to provide a primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control, which can realize the power allocation of wind power and energy storage participating in primary frequency regulation.

[0004] To achieve the above objectives, the present invention provides the following solution:

[0005] A primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control includes the following steps:

[0006] Step 1: Establish a wind-storage integrated system model, specifically as follows:

[0007] Establish a system frequency response model:

[0008] Obtaining wind power increment ΔP W Energy storage system power increment ΔP E Load change ΔP L The active power increment ΔP of other energy units besides the wind and storage system participating in primary frequency regulation G ,in,

[0009]

[0010] In the formula, R G T is the prime mover adjustment coefficient. G T is the time constant of the speed controller. ch T is the turbine time constant. R F is the reheat time constant. H Given the reheater gain, the system frequency response model transfer function is obtained as follows:

[0011]

[0012] In the formula, H, D, and p are the inertial constant, damping constant, and wind power penetration rate, respectively.

[0013] Establish the DFIG model:

[0014] The mechanical power output of the fan is obtained as follows:

[0015]

[0016] In the formula, ρ is the air density, A is the swept area of ​​the wind turbine blade, and R... w Where is the radius of the blade, v is the wind speed, and C is the wind speed. p Where λ is the wind energy utilization coefficient, β is the tip speed ratio, and C is the blade pitch angle. p (λ,β) is a nonlinear function of λ and β. The DFIG model is established as follows:

[0017]

[0018] In the formula, γ is the introduced intermediate variable, ω is the rotor speed of the wind turbine, and the wind turbine is operated in a reduced load mode. Virtual inertial frequency control based on the frequency change rate is added to reduce the amount of speed change during the frequency regulation process.

[0019] Establish a supercapacitor model:

[0020] Construct an equivalent circuit model of a supercapacitor, including an ideal capacitor C and a series equivalent resistance R. es and the parallel equivalent resistance R that reflects the long-term static changes of the supercapacitor eq The voltage of the supercapacitor is:

[0021]

[0022] Supercapacitor energy storage capacity E scss The remaining charge E at time t C (t) and state of charge (SOC) k (t) is:

[0023]

[0024] In the formula, U max Umin These are the highest and lowest operating voltages of the supercapacitor, respectively, in V and P. C (t) represents the supercapacitor's charging and discharging power at time t, in MW, η C E represents the charging and discharging efficiency of supercapacitors. scss The total capacity of the supercapacitor is expressed in MWh. The power storage and SOC constraints are as follows:

[0025]

[0026] In the formula, P rmin P rmax SOC min SOC max These represent the minimum output power, maximum output power, minimum and maximum SOC of the supercapacitor, respectively. Since the voltage of a single supercapacitor is relatively low, multiple supercapacitors can be connected in series and parallel to form a large-capacity supercapacitor array. Let n be the number of supercapacitors connected in series. s The number of parallel connections is n p The equivalent internal resistance and equivalent capacitance are then:

[0027]

[0028] Step 2: Divide the wind speed areas, specifically as follows:

[0029] The minimum cut-in wind speed of DFIG is v. in The maximum cut-out wind speed is v out The DFIG operates at speeds of 0.7–1.2 pu, dividing wind speeds into three zones: low, medium, and high, and two critical wind speed zones.

[0030] Low wind speed area v in -v1, the DFIG rotor speed is constant at 0.7pu, the DFIG does not have frequency modulation capability, only the supercapacitor participates in frequency modulation;

[0031] Critical low wind speed zone v lo -v m1 When the wind speed changes frequently in the low and medium wind speed zones, the active power increment of DFIG participating in frequency regulation changes slowly until it withdraws from frequency regulation.

[0032] The medium wind speed zone v1-v2 consists of the maximum power tracking zone and the constant speed zone;

[0033] Critical high wind speed zone v m2 -v h When the wind speed changes frequently in the medium and high wind speed zones, the active power increment of DFIG participating in frequency regulation changes slowly until it withdraws from frequency regulation.

[0034] High wind speed zone v2-vout The DFIG rotor speed is 1.2 pu, and pitch angle control is introduced to maintain the wind turbine output power at the rated power without load shedding for standby.

[0035] Step 3: Establish a multivariable fuzzy logic control method, specifically as follows:

[0036] Obtain the system frequency and calculate the frequency deviation Δf based on the system frequency, and determine whether Δf meets the condition |Δf|≥0.033Hz;

[0037] If it does not meet the requirements, determine whether the energy storage SOC is ideal. If it is not ideal, perform SOC recovery and set the SOC charge and discharge control coefficient. If it is ideal, recalculate Δf and make a judgment.

[0038] If the conditions are met, the wind speed is checked, and it is determined whether the area is a medium wind speed zone.

[0039] If it is in a medium wind speed area, the DFIG adopts rotor overspeed load reduction control to provide backup capacity, and combines virtual inertial control frequency modulation. The supercapacitor adopts virtual droop control frequency modulation. After the frequency modulation is completed, it is determined whether dΔf / dt has changed polarity. If it has, the DFIG additional power is set to 0 and the supercapacitor frequency modulation is performed. If there is no change, the previous step is returned to perform DFIG and supercapacitor frequency modulation.

[0040] If it is not in the medium wind speed zone, determine whether Δf is greater than 0. If it is less than 0, then perform supercapacitor frequency modulation. If it is greater than 0, determine whether it is in the high wind speed zone. If it is in the high wind speed zone, then perform DFIG pitch angle control and energy storage virtual inertial integrated control. If it is not in the high wind speed zone, then perform supercapacitor frequency modulation.

[0041] DFIG employs rotor overspeed unloading control to provide backup capacity, specifically as follows:

[0042] In the process of participating in power system frequency regulation, the DFIG (Diverterless Geological Injection Generator) uses rotor overspeed load shedding control to provide reserve capacity, while the supercapacitor also shares part of the reserve capacity to reduce wind curtailment. Specifically:

[0043] When the supercapacitor does not provide backup, the backup capacity provided by the DFIG will be used as the total backup capacity required by the wind-storage system, as follows:

[0044] P R (t)=(1-α)P max (t)=d%P max (t) (9)

[0045] In the formula, α is the wind power utilization rate, and P max P represents the maximum power in MPPT operation mode. R Where d% represents the wind curtailment capacity and d% represents the load reduction percentage.

[0046] When the wind and storage combined system provides standby capacity, dynamic load shedding is achieved by adding a SOC adjustment coefficient k, as follows:

[0047] P R '(t)=(1-kα)P max (t) (10)

[0048] Where P R '(t) represents the amount of wind curtailment after energy storage is configured, i.e., the reserve capacity of the DFIG in the wind-storage system. A Logistic regression function is introduced, with SOC as the independent variable, adaptive factors n and P0 as parameters, and k as the dependent variable, to construct an adjustment coefficient model, as shown in the following equation:

[0049]

[0050] In the formula, n is an adaptive factor used to measure the rate of change of the curve, and P0 is the initial value; P max For the maximum output power of energy storage, S SOC This represents the SOC value at the current moment.

