A method for coordinating kinetic energy of a fan rotor with energy storage to regulate frequency

Abstract

The present application relates to kinetic energy storage technical field, specifically to a fan rotor kinetic energy and energy storage coordinated frequency modulation method, including the following steps: obtaining angular velocity and calculating rotor kinetic energy, collecting state of charge and calculating change rate, calculating frequency deviation and combining droop coefficient, matching kinetic energy demand, distributing kinetic energy and energy storage, predicting state of charge and executing constraint adjustment, calculating current and torque and synchronously controlling to generate frequency modulation signal.In the present application, by synchronously extracting angular velocity, rotor kinetic energy value and state of charge change rate and constructing coupling relationship, the power decision takes into account energy stock and state evolution trend, the adjustment coefficient is calculated according to state of charge change rate and combined with interval threshold value to correct reference power, the energy storage output is dynamically adjusted according to state change, the mechanical and battery cooperation degree is improved by kinetic energy proportional distribution, the adjustment continuity and safety margin are improved, and the response consistency and frequency modulation stability are improved by current and torque instruction synchronous control.

Description

Technical Field

[0001] This invention relates to the field of kinetic energy storage technology, and in particular to a method for coordinated frequency regulation of kinetic energy and energy storage of a wind turbine rotor. Background Technology

[0002] The field of kinetic energy storage technology mainly involves technical systems that utilize mechanical or electrical systems to temporarily store energy and release it when needed. It encompasses methods for storing electrical energy through the kinetic energy of rotating bodies and participating in power system regulation. Core aspects include energy accumulation and release mechanisms based on rotating mass, power interaction control with the grid, and coordinated scheduling with other energy storage forms. Energy absorption and release are achieved by controlling the rotational speed of rotating equipment, and regulation is combined with the power system frequency status to participate in and support the grid power balance process. Among these, the traditional wind turbine rotor kinetic energy and energy storage coordinated frequency regulation method refers to a technical solution that utilizes the rotational kinetic energy of the wind turbine rotor and external energy storage devices to jointly participate in grid frequency regulation. Addressing the insufficient frequency support capacity during wind power grid connection, the traditional solution uses the adjustment of the wind turbine's electromagnetic torque to change the rotor speed to release or absorb kinetic energy, combined with the energy storage device injecting or absorbing active power via a converter. Simultaneously, a power allocation strategy is set based on the grid frequency deviation, coordinating the allocation between rotor kinetic energy control and energy storage power output to complete the frequency regulation process.

[0003] In practical operation, existing technologies often directly allocate speed regulation and energy storage output based on frequency deviation. The judgment criteria are biased towards instantaneous power demand and static state quantities, lacking a coordinated characterization of the remaining energy release level on the mechanical side and the rate of battery state evolution. This leads to a situation where power allocation appears balanced on the surface but is unbalanced in the process. For example, when frequency disturbances continue to expand, the rotor releases more energy in the early stages, but the support drops sharply in the later stages as the speed approaches the lower limit. The energy storage side also maintains a high output even when the state of charge is close to the boundary, resulting in insufficient subsequent regulation margin. Existing operating modes also often handle power allocation and operational constraints in separate stages, with the power command formation stage prioritizing meeting the immediate demand. Downward adjustment demands and boundary verification are mostly reflected in the subsequent execution stage, creating a time lag between command feasibility and equipment safety. This can easily lead to problems such as repeated command corrections, response jitter, and increased power tracking deviation. Consequently, the frequency support process becomes discontinuous. If the grid connection point frequency fluctuates frequently, the mechanical side and energy storage side alternately enter a restricted state, causing the support capacity to decay faster. In actual operation, it is also easy to encounter phenomena such as frequent energy storage charging and discharging switching, increased current fluctuations, and sudden torque adjustments. This not only affects the frequency regulation quality but also aggravates the heat load and aging accumulation of components, thus restricting the stability, economy, and maintainability in continuous regulation scenarios. Summary of the Invention

[0004] To achieve the above objectives, the present invention adopts the following technical solution: a method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage, comprising the following steps:

[0005] S1: Acquire the angular velocity signal output by the wind turbine rotor shaft speed sensor, calculate the wind turbine rotor inertia parameters and calculate the rotor kinetic energy value, collect the current state of charge and the state of charge of the previous sampling period output by the energy storage battery management system and calculate the rate of change of state of charge, and fuse the rotor kinetic energy value and the rate of change of state of charge to generate a kinetic energy-charge coupling data set.

[0006] S2: Based on the kinetic energy charge coupling data group, call the wind farm grid connection point frequency measurement device to obtain the grid frequency signal and calculate the frequency deviation. Combine the primary frequency regulation droop coefficient to calculate the target frequency regulation power. Call the wind turbine rotor inertia control unit to output the kinetic energy release power and calculate the kinetic energy participation frequency regulation demand. Match the target frequency regulation power with the kinetic energy participation frequency regulation demand to generate the kinetic energy frequency regulation demand.

[0007] S3: Call the kinetic energy frequency regulation demand and calculate the adjustment coefficient, determine the base power according to the state of charge interval threshold table and calculate the energy storage participation power, allocate the kinetic energy frequency regulation demand proportionally and generate the kinetic energy release power command, and integrate the energy storage participation power and the kinetic energy release power command to generate the wind-storage collaborative power command.

[0008] S4: Call the wind-storage coordinated power command and calculate the state of charge prediction result. Compare the state of charge prediction result with the upper limit threshold of energy storage state of charge and the lower limit threshold of energy storage state of charge. Combine the kinetic energy release power command with the minimum speed limit of the wind turbine to perform constraint judgment and adjust the power component to generate a coordinated constraint power command.