[0051] Set the SOC charge / discharge control coefficient as follows:

[0052] Partition the SOC, where SOC min SOC low SOC high SOC max Let m represent the minimum, slightly lower, slightly higher, and maximum values ​​of the energy storage SOC, respectively, and set to 0.1, 0.45, 0.55, and 0.9. A charge / discharge control coefficient curve m is constructed based on this curve. c and discharge control coefficient curve m d The curve consists of two lines. When the SOC state is good, i.e., S... SOC When ∈(0.45,0.55) m Taking the maximum value, the charge / discharge control coefficient constraint is as follows:

[0053]

[0054]

[0055] In the formula, K max This represents the maximum droop coefficient for energy storage.

[0056] After frequency tuning is completed, determine If a polarity change occurs, set the DFID additional power to 0 and perform supercapacitor frequency modulation, specifically:

[0057] During the inertial response phase, the additional power of the DFIG gradually decreases until it exits frequency modulation as it approaches the lowest frequency point. Simultaneously, the stored energy suddenly increases its power to prevent a secondary frequency drop caused by the power surge when the DFIG exits frequency modulation. When the frequency difference change rate... When the polarity changes, the primary frequency modulation stage begins, and the dominant frequency modulation shifts from the DFIG to the supercapacitor. At this point, the supercapacitor outputs ΔP. E for:

[0058] ΔP E =-K b Δf (14)

[0059] In the formula, K b K at the moment of polarity change of the rate of change of frequency difference p same.

[0060] If the wind speed zone is medium, the DFIG uses rotor overspeed load reduction control to provide backup capacity, and combines it with virtual inertial control frequency modulation. The supercapacitor uses virtual droop control frequency modulation, specifically:

[0061] In the medium wind speed region, the DFIG simulates the grid inertial process using virtual inertial control and employs overspeed load shedding control to provide frequency regulation reserves. The supercapacitor energy storage simulates the primary frequency regulation process using virtual droop control, controlling its active power output based on the frequency deviation according to the droop characteristics. The combined power ΔP of the virtual inertial control and droop control participating in frequency regulation is:

[0062]

[0063] In the formula, K d K represents the virtual inertial control coefficient. p Let m be the droop control coefficient and m be the SOC charging / discharging control coefficient. A three-input, dual-output fuzzy logic controller is established. The three state variables—wind speed v, frequency difference rate of change dΔf / dt, and frequency deviation Δf—are used as inputs to the fuzzy controller. After defuzzification, the fuzzy control output K is obtained. d and K p ,in:

[0064] The wind speed range is set to 6 m / s to 13.5 m / s, the frequency difference rate range is set to -0.6 Hz / s to 0 Hz / s, the frequency deviation range is set to -1 Hz to -0.033 Hz, and the virtual inertia coefficient and droop coefficient range to 5 to 20. The fuzzy subsets of the output quantities include very small, small, relatively small, medium, relatively large, large, and very large. The fuzzy logic rules are as follows:

[0065] When the wind speed is between 7.5 and 9.75 m / s, the virtual inertia coefficient K is... d Take the smallest value and K dIt decreases as the rate of change of frequency difference increases, while the droop coefficient K... p K increases with the increase of the rate of change of frequency deviation, and K increases with the continuous increase of the frequency deviation. d K p It also increases;

[0066] When the wind speed is between 8.5 and 10.9 m / s, K d K decreases as the rate of change of frequency difference increases. p K increases with the increase of the rate of change of frequency difference. When both the rate of change of frequency difference and the frequency deviation are small, K... d As the value of K increases, p When the value is too small, and the rate of change of frequency difference is small while the frequency deviation is large, K... d K p The value of K should be relatively large. When the rate of change of frequency difference is large and the frequency deviation is small, K d K p The value of K should be relatively small. When both the rate of change of frequency difference and the frequency deviation are large, K... d The value of K should be relatively small. p The value should be as large as possible;

[0067] When the wind speed is between 9.75 and 12 m / s, and the rate of change of frequency difference is small, K d The value of K can be appropriately increased. d It decreases as the rate of change of frequency difference increases; when the wind speed is too high, K... d Do not take excessively large values, and at the same time, K p The value of K is positively correlated with both the frequency deviation and the rate of frequency change. When the rate of frequency deviation change is large, the wind speed is greater, and K... d The value of K should be relatively small. p Increase.

[0068] According to specific embodiments provided by the present invention, the following technical effects are disclosed: The wind-storage integrated system primary frequency regulation method based on multivariable fuzzy logic control provided by the present invention realizes the power allocation of wind power and energy storage participating in primary frequency regulation, fully leverages the frequency regulation advantages of DFIG and energy storage, improves the frequency stability of power systems with high wind power penetration, designs a multivariable fuzzy logic controller considering three factors: wind speed, frequency difference rate of change, and frequency deviation, realizes the rational allocation of wind and storage power, dynamically sets the DFIG frequency regulation reserve capacity according to the state of charge (SOC) of the energy storage system, improves the primary frequency regulation capability of the wind-storage integrated system under different wind conditions, and verifies the effectiveness of the proposed strategy on the MATLAB / Simulink simulation platform. The results show that the proposed method improves the primary frequency regulation capability of the wind-storage integrated system, realizes the rational allocation of wind and storage frequency regulation power, reduces the risk of wind curtailment and wind turbine output oscillation, and improves the frequency stability of the system. Attached Figure Description

[0069] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0070] Figure 1 This is a schematic diagram of the system frequency response model structure;

[0071] Figure 2 This is a schematic diagram of a virtual inertial control structure.

[0072] Figure 3a This is a frequency response control diagram;

[0073] Figure 3b This is a power-speed variation curve;

[0074] Figure 4 A simplified equivalent model diagram of a supercapacitor;

[0075] Figure 5 Wind speed zone map;

[0076] Figure 6 This is a diagram illustrating the structure of a fuzzy logic controller.

[0077] Figure 7 The fuzzy inference results are shown in the figure when v = 9.75 m / s;

[0078] Figure 8 A flowchart for wind and energy storage joint frequency regulation;

[0079] Figure 9 This is a topology diagram of a wind-storage system.

[0080] Figure 10 A diagram showing the reserve capacity for frequency modulation.

[0081] Figure 11 For S SOC Comparison of simulation results for a sudden load increase when the load ratio is 0.5;

[0082] Figure 12 For S SOC Comparison of simulation results for a sudden load increase when the load ratio is 0.3;

[0083] Figure 13 For S SOC SOC change curve when = 0.5;

[0084] Figure 14 For S SOCSOC change curve when = 0.3;

[0085] Figure 15 For S SOC Comparison of simulation results for sudden load reduction when the load ratio is 0.5;

[0086] Figure 16a This is a membership function graph for wind speed;

[0087] Figure 16b This is a membership function graph of the rate of change of frequency difference;

[0088] Figure 16c Membership function graph for frequency deviation;

[0089] Figure 16d The membership function graph of virtual inertia coefficient / virtual droop coefficient;

[0090] Figure 17a The diagram shows the fuzzy inference results when v = 6 m / s;

[0091] Figure 17b The fuzzy inference results are shown when v = 13.5 m / s.

[0092] Figure 18 A comparison chart of simulation results for low wind speed areas;

[0093] Figure 19 A comparison chart of simulation results for high wind speed areas;

[0094] Figure 20 A comparison of simulation results for the critical low wind speed zone;

[0095] Figure 21 The figure shows a comparison of simulation results under source load fluctuations. Detailed Implementation

[0096] 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 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.

[0097] The purpose of this invention is to provide a primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control. This method can realize the power allocation of wind power and energy storage participating in primary frequency regulation, fully leverage the frequency regulation advantages of DFIG and energy storage, and improve the frequency stability of power systems with high wind power penetration.

[0098] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0099] The primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control provided in this invention includes the following steps:

[0100] Step 1: Establish a wind-storage integrated system model, specifically as follows:

[0101] Variable-speed wind turbines can capture wind energy to the maximum extent and are 38% more efficient than constant-speed systems. They can also participate in primary frequency regulation by changing the rotor speed. Supercapacitor energy storage has a long cycle life, fast response speed, wide voltage and operating range, and combines the energy storage characteristics of batteries with the power characteristics of capacitors, giving it unique advantages in participating in primary frequency regulation. Therefore, this paper takes a wind-storage system consisting of a doubly-fed induction generator (DFIG) wind farm and supercapacitor energy storage as an example for research.

[0102] Among them, the system frequency response model is established:

[0103] When load fluctuates, the system's active power becomes unbalanced, causing frequency changes. Various energy sources, such as thermal power, wind power, and energy storage, adjust their own active power output to participate in primary frequency regulation. The system frequency response model is as follows: Figure 1 As shown, ΔP W For the increase in wind power, ΔP E For the power increment of the energy storage system, ΔP L Let ΔP be the load change. G For the active power increment of other energy units outside the wind-storage system participating in primary frequency regulation (equivalent to thermal power units), obtain the wind power increment ΔP. W Energy storage system power increment ΔP E Load change ΔP L The active power increment ΔP of other energy units besides the wind and storage system participating in primary frequency regulation G ,in,

[0104]

[0105] In the formula, R G T is the prime mover adjustment coefficient. G T is the time constant of the speed controller. ch T is the turbine time constant. R F is the reheat time constant. H Given the reheater gain, the system frequency response model transfer function is obtained as follows:

[0106]

[0107] In the formula, H, D, and p are the inertial constant, damping constant, and wind power penetration rate, respectively.

[0108] Establish the DFIG model:

[0109] The mechanical power output of the fan is obtained as follows:

[0110]

[0111] In the formula, ρ is the air density, A is the swept area of ​​the wind turbine blade, and R... w Where is the radius of the blade, v is the wind speed, and C is the wind speed. p Where λ is the wind energy utilization coefficient, β is the tip speed ratio, and C is the blade pitch angle. p (λ,β) is a nonlinear function of λ and β. The DFIG model is established as follows:

[0112]

[0113] In the formula, γ is an introduced intermediate variable, and ω is the rotor speed of the wind turbine. To enable the wind turbine to support frequency, the wind turbine operates in a reduced-load mode, and virtual inertial frequency control based on the frequency change rate is added to reduce the speed change during frequency regulation. Its structure diagram is shown below. Figure 2 As shown in Figure 3, the DFIG control response principle and power-speed variation curve are presented. d For the additional power of virtual inertial control, P w It provides power to the DFIG. When the DFIG operates in overspeed unloading mode, it operates stably with the unloading power at point A. At t A As the system frequency decreases, the DFIG responds with frequency fluctuations, and the electromagnetic power changes from point A to point B. The rotor speed decreases, releasing rotor kinetic energy. During the inertial response, the additional power of the DFIG gradually decreases to zero as the rate of frequency change decreases. The electromagnetic power smoothly transitions from point B to point O, where point O is the same operating point as the initial unloaded operating point A. At point O, the rotor speed begins to recover, requiring the absorption of energy from the grid, and the DFIG's output active power decreases. Throughout the entire virtual inertial control process, the electromagnetic power varies along the ABOCA curve, sharing the power surge of the system by releasing rotor kinetic energy.

[0114] Establish a supercapacitor model:

[0115] Construct an equivalent circuit model of a supercapacitor, including an ideal capacitor C and a series equivalent resistance R. es and the parallel equivalent resistance R that reflects the long-term static changes of the supercapacitor eq The frequency modulation time is very short, so R is negligible. eq Impact, simplified equivalent model of supercapacitor, such as Figure 4 As shown, the supercapacitor voltage is:

[0116]

[0117] Supercapacitor energy storage capacity E scss The remaining charge E at time t C (t) and state of charge (SOC) k (t) is:

[0118]

[0119] In the formula, U max U min These are the highest and lowest operating voltages of the supercapacitor, respectively, in V and P. C (t) represents the supercapacitor's charging and discharging power at time t, in MW, η C E represents the charging and discharging efficiency of supercapacitors. scss The total capacity of the supercapacitor is expressed in MWh. The power storage and SOC constraints are as follows:

[0120]

[0121] In the formula, P rmin P rmax SOC min SOC max These represent the minimum output power, maximum output power, minimum and maximum SOC of the supercapacitor, respectively. Since the voltage of a single supercapacitor is relatively low, multiple supercapacitors can be connected in series and parallel to form a large-capacity supercapacitor array. Let n be the number of supercapacitors connected in series. s The number of parallel connections is n p The equivalent internal resistance and equivalent capacitance are then:

[0122]

[0123] like Figure 5 As shown, step 2: divide the wind speed areas, specifically as follows:

[0124] The minimum cut-in wind speed of DFIG is v in The maximum cut-out wind speed is v out The operating speed is typically 0.7–1.2 pu. Wind speed is divided into three zones: low, medium, and high, and two critical wind speed zones. Among them:

[0125] Low wind speed area (v in -v1): In this wind speed range, the rotor speed of the wind turbine is constant at 0.7 pu. At this time, the DFIG does not have frequency modulation capability, and only the supercapacitor participates in frequency modulation.

[0126] Critical low wind speed zone (v lo -v m1): In the frequency regulation strategy of this wind speed zone, a first-order inertial element is added. When the wind speed changes frequently in the low and medium wind speed zones, the active power increment of DFIG participating in frequency regulation changes slowly until it exits frequency regulation, thus avoiding the problem of wind turbine output oscillation.

[0127] Medium wind speed zone [v1-v2): consists of the maximum power point tracking (MPPT) zone and the constant speed zone.

[0128] Critical high wind speed zone (v m2 -v h ): Frequency modulation strategy is the same as in the critical low wind speed zone.

[0129] High wind speed area [v2-v out Within this wind speed range, the wind turbine rotor speed is maintained at 1.2 pu, and pitch angle control is introduced to keep the wind turbine output power at the rated power without load shedding to reduce wind curtailment.

[0130] When the frequency rises, the active power output can be responded to by rotor overspeed control or pitch angle control to regulate frequency fluctuations. At the same time, the energy storage system is charged from the grid to participate in primary frequency regulation. Compared with the frequency decline, its frequency regulation process is easier to achieve load reduction operation. At low wind speeds, the wind turbine rotor kinetic energy is insufficient, and it is impossible to reduce the active power output of the wind turbine by rotor overspeed. The energy storage system completes the frequency regulation task. At medium wind speeds, due to the slow response speed of pitch angle control and the easy wear of mechanical parts, DFIG adopts rotor overspeed combined with virtual inertial control, and the energy storage system adopts virtual droop control. The high wind speed operation time accounts for a small proportion. At high wind speeds, the DFIG rotor speed is maintained at 1.2 pu, and the active power output is reduced by pitch angle control. At the same time, the energy storage system also shares part of the frequency regulation task.