[0009] S5: Call the coordinated constraint power command and calculate the energy storage charging and discharging current command, convert the kinetic energy release power command into the electromagnetic torque command of the wind turbine rotor-side converter, and drive the energy storage converter PWM drive unit and the wind turbine rotor-side converter control unit to perform synchronous control to generate a wind-storage joint frequency modulation control signal.

[0010] As a further aspect of the present invention, the kinetic energy-charge coupling data set includes kinetic energy characterization items, charge evolution items, and state correlation items; the kinetic energy frequency regulation demand includes power matching items, energy release demand values, and frequency regulation margin values; the wind-storage coordinated power command includes energy storage output items, kinetic energy release items, and power coordination items; the coordinated constraint power command includes charge boundary constraint items, speed lower limit constraint items, and limit correction items; the wind-storage joint frequency regulation control signal includes current control items, torque control items, and synchronous drive items.

[0011] As a further aspect of the present invention, the process of calculating the inertia parameters of the wind turbine rotor and calculating the rotor kinetic energy value includes sampling the angular velocity signal output by the wind turbine rotor shaft speed sensor at fixed time intervals, calculating the change in angular acceleration using the angular velocity signal of three consecutive sampling cycles, and correcting the inertia parameters by combining the preset rated inertia coefficient of the wind turbine rotor to obtain the dynamic inertia parameters.

[0012] As a further aspect of the present invention, the process of calculating the rotor kinetic energy value includes performing corresponding calculations based on the dynamic inertia parameters and the angular velocity signal output by the wind turbine rotor shaft speed sensor.

[0013] As a further aspect of the present invention, the specific steps of S1 are as follows:

[0014] S101: Obtain the angular velocity signal output by the fan rotor shaft speed sensor, call the rotor inertia parameter, perform kinetic energy numerical calculation on the angular velocity signal and perform dimensionless processing to generate the rotor kinetic energy value.

[0015] S102: Obtain the current state of charge (SOC) and the SOC of the previous sampling period from the energy storage battery management system, perform differential calculation on the two sets of SOC data, and normalize them in combination with the sampling time interval to obtain the rate of change of SOC.

[0016] S103: Based on the rotor kinetic energy value and the rate of change of charge, perform time identifier alignment processing, synchronize and rearrange according to the sampling period index, and perform pairing and integration of the two types of sequences to form a unified structure data frame set and generate a kinetic energy-charge coupling data group.

[0017] As a further aspect of the present invention, the specific steps of S2 are as follows:

[0018] S201: Based on the kinetic energy load coupling data group, call the wind farm grid connection point frequency measurement device to obtain the grid frequency signal, extract the sampling time identifier to construct the frequency sequence, and perform frequency deviation value calculation with the rated frequency reference. Perform consistency processing on the obtained sequence to obtain the frequency deviation sequence.

[0019] S202: Based on the frequency deviation sequence, call the first frequency modulation droop coefficient, perform proportional mapping calculation on the frequency deviation sequence to form the target frequency modulation power sequence, and simultaneously call the wind turbine rotor inertial control unit to output the kinetic energy release power sequence, and perform participation quantity calculation processing on the power sequence to obtain the frequency modulation power demand difference.

[0020] S203: Based on the frequency modulation power demand difference, perform matching processing, align the target frequency modulation power sequence with the kinetic energy release power participation amount at the corresponding positions, perform constraint condition judgment and structural reconstruction processing on the difference sequence, and generate the kinetic energy frequency modulation demand amount.

[0021] As a further aspect of the present invention, the specific steps of S3 are as follows:

[0022] S301: Based on the required kinetic energy frequency modulation, call the rate of change of state of charge in the kinetic energy charge coupling data group, perform adjustment coefficient calculation on the rate of change of state of charge, and combine it with the sampling period identifier to sort the sequence and obtain the charge adjustment coefficient.

[0023] S302: Call the state of charge interval threshold table according to the charge regulation coefficient, retrieve the interval where the current state of charge is located and extract the corresponding reference power, perform energy storage participation power calculation processing on the reference power, and at the same time perform proportional allocation calculation by combining the kinetic energy frequency regulation demand and the rotor kinetic energy value to obtain the kinetic energy release power command.

[0024] S303: Based on the combined processing of the kinetic energy release power command and the energy storage participation power execution, the two types of power sequences are aligned by time index and superimposed to generate a wind-storage coordinated power command.

[0025] As a further aspect of the present invention, the specific steps of S4 are as follows:

[0026] S401: Based on the wind-storage coordinated power command, call the current state of charge, extract the state of charge sequence and perform prediction value calculation processing to form a state of charge change trend data sequence. At the same time, call the upper limit threshold of the energy storage state of charge and the lower limit threshold of the energy storage state of charge, perform interval comparison judgment on the prediction sequence, and obtain the state of charge constraint judgment result.

[0027] S402: Based on the state of charge constraint determination result, call the kinetic energy release power command, extract the fan rotor operating state parameters and retrieve the minimum speed limit value of the fan, perform constraint condition judgment on the kinetic energy release power command, identify the power component that meets the limit trigger condition and perform marking processing to obtain the restricted power component identifier set;

[0028] S403: Based on the constrained power component identifier set, perform adjustment processing, perform amplitude correction and structural rearrangement on the marked power components according to the constraint judgment results, and perform unified integration processing with the unconstrained power components to form a power sequence set that satisfies multiple constraint conditions, and generate a coordinated constraint power command.

[0029] As a further aspect of the present invention, the specific steps of S5 are as follows:

[0030] S501: According to the coordinated constraint power command, the energy storage DC bus voltage detection unit is called to obtain the DC bus voltage measurement sequence and construct the voltage state dataset. The coordinated constraint power command and the voltage measurement sequence are processed to perform current command calculation to obtain the energy storage current command.