[0131] When the frequency decreases, the participation of the wind-storage combined system in power system frequency regulation can be simply divided into two stages: inertial response and primary frequency regulation. In the inertial response stage, the virtual inertial control of the DFIG plays a major role; in the primary frequency regulation stage, the virtual droop control of the energy storage plays a major role. The DFIG possesses different rotational speeds and kinetic energy under different wind speed scenarios, resulting in varying frequency regulation capabilities. Therefore, wind speed regions must be divided according to the operating status and frequency regulation capabilities of the wind turbines under different wind speed scenarios. Based on this, a multivariable fuzzy logic controller is designed to obtain dynamic virtual inertial coefficients and droop coefficients, thereby realizing the power allocation for wind power and energy storage participating in primary frequency regulation.

[0132] like Figure 8 As shown, step 3: Establish a multivariate fuzzy logic control method, specifically as follows:

[0133] Obtain the system frequency and calculate the frequency deviation Δf based on the system frequency, and determine whether Δf meets the condition |Δf|≥0.033Hz;

[0134] If it does not meet the requirements, determine whether the energy storage SOC is ideal. If it is not ideal, perform SOC recovery and set the SOC charge and discharge control coefficient. If it is ideal, recalculate Δf and make a judgment.

[0135] If the conditions are met, the wind speed is checked, and it is determined whether the area is a medium wind speed zone.

[0136] If the wind speed zone is medium, the DFIG uses rotor overspeed unloading control to provide backup capacity, and combines it with virtual inertial control frequency modulation. The supercapacitor uses virtual droop control frequency modulation. After frequency modulation is completed, a judgment is made. If a polarity change occurs, set the DFIG additional power to 0 and perform supercapacitor frequency modulation. If no change occurs, return to the previous step to perform DFIG and supercapacitor frequency modulation.

[0137] If it is not in the medium wind speed zone, determine whether Δf is greater than 0. If it is less than 0, then perform supercapacitor frequency modulation. If it is greater than 0, determine whether it is in the high wind speed zone. If it is in the high wind speed zone, then perform DFIG pitch angle control and energy storage virtual inertial integrated control. If it is not in the high wind speed zone, then perform supercapacitor frequency modulation.

[0138] In step 3, the DFIG employs rotor overspeed unloading control to provide backup capacity, specifically as follows:

[0139] In the process of wind-storage combined systems participating in power system frequency regulation, the DFIG (Diverterless Geological Injection Generator) uses rotor overspeed load shedding control to provide reserve capacity, while the supercapacitor also shares some reserve capacity to reduce wind curtailment. When the State of Charge (SOC) is low, energy storage provides less reserve capacity, while wind power provides more; when the SOC is high, energy storage provides more reserve capacity, while wind power provides less, thereby achieving the goal of reducing wind curtailment.

[0140] When the supercapacitor does not provide backup, the backup capacity provided by the DFIG will be used as the total backup capacity required by the wind-storage system, as follows:

[0141] P R (t)=(1-α)P max (t)=d%P max (t) (9)

[0142] In the formula, α is the wind power utilization rate, and P max P represents the maximum power in MPPT operation mode. R Where d% represents the wind curtailment capacity and d% represents the load reduction percentage.

[0143] When the wind and storage combined system provides standby capacity, dynamic load shedding is achieved by adding a SOC adjustment coefficient k, as follows:

[0144] P R '(t)=(1-kα)P max(t) (10)

[0145] Where P R '(t) represents the amount of wind curtailment after energy storage is configured, i.e., the reserve capacity of the DFIG in the wind-storage system. A Logistic regression function is introduced, with SOC as the independent variable, adaptive factors n and P0 as parameters, and k as the dependent variable, to construct an adjustment coefficient model, as shown in the following equation:

[0146]

[0147] In the formula, n is an adaptive factor used to measure the rate of change of the curve, and P0 is the initial value; P max For the maximum output power of energy storage, S SOC This represents the SOC value at the current moment.

[0148] If the result is not ideal, then SOC recovery will be performed, specifically as follows:

[0149] When the SOC state of the energy storage system is not ideal, the output of the energy storage system should be adjusted in real time according to the SOC state during the frequency regulation process.

[0150] Partition the SOC, where SOC min SOC low SOC high SOC max Let m represent the minimum, slightly lower, slightly higher, and maximum values ​​of the energy storage SOC, respectively, and set to 0.1, 0.45, 0.55, and 0.9. A charge / discharge control coefficient curve m is constructed based on this curve. c and discharge control coefficient curve m d The curve consists of two lines. When the SOC state is good, i.e., S... SOC When m ∈ (0.45, 0.55), the charge / discharge control coefficient constraint is as follows:

[0151]

[0152]

[0153] In the formula, K max This represents the maximum droop coefficient for energy storage.

[0154] After frequency tuning is completed, determine If a polarity change occurs, set the DFID additional power to 0 and perform supercapacitor frequency modulation, specifically:

[0155] During the inertial response phase, the additional power of the DFIG gradually decreases until it exits frequency modulation as it approaches the lowest frequency point. Simultaneously, the stored energy suddenly increases its power to prevent a secondary frequency drop caused by the power surge when the DFIG exits frequency modulation. When the frequency difference change rate... When the polarity changes, the primary frequency modulation stage begins, and the dominant frequency modulation shifts from the DFIG to the supercapacitor. At this point, the supercapacitor outputs ΔP. E for:

[0156] ΔP E =-K b Δf (14)

[0157] In the formula, K b K at the moment of polarity change of the rate of change of frequency difference p same.

[0158] If the wind speed zone is medium, the DFIG uses rotor overspeed load reduction control to provide backup capacity, and combines it with virtual inertial control frequency modulation. The supercapacitor uses virtual droop control frequency modulation, specifically:

[0159] In the medium wind speed region, the DFIG simulates the grid inertial process using virtual inertial control and employs overspeed load shedding control to provide frequency regulation reserves. The supercapacitor energy storage simulates the primary frequency regulation process using virtual droop control, controlling its active power output based on the frequency deviation according to the droop characteristics. The combined power ΔP of the virtual inertial control and droop control participating in frequency regulation is:

[0160]

[0161] In the formula, K d K represents the virtual inertial control coefficient. p Let denoted as droop control coefficient and m as SOC charging / discharging control coefficient. The ability of DFIG to participate in primary frequency regulation varies under different wind conditions. However, it is difficult to establish an accurate mathematical model between the control coefficient values ​​and wind conditions, frequency difference rate of change, and frequency deviation. Fuzzy logic control has advantages such as ease of use, strong robustness, and high fault tolerance. To implement the proposed frequency regulation method, a three-input, dual-output fuzzy logic controller is established. The wind speed v, frequency difference rate of change dΔf / dt, and frequency deviation Δf are used as inputs to the fuzzy controller. After defuzzification, the fuzzy control output K is obtained. d and K p ,like Figure 6 As shown, a fuzzy logic controller typically consists of four parts: input fuzzification, rule base, fuzzy inference, and output defuzzification.