[0031] S502: Call the kinetic energy release power command according to the energy storage current command, perform torque mapping calculation on the kinetic energy release power command, and combine it with the wind turbine rotor operating status parameters for consistency processing to obtain the electromagnetic torque command;

[0032] S503: Based on the electromagnetic torque command and the energy storage current command, the energy storage converter PWM drive unit and the wind turbine rotor-side converter control unit are invoked to perform synchronous modulation and timing alignment processing on the two types of control sequences, and generate a wind-storage joint frequency modulation control signal.

[0033] As a further aspect of the present invention, the process of calculating the rate of change of state of charge includes performing a difference calculation on the current state of charge output by the energy storage battery management system and the state of charge of the previous sampling period, and obtaining the rate of change of state of charge by combining the preset fixed sampling time interval.

[0034] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0035] In this invention, by synchronously extracting angular velocity, rotor kinetic energy value, and the rate of change of state of charge and establishing a coupling relationship, the power decision takes into account both energy storage and state evolution trend. The adjustment coefficient is calculated based on the rate of change of state of charge and the reference power is corrected by combining the interval threshold, so that the energy storage output is dynamically adjusted with the state change. Combined with the kinetic energy ratio distribution, the degree of mechanical and battery coordination is improved, the adjustment continuity and safety margin are improved, and the response consistency and frequency regulation stability are improved by synchronous control of current and torque commands. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0037] Figure 1 This is a schematic diagram of the steps of the present invention;

[0038] Figure 2 This is a detailed schematic diagram of S1 of the present invention;

[0039] Figure 3 This is a detailed schematic diagram of S2 of the present invention;

[0040] Figure 4 This is a detailed schematic diagram of S3 of the present invention;

[0041] Figure 5 This is a detailed schematic diagram of S4 of the present invention;

[0042] Figure 6This is a detailed schematic diagram of S5 of the present invention. Detailed Implementation

[0043] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0044] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0045] Please see Figure 1 This invention provides a method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage, comprising the following steps:

[0046] S1: Obtain the angular velocity signal output by the wind turbine rotor shaft speed sensor, call the wind turbine rotor inertia parameter to calculate the rotor kinetic energy value, and at the same time collect the current state of charge output by the energy storage battery management system and the state of charge of the previous sampling period and calculate the rate of change of state of charge. Synchronize and organize the rotor kinetic energy value and the rate of change of state of charge to generate a kinetic energy-charge coupling data set.

[0047] S2: Based on the kinetic energy charge coupling data group, call the wind farm grid connection point frequency measurement device to obtain the grid frequency signal and calculate the frequency deviation. Combine the primary frequency regulation droop coefficient to calculate the target frequency regulation power. At the same time, call the wind turbine rotor inertial control unit to output the kinetic energy release power and calculate the kinetic energy participation frequency regulation demand. Match the target frequency regulation power with the kinetic energy participation frequency regulation demand to generate the kinetic energy frequency regulation demand.

[0048] S3: Calculate the regulation coefficient based on the rate of change of state of charge in the kinetic energy charge coupling data group according to the kinetic energy frequency regulation demand, determine the reference power corresponding to the current state of charge according to the state of charge interval threshold table and calculate the energy storage participation power, and at the same time, proportionally allocate the kinetic energy frequency regulation demand by combining the wind turbine rotor kinetic energy value and obtain the kinetic energy release power command. Combine the energy storage participation power and the kinetic energy release power command to generate the wind-storage collaborative power command.

[0049] S4: Based on the wind-storage coordinated power command, the current state of charge is called to calculate the state of charge prediction result, and it is compared with the upper limit threshold of the energy storage state of charge and the lower limit threshold of the energy storage state of charge. At the same time, the constraint judgment is combined with the kinetic energy release power command and the minimum speed limit of the wind turbine. The power component that triggers the constraint condition is adjusted to generate the coordinated constraint power command.

[0050] S5: Based on the coordinated constraint power command, the energy storage DC bus voltage detection unit is called to calculate the energy storage charging and discharging current command. At the same time, the kinetic energy release power command is converted into the electromagnetic torque command of the wind turbine rotor-side converter. The energy storage converter PWM drive unit and the wind turbine rotor-side converter control unit perform synchronous control to generate the wind-storage joint frequency regulation control signal.

[0051] The kinetic energy-charge coupling data set includes kinetic energy characterization items, charge evolution items, and state correlation items; the kinetic energy frequency regulation demand includes power matching items, energy release demand values, and frequency regulation margin values; the wind-storage coordinated power command includes energy storage output items, kinetic energy release items, and power coordination items; the coordinated constraint power command includes charge boundary constraint items, speed lower limit constraint items, and limit correction items; the wind-storage joint frequency regulation control signal includes current control items, torque control items, and synchronous drive items.

[0052] Please see Figure 2 The specific steps of S1 are as follows:

[0053] S101: Obtain the angular velocity signal output by the fan rotor shaft speed sensor, call the rotor inertia parameter, perform kinetic energy numerical calculation on the angular velocity signal and perform dimensionless processing to generate the rotor kinetic energy value.

[0054] Rotor speed records from the rotor shaft speed sensor are continuously read according to the sampling period. Each record is converted into angular velocity. Then, the rotor inertia parameters filed in the factory test report of the same fan are retrieved, and kinetic energy quantification is performed point by point. During implementation, missing values ​​and sudden jump values ​​are first removed. Specifically, records where the speed difference between two adjacent sampling points exceeds 0.8 revolutions per minute are marked as abnormal, and then the average of the two valid points is used to fill the gap. Then, the speed is converted from revolutions per minute to angular displacement per second, and then combined with the rotor inertia into the kinetic energy calculation logic to obtain the raw kinetic energy value measured in joules. Finally, it is uniformly converted into a megajoule sequence and three decimal places are retained. Taking a wind turbine with a sampling interval of 0.1 seconds and a rotor inertia of 4,200,000 kg / m² as an example, sampling times of 12.0 rpm, 11.8 rpm, and 11.5 rpm were continuously obtained. The converted angular velocities were 1.257, 1.236, and 1.204, respectively, and the corresponding kinetic energies were 3.318 MJ, 3.208 MJ, and 3.044 MJ, respectively. This forms a rotor kinetic energy value that can be continuously referenced along the time axis. Public wind turbine data shows that the common grid connection frequency for variable speed onshore turbines is 50 Hz, the cut-in wind speed of representative models is 3 m / s, and the cut-out wind speed is 18 m / s to 21 m / s. The rated speed of the reference unit is around 12.1 rpm. Therefore, the above-mentioned operating point of 12.0 rpm is in the common range that can be implemented.

[0055] S102: Obtain the current state of charge (SOC) and the SOC of the previous sampling period from the energy storage battery management system, perform differential calculation on the two sets of SOC data, and normalize them in combination with the sampling time interval to obtain the rate of change of SOC.

[0056] The current state of charge (SOC) and the SOC of the previous sampling period are read at a sampling rate of 0.1 seconds, consistent with the previous period. First, an integrity check is performed, then the difference between two adjacent records is calculated, and finally, normalization is performed using the sampling time interval. In practice, the SOC is uniformly retained to 0.1 percentage points. When duplicate reports are encountered within the same period, only the record with the latest timestamp is retained. When a null value is returned, the value of the previous valid period is directly referenced without introducing additional interpolation. In the example, the battery management record continuously provides 62.4%, 62.3%, and 62.1%. The change in SOC at the second sampling point relative to the previous point is a decrease of 0.1 percentage points, and the change in SOC at the third sampling point relative to the previous point is a decrease of 0.2 percentage points. Dividing by the 0.1-second sampling interval, the rate of change of SOC is obtained as a decrease of 1.0 percentage point per second and a decrease of 2.0 percentage points per second, respectively. To maintain data flow in subsequent segments, the decreasing direction is recorded as a negative value in the sequence, the charging direction as a positive value, and zero change is recorded as 0. After this sorting, the change in state of charge retains the original measurement sequence and can be directly used in the interval retrieval, power regulation coefficient setting and constraint determination in the following text. Public information defines the state of charge as the percentage of the current energy storage capacity relative to the total available energy capacity, with a value range corresponding to 0% to 100% from completely empty to fully charged, and this value directly affects the energy storage's ability to provide continuous ancillary services to the grid.

[0057] S103: Based on the rotor kinetic energy value and the rate of change of state of charge, perform time stamp alignment processing, perform synchronous rearrangement according to the sampling period index, and perform pairing and integration of the two types of sequences to form a unified structure data frame set, generating kinetic energy and charge coupling data group;

[0058] The aforementioned rotor kinetic energy value and state of charge change rate sequence are mapped one-to-one according to the sampling time identifier. First, a unified time column is established with 0.1 seconds as the index. Then, the two types of sequences are rearranged into the same data frame according to the same index. During implementation, the timestamps are checked for misalignment. If the deviation does not exceed 0.02 seconds, they are directly merged into the nearest sampling point. If the deviation exceeds 0.02 seconds, the misaligned item is deleted and the gap number is recorded to prevent cross-cycle splicing in subsequent power allocation. Taking this embodiment as an example, the rotor kinetic energy value is 3.318 megajoules, 3.208 megajoules, and 3.044 megajoules at 10.0 seconds, 10.1 seconds, and 10.2 seconds, respectively. The state of charge change rate is recorded as 0, decreasing by 1.0 percentage point per second, and decreasing by 2.0 percentage point per second at the same time index, respectively. After rearrangement, three coupled records can be formed. To facilitate continuous access for subsequent frequency deviation, frequency modulation requirements, and power allocation, the sampling time, rotor kinetic energy value, current state of charge, previous cycle state of charge, and state of charge change rate are all retained in the same record. After this processing, the coupled data corresponding to the 10.1-second index is fixed at 3.208 megajoules, 62.3%, 62.4%, decreasing by 1.0 percentage point per second, and the coupled data corresponding to the 10.2-second index is fixed at 3.044 megajoules, 62.1%, 62.3%, decreasing by 2.0 percentage point per second. Subsequent segments are directly processed using the results from this set.

[0059] Table 1 Kinetic-charge coupling data table

[0060]

[0061] As shown in Table 1, the three records in the table have been unified in terms of time index. In the following text, whenever the sampling results of 10.1 seconds and 10.2 seconds are called, the data of the corresponding row in Table 1 will be directly referenced, and no resampling or recalculation will be performed.

[0062] Please see Figure 3 The specific steps of S2 are as follows:

[0063] S201: Based on the kinetic energy load coupling data group, call the wind farm grid connection point frequency measurement device to obtain the grid frequency signal, extract the sampling time identifier to construct the frequency sequence, and perform frequency deviation value calculation with the rated frequency reference. Perform consistency processing on the obtained sequence to obtain the frequency deviation sequence.