[0162] The wind speed range is set to [6, 13.5] m / s, the frequency difference rate of change range is set to [-0.6, 0] Hz / s, the frequency deviation range is set to [-1, -0.033] Hz, and the virtual inertia coefficient and droop coefficient range are both set to [5, 20]. The fuzzy subsets of the output are all {VS (very small), S (small), RS (small), M (medium), RB (large), B (large), VB (very large)}, and their membership function curves are shown in Figure 16. The fuzzy logic rules are as follows:

[0163] (1) When the wind speed is low (7.5~9.75m / s), the frequency regulation capability of the DFIG is relatively weak. Excessive changes in rotor speed can easily cause the doubly-fed wind turbine generator to disconnect from the grid. Therefore, regardless of how the frequency difference change rate and frequency deviation change, the virtual inertia coefficient K d Both should be taken as small as possible and K d The frequency difference decreases as the rate of change of frequency difference increases, while supercapacitor energy storage should undertake the main frequency regulation task. p It increases with the rate of change of frequency deviation. As the frequency deviation continues to increase, K... d K p It should also be increased appropriately.

[0164] (2) When the wind speed is moderate (8.5~10.9m / s), the frequency modulation capability of DFIG is the strongest. d K decreases as the rate of change of frequency difference increases. p It increases with the increase of the rate of change of frequency deviation. When both the rate of change of frequency deviation and the frequency deviation are small, K... d The value of K can be appropriately increased. p The value should be relatively small; when the rate of change of frequency difference is small and the frequency deviation is large, K d K p The value of K should be relatively large; when the rate of change of frequency difference is large and the frequency deviation is small, K should be... d K p The value of K should be relatively small; when both the rate of change of frequency difference and the frequency deviation are large, K... d The value of K should be relatively small. p The value should be as large as possible.

[0165] (3) When the wind speed is high (9.75~12m / s), the frequency modulation capability of the DFIG is weak, and its output additional power capability is small. When the frequency difference change rate is small, K d The value of K can be appropriately increased. d It decreases as the rate of change of frequency difference increases. However, when the wind speed is too high, K... d The value of K should not be too large, and K should also be... p The value of K is positively correlated with both the frequency deviation and the rate of frequency change; when the rate of frequency deviation change is large, to avoid excessive kinetic energy release, the higher the wind speed, the greater the value of K. d The value of K should be relatively small.p It can be increased appropriately;

[0166] The established fuzzy rule table is shown in Table 1. When the wind speed is moderate, dΔf / dt and Δf are used as the X and Y axes, respectively, and K is used as the Y axis. d K p A three-dimensional relationship diagram along the Z-axis is shown below. Figure 7 As shown in Figure 17, the three-dimensional relationship diagrams for lower and higher wind speeds are respectively.

[0167] Table 1 Fuzzy Logic Inference Table

[0168]

[0169]

[0170]

[0171] Wind-storage combined system topology as follows Figure 9 As shown. To verify the feasibility and effectiveness of the proposed wind-storage combined system participating in primary frequency regulation strategy, this invention builds a simulation model on the MATLAB / Simulink platform. The parameters are standardized per unit with a unit capacity of 100MW, the load is taken as 100MW, the wind power accounts for 20%, the rated power of the energy storage system is configured as 10% of the wind farm, i.e. 2MW / 2MW·1min, 2.7V / 3000F supercapacitors are selected, and 278 series and 53 parallel supercapacitors are used to form the energy storage device required for primary frequency regulation. The total capacitance of the supercapacitors is 572F, the reference frequency is 50Hz, and the upper and lower limits of the primary frequency regulation dead zone are set to ±0.033Hz.

[0172] A simulation analysis was conducted to reduce wind curtailment due to dynamic frequency regulation reserve capacity: A long-term reserve of 10% of the rated power (2MW) in the DFIG would result in significant wind curtailment; while maximum power point tracking (MPPT) operation maximizes wind energy utilization, it fails to meet the requirement of primary frequency regulation capability for renewable energy power plants. For example... Figure 10 As shown, using the DFIG dynamic load shedding strategy proposed in this paper, the reserved frequency regulation reserve capacity of the wind turbine is 0.8569MW and 3.132MW when the state of charge (SOC) of the supercapacitor energy storage is 0.5 and 0.3, respectively. The smaller the SOC of the energy storage system, the worse its ability to participate in primary frequency regulation, and the larger the frequency regulation reserve capacity reserved by the DFIG. After the SOC of the energy storage system exceeds 0.36, the reserve capacity reserved by dynamic load shedding is less than the reserve capacity of a constant load shedding ratio of 10%. When the wind-storage system does not participate in primary frequency regulation, it will restore the SOC of the supercapacitor energy storage, making SOC... SOC Maintain a good condition (S) SOC∈(0.45,0.55)), therefore, the dynamic load shedding strategy can keep the DFIG frequency regulation reserve capacity around 0.85MW, and the remaining frequency regulation task is completed by the energy storage system, which greatly reduces the amount of wind curtailment;

[0173] Simulation analysis of wind-storage combined frequency regulation in the medium wind speed region: When the wind speed is in the medium wind speed region, the effectiveness and superiority of the proposed wind-storage combined system participating in primary frequency regulation strategy are verified by simulation. Simulation results in the low wind speed and high wind speed regions are as follows. Figure 18 and Figure 19 As shown, with a simulated wind speed of 10 m / s, and a sudden load increase of 0.03 pu in the system at 10 s, simulations were performed comparing the wind-storage joint frequency regulation method with the proposed method under constant K method, without additional control, only wind turbine frequency regulation (virtual inertial control and overspeed load reduction), and with initial SOCs of 0.5 and 0.3. The rotational speed and active power in the figure are per-unit values. SOC Simulation results of sudden load increase are as follows Figure 11 , 12 As shown, the SOC variation curve is as follows: Figure 13 , 14 As shown;

[0174] Depend on Figure 11 , 12 It can be seen that when the load suddenly increases by 0.03 pu at 10s, the system frequency drops rapidly, with the highest frequency drop rate and drop depth without additional control. While the frequency drop rate and maximum frequency deviation are improved when only the fan participates in primary frequency regulation, the excessive release of DFIG rotor kinetic energy due to the lack of energy storage participation causes the rotor speed to drop excessively, even below the minimum value of 0.7 pu. Both the constant K method and the method presented in this paper show significant improvements in the frequency drop rate and maximum frequency deviation. Compared with the constant K method, the method presented in this paper, when the initial S... SOC When the value is 0.5, the lowest frequency point increases by 0.0135Hz. SOC When the value is 0.3, the lowest frequency point is increased by 0.0201Hz; compared with the method without additional control, when the initial S SOC When the value is 0.5, the lowest frequency point increases by 0.1694 Hz. SOC When the initial state of charge is 0.3, the lowest frequency point increases by 0.1244 Hz. Therefore, the worse the initial state of charge, the more pronounced the advantage of the wind-storage frequency regulation strategy proposed in this paper becomes, because the fuzzy logic controller can adjust the control coefficients in real time based on the rate of change of frequency difference and frequency deviation. Figure 11 and Figure 12 As shown in Figure (b), the rotor speed of the wind-storage combined frequency regulation strategy proposed in this invention is significantly higher than that of the rotor speed when only the wind turbine participates in frequency regulation, and slightly lower than that of the constant K method, which can fully release the rotor kinetic energy without exceeding the limit; Figure 11 and Figure 12As shown in Figure (c), when the DFIG exits frequency modulation, the supercapacitor's energy storage can compensate for the power loss during DFIG exit, preventing a secondary frequency drop during the modulation process; from Figure 11 and Figure 12 As shown in Figure (d), during the speed recovery period, the difference between the mechanical power and the electromagnetic power output during unloaded operation is used entirely for rotor acceleration and does not participate in primary frequency regulation.