[0064] Based on the coupling record formed in Table 1, the grid connection point frequency measurement sequence is read again, and the frequency values ​​are entered into the same batch of data according to the same time index. During implementation, invalid samples caused by phase loss or communication interruption are first screened out, and then the frequency values ​​are retained to 0.001 Hz. Subsequently, the difference between the frequency and the rated frequency is calculated point by point to obtain the frequency deviation sequence. Taking three sampling points at 10.0 seconds, 10.1 seconds, and 10.2 seconds as examples, the measured grid connection point frequencies are 49.920 Hz, 49.900 Hz, and 49.880 Hz, respectively. After adjusting to the rated frequency of 50.000 Hz, the corresponding frequency deviations are a decrease of 0.080 Hz, a decrease of 0.100 Hz, and a decrease of 0.120 Hz, respectively. To ensure unified calculation in the future, all decreasing deviations are recorded as positive frequency modulation demand, and increasing deviations are recorded as negative frequency modulation demand. Small fluctuations with an absolute value less than 0.010 Hz are merged into 0 and are no longer included in the frequency modulation power calculation. The resulting frequency deviation sequence can be directly matched with the kinetic energy charge coupling record in Table 1 by time index. The index at 10.1 seconds corresponds to a fixed decrease of 0.100 Hz, and the index at 10.2 seconds corresponds to a fixed decrease of 0.120 Hz. The subsequent power demand difference can be derived based on this. The rated frequency of 50 Hz is a commonly used benchmark for AC grid-connected scenarios, and publicly available wind turbine data also indicates that there is a 50 Hz range for the unit's electrical frequency configuration.

[0065] S202: Based on the frequency deviation sequence, call the primary frequency modulation droop coefficient, perform proportional mapping calculation on the frequency deviation sequence to form the target frequency modulation power sequence, and simultaneously call the wind turbine rotor inertial control unit to output the kinetic energy release power sequence, and perform participation quantity calculation processing on the power sequence to obtain the frequency modulation power demand difference.

[0066] The frequency deviation sequence is fed point by point into the primary frequency modulation droop coefficient calculation segment. First, the proportional mapping is completed according to the droop coefficient selected in the experiment. Then, the power participation of the kinetic energy release in the same period is read, and the difference is then calculated to obtain the frequency modulation power demand difference. This droop coefficient is not directly taken from empirical values, but is screened based on three sets of settings in the grid connection test, respectively taking 6 MW / Hertz, 8 MW / Hertz, and 10 MW / Hertz. Under the same frequency drop condition, the lowest frequency point and power swing within 30 seconds are compared. It is measured that the lowest frequency point of 8 MW / Hertz is 49.890 Hz and the power swing is 0.41 MW. Taking into account both response amplitude and fluctuation degree, 8 MW / Hertz is fixed in this embodiment. Substituting the 0.100 Hz drop at 10.1 seconds into the textual calculation logic, the target frequency modulation power is obtained as 0.800 MW. Then, the kinetic energy release participation amount of 0.350 MW under the same index is read. This participation amount is calculated from the instantaneous release limit of 0.730 MW derived from the aforementioned kinetic energy drop, and then multiplied by the participation ratio of 0.48 verified on-site. The difference between 0.800 MW and 0.350 MW is the difference in frequency modulation power demand at 10.1 seconds, which is 0.450 MW. Similarly, at 10.2 seconds, the target frequency modulation power is 0.960 MW, the kinetic energy release participation amount is 0.410 MW, and the demand difference is 0.550 MW.

[0067] S203: Based on the frequency modulation power demand difference, perform matching processing, align the target frequency modulation power sequence with the kinetic energy release power participation amount, perform constraint condition judgment and structural reconstruction processing on the difference sequence, and generate the kinetic energy frequency modulation demand amount;

[0068] After aligning the target frequency-modulated power sequence and the kinetic energy release power participation point by point, constraint judgment and structural reconstruction are performed according to the same time index, and cross-time value taking is not allowed. During implementation, it is first determined whether the target frequency-modulated power is greater than 0. If it is not greater than 0, the point is directly transferred to the zero demand record; then it is determined whether the kinetic energy release participation exceeds the target frequency-modulated power. If it exceeds, the excess part is truncated to 0, and no negative demand is formed; if it does not exceed, the difference between the two is retained as the effective demand. Taking 10.1 seconds as an example, the target frequency-modulated power of 0.800 MW is greater than 0, and the kinetic energy release participation of 0.350 MW is less than the target frequency-modulated power, so the effective difference is maintained at 0.450 MW; taking 10.2 seconds as an example, the effective difference is maintained at 0.550 MW. Subsequently, the effective difference is repackaged with the rotor kinetic energy value, current state of charge, and state of charge change rate in Table 1 to form the kinetic energy frequency-modulated demand record directly referenced in subsequent power allocation. After this arrangement, the record corresponding to the 10.1-second index is fixed at a frequency regulation demand of 0.450 MW, rotor kinetic energy of 3.208 MJ, current state of charge (SOC) of 62.3%, and a rate of SOC change decreasing by 1.0 percentage point per second; the record corresponding to the 10.2-second index is fixed at a frequency regulation demand of 0.550 MW, rotor kinetic energy of 3.048 MJ, current SOC of 62.1%, and a rate of SOC change decreasing by 2.0 percentage points per second. The advantage of this calculation logic is that by synchronously comparing the target frequency regulation power and the kinetic energy participation, records with negative differences are eliminated first, preventing reverse compensation terms from appearing during subsequent power allocation.

[0069] Please see Figure 4 The specific steps of S3 are as follows:

[0070] S301: Based on the kinetic energy frequency regulation demand, call the rate of change of state of charge in the kinetic energy charge coupling data group, perform adjustment coefficient calculation on the rate of change of state of charge, and combine it with the sampling period identifier to sort the sequence and obtain the charge adjustment coefficient.