[0175] Depend on Figure 13 , 14 It can be seen that when the initial S SOC When the initial SOC is 0.5, and the simulation time is less than 30 seconds, the two methods have almost no difference in their impact on SOC, and the active power increment of the energy storage system is also very similar. After 30 seconds, due to the effect of constraint control, the SOC change curve using the method of this invention is smoother; when the initial SOC is less than 0.5, the SOC change curve using the method of this invention is smoother. SOC When the initial SOC is 0.3, the state of charge (SOC) without constraint control is significantly worse than that of the method presented in this paper, and it prematurely exits frequency modulation at 36.7s due to reaching the lower limit of SOC. Therefore, the initial SOC... SOC The smaller the value, the faster the energy storage will reach the lower limit of SOC without constraint control, causing the energy storage system to exit the frequency regulation prematurely and fail to complete the frequency regulation task.

[0176] Similarly, with a simulated wind speed of 10 m / s, and a sudden load reduction of 0.03 pu at 10 s, a simulation comparison was conducted between the proposed method and a system without additional control, relying solely on fan frequency regulation (a combination of virtual inertial control and overspeed load reduction), with an initial SOC of 0.5. The simulation results are as follows: Figure 15 As shown

[0177] Depend on Figure 15 It can be seen that when the load is suddenly reduced by 0.03 pu at 10 seconds, the system frequency rises rapidly. Without additional control, the rate and depth of frequency rise are both the largest. While the rate of frequency rise and maximum frequency deviation are improved when only the fan participates in primary frequency regulation, the kinetic energy absorbed by the increased rotor speed exceeds the limit of 1.2 pu. The method presented in this paper has the smallest rate of frequency rise and maximum frequency deviation. Compared with the method without additional control, the highest frequency point is reduced by 0.1511 Hz; compared with the method with only fan frequency regulation, the highest frequency point is reduced by 0.129 Hz. Figure 15 As shown in Figure (b), the rotor speed of the method of the present invention is significantly lower than that of the rotor speed when only the fan participates in frequency regulation. It can fully absorb rotor kinetic energy without exceeding the limit, and the disturbance to the system during the recovery process is also smaller.

[0178] Detailed parameters of the simulation system are shown in Tables 2 and 3:

[0179] Table 1 Power Parameters

[0180]

[0181] Table 2 Parameters of Wind Turbine and Energy Storage System

[0182]

[0183] Simulation analysis of low and high wind speed areas under sudden load increase of the system is as follows:

[0184] In low wind speed areas, due to the weak wind, the DFIG can only operate at the minimum speed of 0.7 pu and does not have primary frequency regulation capability. It cannot increase or decrease the active power output of the wind turbine by controlling the rotor speed. At this time, the energy storage system can independently complete all frequency regulation tasks, so that the wind-storage combined power generation system can also participate in the primary frequency regulation of the system in low wind speed areas.

[0185] In high wind speed areas, due to the strong winds, the wind turbine rotor speed is maintained at 1.2 pu, and pitch angle control is introduced to keep the wind turbine output power at the rated power without load shedding. In this case, the energy storage system alone completes all frequency regulation tasks. When the wind speed is in both low and high wind speed areas, only one frequency regulation is performed using the energy storage system. The frequency regulation process is similar in both cases, so only the low wind speed area is analyzed. The simulated wind speed is 5 m / s, the initial SOC is 0.5, and the system experiences a sudden load increase of 0.03 pu at 10 s. The simulation results are as follows: Figure 18 As shown;

[0186] Figure 18 The system frequency curves were compared in low-wind-speed areas with and without additional control using the method of this invention. Simulation results show that the minimum frequency of the proposed wind-storage joint frequency regulation strategy increased by 0.1736 Hz, and the rate of frequency decline also slowed. The active power increment of the energy storage system was slightly higher than the simulation results in the medium-wind-speed area. This is because in the medium-wind-speed area, the DFIG participates in primary frequency regulation, sharing some of the frequency regulation task, resulting in a relatively smaller active power increment of the energy storage system and a slightly higher minimum frequency than in the low-wind-speed area. Therefore, although the DFIG cannot participate in frequency regulation in low and high-wind-speed areas, the introduction of the energy storage system ensures that the wind farm can meet the primary frequency regulation requirements regardless of its operating conditions.

[0187] Simulation analysis of low and high wind speed areas during sudden load reduction of the system is as follows:

[0188] In low wind speed areas, the strategy is the same as for sudden load increases in the system: the energy storage system must independently complete all frequency regulation tasks. In high wind speed areas, the DFIG rotor speed is maintained at 1.2 pu, and pitch angle control is used to reduce the active power output of the wind turbine. Simultaneously, the energy storage system also shares some of the frequency regulation tasks. Therefore, only the situation in high wind speed areas is analyzed. The simulated wind speed is 15 m / s, the initial SOC is 0.5, and the system suddenly reduces the load by 0.03 pu at 10 s. The simulation results are as follows: Figure 19 As shown

[0189] Figure 19 The system frequency curves were compared in high wind speed areas with and without additional control using the proposed method. Simulation results show that the frequency peak using the proposed method is reduced by 0.2 Hz, and the frequency rise rate is also significantly decreased. Under the wind-storage joint frequency regulation strategy, the DFIG reduces the output active power by increasing the pitch angle, with a pitch angle change of 0.653°; simultaneously, the energy storage system absorbs active power from the grid to participate in primary frequency regulation. Therefore, the frequency regulation strategy proposed in this invention can meet the system's primary frequency regulation requirements in both low and high wind speed areas.

[0190] Frequent wind speed variations within a given wind speed range can cause frequent switching of the frequency regulation strategy, easily leading to oscillations in turbine output. To address this, a first-order inertial element is added to the frequency regulation strategy in these two critical wind speed zones. This causes the active power increment of the turbine participating in frequency regulation to change slowly until it withdraws, thus avoiding the problem of turbine output oscillations. Taking the critical low wind speed zone as an example, with an initial SOC of 0.5, a sudden load increase of 0.03 pu at 10s, an initial wind speed of 6.2 m / s, and a sudden change in wind speed to a random speed between 5.5 m / s and 6.5 m / s at 11s, the simulation results are as follows: Figure 20 As shown;

[0191] At 11 seconds, the wind speed becomes a random wind speed within the range of 5.5 m / s to 6.5 m / s. Without an additional first-order inertial element, the DFIG output oscillates with frequent changes in wind speed zones until the frequency begins to rise and then levels off. Using the method of this invention, for random wind speeds within the critical low-wind-speed zone, if there is a sudden transition from a medium-wind-speed zone to a low-wind-speed zone, the DFIG output will not change abruptly but will slowly decrease to zero. Subsequently, as long as the wind speed remains within this range, the DFIG will not participate in any frequency modulation until the wind speed exceeds 6.5 m / s, at which point the frequency modulation strategy will switch.