[0071] The kinetic energy frequency regulation demand is correlated point-by-point with the rate of change of state of charge (SOC) in Table 1. First, the regulation coefficient is retrieved based on the interval into which the absolute value of the SOC falls. Then, the SOC regulation coefficients are rearranged according to the sampling time to form a sequence. The interval values ​​are determined by statistics from 120 sets of charging and discharging conditions: 1.00 for absolute values ​​below 0.5 percentage points per second, 0.90 for 0.5 to 1.5 percentage points per second, 0.75 for 1.5 to 3.0 percentage points per second, and 0.55 for greater than 3.0 percentage points per second. The interval selection is based on the remaining available power and temperature rise records after the same unit has run continuously for 20 minutes under frequency support scenarios. The 0.75 range shows the smallest deviation between response amplitude and charge retention capability. Substituting the aforementioned data, the absolute value of the indexed state of charge change rate at 10.1 seconds is 1.0 percentage point per second, falling within the range of 0.5 to 1.5, corresponding to a charge regulation coefficient of 0.90; the absolute value of the indexed state of charge change rate at 10.2 seconds is 2.0 percentage points per second, falling within the range of 1.5 to 3.0, corresponding to a charge regulation coefficient of 0.75. During processing, this coefficient is stored alongside the frequency regulation demand at the same time. Therefore, the 10.2-second record is written as a frequency regulation demand of 0.550 MW, a charge regulation coefficient of 0.75, a current state of charge of 62.1%, and a rotor kinetic energy of 3.044 MJ. Subsequent energy storage power and rotor release power are all based on this record.

[0072] S302: Call the state of charge interval threshold table according to the charge regulation coefficient, retrieve the interval where the current state of charge is located and extract the corresponding reference power, perform energy storage participation power calculation processing on the reference power, and at the same time perform proportional allocation calculation by combining the kinetic energy frequency regulation demand and the rotor kinetic energy value to obtain the kinetic energy release power command.

[0073] After determining the charge regulation coefficient, the current state of charge (SOC) is used to find the SOC interval threshold table, extract the corresponding reference power, and then multiply the reference power by the SOC to obtain the energy storage participation power. Simultaneously, the kinetic energy release power command is allocated based on the kinetic energy frequency regulation demand and the rotor kinetic energy margin. The reference power table is compiled from 100 grid-connected frequency regulation tests, with 0.15 MW for the 20% to 35% range, 0.30 MW for the 35% to 50% range, 0.45 MW for the 50% to 70% range, and 0.35 MW for the 70% to 85% range. The values ​​are based on the maximum stable power that can maintain a DC voltage fluctuation of no more than 25 volts after 120 seconds of continuous discharge in each range. Substituting the current SOC of 62.1% from the 10.2-second index into the search, it falls within the 50% to 70% range, corresponding to a reference power of 0.45 MW; multiplying this by the aforementioned SOC of 0.75 yields an energy storage participation power of 0.3375 MW. Subsequently, 0.3375 MW was deducted from the 0.550 MW kinetic energy frequency regulation demand at this point, leaving 0.2125 MW as the amount to be shared by the rotor side; combined with the fact that the rotor kinetic energy of 3.048 MJ in 10.2 seconds is within the release range, the kinetic energy release power command at this sampling point was determined to be 0.2125 MW.

[0074] Table 2 Charge Threshold and Reference Power Table

[0075]

[0076] Referring to Table 2, the 62.1% state of charge at the 10.2-second index corresponds to a base power of 0.45 MW. After combining this with an adjustment factor of 0.75, we obtain a storage power of 0.3375 MW. This result has been incorporated into the kinetic energy release distribution in the latter half of this section.

[0077] S303: Based on the combined processing of kinetic energy release power command and energy storage participation power execution, the two types of power sequences are aligned by time index and superimposed to generate wind-storage coordinated power command;

[0078] After aligning the kinetic energy release power command and the energy storage participation power according to the time index, they are superimposed to form a wind-storage coordinated power command at the same sampling point, and the result is written sequentially into the power sequence set. During implementation, the time identifiers of the two branches are not recalculated; instead, the same index data output by S302 is directly called. Taking the 10.2-second index as an example, the aforementioned energy storage participation power is 0.3375 MW, and the kinetic energy release power command is 0.2125 MW. After superposition, the resulting wind-storage coordinated power command is 0.550 MW, which exactly matches the kinetic energy frequency regulation demand at that point formed by S203. If the deviation between the superimposed result and the demand exceeds 0.010 MW, the sampling point is recorded as a mismatch point, and the process is reverted to the previous segment to recheck the reference power range; in this embodiment, no mismatch occurred at 10.1 seconds and 10.2 seconds. The resulting coordinated power sequence can then be directly used for charge trend prediction and multiple constraint correction. The unified record at the 10.2-second index is fixed at 0.550 MW for coordinated power, 0.3375 MW for the energy storage component, and 0.2125 MW for the rotor component. Subsequent constraint judgments are all based on this set of values. The advantage of this operational logic is that by superimposing the energy storage component and the rotor component at the same time point, subsequent constraint processing can be directly executed on a single power sequence, eliminating the need to correct two separate sets of commands.

[0079] Please see Figure 5 The specific steps of S4 are as follows:

[0080] S401: Based on the wind-storage coordinated power command, the current state of charge is invoked, the state of charge sequence is extracted and the predicted value is calculated and processed to form a data sequence of the state of charge change trend. At the same time, the upper limit threshold of the energy storage state of charge and the lower limit threshold of the energy storage state of charge are invoked, and the interval comparison judgment is performed on the predicted sequence to obtain the state of charge constraint judgment result.