[0192] To verify the effectiveness of the proposed frequency regulation strategy under random wind speed and random load fluctuations, a simulation analysis was conducted on the frequency regulation situation when the source and load fluctuate simultaneously.

[0193] from Figure 21 It can be seen that when wind speed and load fluctuate randomly, the frequency deviation and frequency change rate of the method of this invention are greatly improved. The DFIG and energy storage system can select appropriate control coefficients according to the changing wind speed, frequency deviation and frequency change rate, and respond to frequency fluctuations by changing the active power output, thereby mitigating the frequency change rate and improving the highest / lowest frequency points.

[0194] This method proposes a dynamic frequency regulation reserve capacity that takes into account the state of charge of the energy storage system. Based on this, overspeed load shedding control is applied to the DFIG to achieve primary frequency regulation of the wind-storage system while avoiding excessive load shedding, thereby reducing wind curtailment.

[0195] The proposed method uses multivariable fuzzy logic control to determine the virtual inertia coefficient and virtual droop coefficient based on wind speed, frequency difference rate of change, and frequency deviation, thereby achieving a reasonable allocation of active power output between the DFIG and the energy storage system.

[0196] Compared with three scenarios—no additional control, wind turbine-only frequency regulation, and the fixed-K method—the proposed wind-storage combined primary frequency regulation control strategy effectively mitigates the system frequency change rate, improves the minimum and maximum frequency points, and prevents secondary frequency drops or rises. In the numerical examples, compared with the fixed-K method and no additional control, under the conditions of an initial SSOC of 0.5 and a sudden load increase of 0.03 pu, the minimum frequency point using the proposed method increased by 0.0135 Hz and 0.1694 Hz, respectively. Compared with wind turbine-only frequency regulation and no additional control, under the conditions of an initial SSOC of 0.5 and a sudden load decrease of 0.03 pu, the maximum frequency point using the proposed method decreased by 0.129 Hz and 0.1511 Hz, respectively.

[0197] This invention provides a primary frequency regulation method for wind-storage integrated systems based on multivariable fuzzy logic control. This method enables power allocation for wind power and energy storage participating in primary frequency regulation, fully leveraging the frequency regulation advantages of DFIG and energy storage, and improving the frequency stability of power systems with high wind power penetration. The method designs a multivariable fuzzy logic controller considering three factors: wind speed, frequency difference rate of change, and frequency deviation, to achieve reasonable allocation of wind and storage power. It dynamically sets the DFIG frequency regulation reserve capacity based on the state of charge (SOC) of the energy storage system, enhancing the primary frequency regulation capability of the wind-storage integrated system under different wind conditions. The effectiveness of the proposed strategy is verified on the MATLAB / Simulink simulation platform. The results show that the proposed method improves the primary frequency regulation capability of the wind-storage integrated system while achieving reasonable allocation of wind and storage frequency regulation power, reducing wind curtailment and wind turbine output oscillation risks, and improving the frequency stability of the system.

[0198] This document uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. Furthermore, those skilled in the art will recognize that, based on the ideas of the present invention, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for primary frequency regulation of a wind storage combined system based on multivariable fuzzy logic control, characterized in that, Includes the following steps: Step 1: Establish a wind-storage integrated system model; Step 2: Divide the wind speed zones; Step 3: Establish a multivariable fuzzy logic control method; In step 3, a multivariate fuzzy logic control method is established, specifically as follows: Obtain the system frequency and calculate the frequency deviation based on the system frequency. ,judge Does it meet the requirements? ; If it does not meet the requirements, determine whether the energy storage SOC is ideal. If it is not ideal, perform SOC recovery and set the SOC charge / discharge control coefficient. If it is ideal, recalculate. And make a judgment; If the conditions are met, the wind speed is checked, and it is determined whether the area is a medium wind speed zone. If the wind speed zone is medium, the DFIG uses rotor overspeed unloading control to provide backup capacity, and combines it with virtual inertial control frequency modulation. The supercapacitor uses virtual droop control frequency modulation. After frequency modulation is completed, a judgment is made. If a polarity change occurs, set the DFIG additional power to 0 and continue frequency modulation of the supercapacitor. If no change occurs, return to the previous step to perform frequency modulation of the DFIG and supercapacitor. If it does not belong to the medium wind speed area, then judge If the value is greater than 0, then supercapacitor frequency modulation is performed; if the value is greater than 0, then it is determined whether the area belongs to a high wind speed zone. If it does, then DFIG pitch angle control and energy storage virtual inertial integrated control are performed; if it does not, then supercapacitor frequency modulation is performed. DFIG employs rotor overspeed unloading control to provide backup capacity, specifically as follows: In the process of participating in power system frequency regulation, the DFIG (Diverterless Geological Injection Generator) uses rotor overspeed load shedding control to provide reserve capacity, while the supercapacitor also shares part of the reserve capacity to reduce wind curtailment. Specifically: When the supercapacitor does not provide backup, the backup capacity provided by the DFIG will be used as the total backup capacity required by the wind-storage system, as follows: (9) In the formula, α For wind power utilization rate, P max This is the maximum power in MPPT operation mode. P R For wind curtailment capacity, d % represents the load reduction percentage; When a wind-storage combined system provides standby capacity, a SOC adjustment factor is added. k To achieve dynamic load reduction, the following is required: （10） in To configure the curtailed wind capacity after energy storage, i.e., the reserve capacity of the DFIG in the wind-storage system, a Logistic regression function is introduced, with SOC as the independent variable and an adaptive factor. n and P 0 is a parameter variable. k With the dependent variable as the moderating coefficient, a model is constructed as shown in the following equation: （11） In the formula, n This is an adaptive factor used to measure how quickly the curve changes. P 0 is the initial value; P max To maximize the output power of energy storage, S SOC This represents the SOC value at the current moment. Set the SOC charge / discharge control coefficient as follows: Partition the SOC, where SOC min SOC low SOC high SOC max Let $\mathbf$ represent the minimum, slightly lower, slightly higher, and maximum values ​​of the energy storage SOC, respectively, and set to 0.1, 0.45, 0.55, and 0.9, to construct the charge / discharge control coefficient curves. m Charge and discharge control coefficient curve m From the charging control coefficient curve m c and discharge control coefficient curve m d The curves consist of two parts; when the SOC (State of Charge) is relatively good, that is... hour m Taking the maximum value, the charge / discharge control coefficient constraint is as follows: （12） （13） In the formula, K max This represents the maximum droop coefficient for energy storage.