[0081] The energy storage component in the wind-storage coordinated power command is read, and a short-term charge change trend sequence is constructed based on the current state of charge. The predicted result is then compared with preset upper and lower thresholds. The thresholds are not directly given but determined by cycle life tests. Group verification is performed on upper limits of 88%, 85%, and 82% and lower limits of 15%, 20%, and 25%. Under the same 1200 shallow cycle conditions, the capacity retention rate of upper limit 85% and lower limit 20% is the highest. Therefore, 85% and 20% are fixed in this embodiment. In the example, the aforementioned energy storage participation power of 0.3375 MW is indexed at 10.2 seconds. The rated energy storage capacity is taken as 5 MW. Based on the continuous discharge condition of the next 120 seconds, the predicted discharge is 11.25 kWh, corresponding to a decrease in state of charge of 0.225 percentage points. After deducting the current state of charge of 62.1%, the predicted value is 61.875%. Comparing 61.875% with 85% and 20% respectively, the results are within the threshold range. Therefore, the result of the charge state constraint determination for this sampling point is recorded as passed. Following the same logic, the predicted value of the 10.1-second index also remains within the threshold range. Therefore, the subsequent constraint source is transferred to the rotor-side operating parameter determination, rather than being directly truncated by the charge state.

[0082] S402: Based on the state of charge constraint determination result, call the kinetic energy release power command, extract the wind turbine rotor operating state parameters and retrieve the minimum speed limit value of the wind turbine, perform constraint condition judgment on the kinetic energy release power command, identify the power component that meets the limit triggering condition and mark it to obtain the restricted power component identifier set;

[0083] After the state of charge determination is passed, the rotor operating state parameters corresponding to the kinetic energy release power command are extracted, and the minimum speed limit value is retrieved. A limit trigger judgment is performed for each sampling point. The minimum speed limit value is determined by three sets of unit inertia support tests, using speeds of 10.5 rpm, 11.0 rpm, and 11.5 rpm respectively. Under the same frequency drop, the peak torsional vibration of the main shaft and the recovery time are compared. The peak torsional vibration of the main shaft at 11.0 rpm is controlled within 8.2% of the rated torque, and the recovery time is less than 7 seconds; therefore, it is fixed as the minimum speed limit value in this embodiment. Substituting the 10.2-second index into the aforementioned kinetic energy release power command of 0.2125 MW and the current speed of 11.5 rpm, and calculating based on the release process over the next 5 seconds, the predicted speed will drop to 9.27 rpm, which is lower than the 11.0 rpm limit value. Therefore, the rotor power component at this sampling point is marked as a restricted component; the energy storage component of 0.3375 MW at the same sampling point does not trigger the speed limit and remains unmarked. After this arrangement, the constrained power component identifier set corresponding to the 10.2-second index is recorded as rotor component constrained and energy storage component unconstrained.

[0084] S403: Based on the constrained power component identifier set, perform adjustment processing, perform amplitude correction and structural rearrangement on the marked power components according to the constraint judgment results, and integrate them with the unconstrained power components to form a set of power sequences that meet multiple constraint conditions, and generate a coordinated constraint power instruction.

[0085] Based on the constrained power component identifier set, the marked rotor components undergo amplitude correction, and are then rearranged and integrated with the unconstrained components to obtain a power sequence that satisfies both state of charge and minimum speed constraints. During correction, the allowable release limit is first calculated backward from the minimum speed limit of 11.0 rpm, and then the rotor component is reduced without altering the energy storage component. For example, at the 10.2-second index, the original rotor component of 0.2125 MW would reduce the predicted speed to 10.92 rpm, so this component is reduced to 0.1500 MW; the energy storage component remains unchanged at 0.3375 MW, resulting in a coordinated constraint power command of 0.4875 MW after reorganization. Since the total power after correction is lower than the original coordinated power of 0.550 MW, the deficit of 0.0625 MW is registered as a constrained shortfall and will not be compensated in this cycle. When performing the same judgment on the 10.1-second index, if the predicted speed does not reach the minimum speed limit, the power sequence remains unchanged. The experimental results show that after the minimum speed constraint is added, the rotor side over-release at the 10.2-second sampling point is directly reduced by 0.0625 megawatts, and subsequent torque commands no longer fall into the restricted range.

[0086] Please see Figure 6 The specific steps of S5 are as follows:

[0087] S501: Call the energy storage DC bus voltage detection unit according to the coordinated constraint power command, obtain the DC bus voltage measurement sequence and construct the voltage state dataset, perform current command calculation processing on the coordinated constraint power command and voltage measurement sequence to obtain the energy storage current command;

[0088] After the coordinated power command is formed, the energy storage component is extracted and paired point-by-point with the DC bus voltage measurement sequence to calculate the energy storage current command. During implementation, the DC bus voltage is first adjusted using a 5-point moving average to remove instantaneous spikes, and then the energy storage component is converted into DC-side current. Public research on wind turbine energy storage integrated into the DC link uses DC link voltage control as a key variable. Existing doubly-fed induction generator (DFIG) literature demonstrates that the DC bus voltage is stabilized at around 1150 volts. Therefore, in this embodiment, using 1128 volts, 1132 volts, and 1129 volts as continuous sampling points is within a reasonable range. Substituting a 10.2-second index, after the aforementioned coordination, the energy storage component remains 0.3375 MW. Paired with a DC bus voltage of 1129 volts, a DC-side current command of 299 amps is obtained. If a 5-point average of 1130 volts is used for calculation, a smoothed current command of around 299 amps is obtained. The difference between the two results is no more than 1 amp, so the value written into the command sequence is 299 amps. The result is still within the allowable range compared to the preset current limit of 420 amps, so the energy storage current command output in this step is directly used for the next stage of torque mapping.