2. The primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control according to claim 1, characterized in that, In step 1, a wind-storage integrated system model is established, specifically as follows: Establish a system frequency response model: Obtaining wind power increment Δ P W Energy storage system power increment Δ P E Load change Δ P L The active power increment Δ of other energy units besides the wind and storage system participating in primary frequency regulation P G ,in, （1） In the formula, R G This is the prime mover adjustment coefficient. T G The time constant of the speed controller, T ch The time constant of the steam turbine. T R The reheat time constant is F H Given the reheater gain, the system frequency response model transfer function is obtained as follows: （2） In the formula, H , D , p These are the inertial constant, damping constant, and wind power penetration rate, respectively. Establish the DFIG model: The mechanical power output of the fan is obtained as follows: （3） In the formula, ρ air density, A The swept area of ​​the wind turbine blades. R w Let be the radius of the blade. v For wind speed, C p The wind energy utilization coefficient, λ For the tip speed ratio, β The pitch angle is the propeller angle. C p ( λ , β ) is about λ , β The nonlinear function is used to establish the DFIG model as follows: （4） In the formula, γ For the introduction of intermediate variables, ω To adjust the rotor speed of the wind turbine, the wind turbine is operated in a reduced-load mode, and virtual inertial frequency control based on the frequency change rate is added to reduce the amount of speed change during frequency regulation. Establish a supercapacitor model: Construct an equivalent circuit model of a supercapacitor, including an ideal capacitor. C Series equivalent resistance R es and the parallel equivalent resistance reflecting the long-term static changes of the supercapacitor. R eq The voltage of the supercapacitor is: （5） Supercapacitor energy storage capacity E scss ,exist t Remaining battery power at any given time E C ( t and State of Charge (SOC) k ( t )for: （6） In the formula, U max , U min These are the highest and lowest operating voltages of the supercapacitor, respectively, in volts (V). P C ( t )for t The charging and discharging power of a supercapacitor at any given time, measured in MW. η C To improve the charging and discharging efficiency of supercapacitors. E scss The total capacity of the supercapacitor is expressed in MWh. The power storage and SOC constraints are as follows: （7） In the formula, P rmin , P rmax SOC min SOC max These represent the minimum output power, maximum output power, minimum and maximum SOC of the supercapacitor, respectively. Because the voltage of a single supercapacitor is relatively low, multiple capacitors are connected in series and parallel to form a large-capacity supercapacitor array. Let the number of supercapacitors connected in series be... n s The number of parallel connections is n p Then the equivalent internal resistance and equivalent capacitance are: （8）。 3. The primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control according to claim 2, characterized in that, In step 2, wind speed zones are divided, specifically as follows: The minimum cut-in wind speed of DFIG is obtained as follows v in Maximum cut-out wind speed is v out The DFIG operates at speeds of 0.7 to 1.2 pu, dividing wind speeds into three zones: low, medium, and high, and two critical wind speed zones. Low wind speed area v in - v 1. The DFIG rotor speed is constant at 0.7 pu. The DFIG does not have frequency modulation capability; only the supercapacitor participates in frequency modulation. Critical low wind speed zone v lo - v m1 When the wind speed changes frequently in the low and medium wind speed zones, the active power increment of DFIG participating in frequency regulation changes slowly until it withdraws from frequency regulation. Medium wind speed zone v 1- v 2. It consists of a maximum power tracking region and a constant speed region; Critical high wind speed zone v m2 - v h When the wind speed changes frequently in the medium and high wind speed zones, the active power increment of DFIG participating in frequency regulation changes slowly until it withdraws from frequency regulation. High wind speed area v 2- v out The DFIG rotor speed is 1.2 pu, and pitch angle control is introduced to maintain the wind turbine output power at the rated power without load shedding for standby.

4. The primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control according to claim 1, characterized in that, After frequency tuning is completed, determine If a polarity change occurs, set the DFIG additional power to 0 and perform supercapacitor frequency modulation, specifically as follows: During the inertial response phase, the additional power of the DFIG gradually decreases until it exits frequency modulation as it approaches the lowest frequency point. Simultaneously, the stored energy suddenly increases its power to prevent a secondary frequency drop caused by the power surge when the DFIG exits frequency modulation. When the frequency difference change rate... When the polarity changes, the primary frequency modulation stage begins, and the dominant frequency modulation shifts from the DFIG to the supercapacitor. At this point, the supercapacitor output Δ P E for: （14） In the formula, K b With the moment of polarity change of the rate of change of frequency difference K p same.

5. The primary frequency regulation method for a wind-storage integrated system based on multivariable fuzzy logic control according to claim 4, characterized in that, If the wind speed zone is medium, the DFIG uses rotor overspeed load reduction control to provide backup capacity, and combines it with virtual inertial control frequency modulation. The supercapacitor uses virtual droop control frequency modulation, specifically: In the medium wind speed range, the DFIG simulates the grid inertial process using virtual inertial control and employs overspeed load shedding control to provide frequency regulation reserves. The supercapacitor energy storage simulates the primary frequency regulation process using virtual droop control, controlling its active power output based on frequency deviation according to the droop characteristics. The virtual inertial control and droop control work together to participate in frequency regulation. for: （15） In the formula, K d These are virtual inertial control coefficients. K p This is the droop control coefficient. m To determine the SOC charging and discharging control coefficients, a three-input, dual-output fuzzy logic controller is established to control the wind speed. v Frequency variation rate and frequency deviation The three state variables are used as inputs to the fuzzy controller, and the fuzzy control output is obtained after defuzzification. K d and K p ,in: The wind speed range is set to 6 m / s to 13.5 m / s, the frequency difference rate range is set to -0.6 Hz / s to 0 Hz / s, the frequency deviation range is set to -1 Hz to -0.033 Hz, and the virtual inertia coefficient and droop coefficient range to 5 to 20. The fuzzy subsets of the output quantities include very small, small, relatively small, medium, relatively large, large, and very large. The fuzzy logic rules are as follows: When the wind speed is between 7.5 and 9.75 m / s, the virtual inertia coefficient will be... K d Take the smaller value and K d It decreases as the rate of change of frequency difference increases, while the droop coefficient... K p It increases with the increase of the rate of change of frequency deviation, and as the frequency deviation continues to increase, K d , K p It also increases; When the wind speed is between 8.5 and 10.9 m / s, K d It decreases as the rate of change of frequency difference increases. K p It increases with the increase of the rate of change of frequency difference; when both the rate of change of frequency difference and the frequency deviation are small, K d As the value of increases, K p If the value is too small, when the rate of change of frequency difference is small and the frequency deviation is large, the frequency deviation will be too large. K d , K p The value should be relatively large when the rate of change of frequency difference is large and the frequency deviation is small. K d , K p The value of should be relatively small, especially when both the rate of change of frequency difference and the frequency deviation are large. K d The value should be relatively small. K p The value should be as large as possible; When the wind speed is between 9.75 and 12 m / s, and the rate of change of frequency difference is small, K d The value should be increased appropriately. K d It decreases as the rate of change of frequency difference increases, but decreases when the wind speed is too high. K d Do not take excessively large values, and at the same time K p The value is positively correlated with both the frequency deviation and the rate of frequency change; the larger the rate of frequency deviation change, the greater the wind speed. K d The value should be relatively small. K p Increase.

Images