[0089] S502: Based on the energy storage current command, the kinetic energy release power command is invoked, the torque mapping calculation is performed on the kinetic energy release power command, and the consistency is adjusted in combination with the wind turbine rotor operating status parameters to obtain the electromagnetic torque command;

[0090] After the energy storage current command is determined, the kinetic energy release power command is invoked to perform torque mapping on the rotor-side power sequence and compile it into a unified record with the rotor operating parameters of the same period. During implementation, mapping is only performed on rotor components that are not further restricted. Taking a 10.2-second index as an example, the rotor component corrected by S403 is 0.1500 MW, and the current speed is 11.5 rpm. First, the angular displacement per second is calculated to be 1.204, and then the target electromagnetic torque of 124600 N·m is calculated according to the power-angular velocity correspondence. To prevent abrupt changes in commands between adjacent periods, the slope of the next two periods is checked again. If the torque increment exceeds 15000 N·m, it is truncated to the upper limit. In this embodiment, the electromagnetic torque of the previous period was 117800 N·m, and the increment in this period is 6800 N·m, which does not exceed the limit; therefore, 124600 N·m is maintained and written into the control sequence. The resulting record is an electromagnetic torque command of 124,600 N·m, an energy storage current command of 299 A, and a coordination constraint power command of 0.4875 MW under a 10.2-second index. Subsequent modulation will be carried out synchronously based on this set of results.

[0091] S503: Based on electromagnetic torque command and energy storage current command, call the PWM drive unit of energy storage converter and the control unit of wind turbine rotor-side converter to perform synchronous modulation and timing alignment processing on the two types of control sequences, and generate wind-storage joint frequency modulation control signal;

[0092] Electromagnetic torque commands and energy storage current commands are loaded into the same control cycle. Sampling timing alignment is completed first, and then a joint wind-storage frequency modulation control signal is generated. During implementation, the 124600 N·m torque command and 299 A current command under a 10.2-second index are simultaneously written into the same cycle buffer. Subsequently, the torque modulation sequence and current modulation sequence are output in a unified clock trigger order to avoid phase misalignment where one side updates prematurely while the other lags behind. If the timestamp difference between the two types of commands exceeds 0.5 ms, the entire cycle is delayed by one control beat for re-alignment; in this embodiment, the difference is 0.2 ms, so joint modulation is directly initiated. Bench testing showed that after adopting this joint control, the lowest frequency point increased from 49.860 Hz under traditional distributed control to 49.900 Hz, the number of state-of-charge violations decreased from 3 to 0, and the torque peak-to-valley fluctuation rate decreased from 8.4% to 4.6%. The experimental results show that the joint control signal after coordination and constraint can maintain a more stable current and torque output under the same frequency disturbance, and fix the aforementioned limited deficit within a controllable range. Therefore, the generated wind-storage joint frequency regulation control signal is the final stage control result that can be directly executed.

[0093] Table 3 Comparison of Joint FM Effects

[0094]

[0095] As shown in Table 3, this embodiment retains a higher output power after constraint under the same disturbance conditions, while no state of charge overflow occurs. The 0.4875 MW in the table is the result of the coordinated constraint power command obtained in S403, which is subsequently converted into an electromagnetic torque command of 124600 N·m and an energy storage current command of 299 amps in S502 and S503, respectively, forming a coherent implementation data chain.

[0096] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage, characterized in that, Includes the following steps: S1: Acquire the angular velocity signal output by the wind turbine rotor shaft speed sensor, calculate the wind turbine rotor inertia parameters and calculate the rotor kinetic energy value, collect the current state of charge and the state of charge of the previous sampling period output by the energy storage battery management system and calculate the rate of change of state of charge, and fuse the rotor kinetic energy value and the rate of change of state of charge to generate a kinetic energy-charge coupling data set. S2: Based on the kinetic energy charge coupling data group, call the wind farm grid connection point frequency measurement device to obtain the grid frequency signal and calculate the frequency deviation. Combine the primary frequency regulation droop coefficient to calculate the target frequency regulation power. Call the wind turbine rotor inertia control unit to output the kinetic energy release power and calculate the kinetic energy participation frequency regulation demand. Match the target frequency regulation power with the kinetic energy participation frequency regulation demand to generate the kinetic energy frequency regulation demand. S3: Call the kinetic energy frequency regulation demand and calculate the adjustment coefficient, determine the base power according to the state of charge interval threshold table and calculate the energy storage participation power, allocate the kinetic energy frequency regulation demand proportionally and generate the kinetic energy release power command, and integrate the energy storage participation power and the kinetic energy release power command to generate the wind-storage collaborative power command. S4: Call the wind-storage coordinated power command and calculate the state of charge prediction result. Compare the state of charge prediction result with the upper limit threshold of energy storage state of charge and the lower limit threshold of energy storage state of charge. Combine the kinetic energy release power command with the minimum speed limit of the wind turbine to perform constraint judgment and adjust the power component to generate a coordinated constraint power command. S5: Call the coordinated constraint power command and calculate the energy storage charging and discharging current command, convert the kinetic energy release power command into the electromagnetic torque command of the wind turbine rotor-side converter, and drive the energy storage converter PWM drive unit and the wind turbine rotor-side converter control unit to perform synchronous control to generate a wind-storage joint frequency modulation control signal.

2. The method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage according to claim 1, characterized in that, The kinetic energy-charge coupling data set includes kinetic energy characterization items, charge evolution items, and state correlation items; the kinetic energy frequency regulation demand includes power matching items, energy release demand values, and frequency regulation margin values; the wind-storage coordinated power command includes energy storage output items, kinetic energy release items, and power coordination items; the coordinated constraint power command includes charge boundary constraint items, speed lower limit constraint items, and limit correction items; the wind-storage joint frequency regulation control signal includes current control items, torque control items, and synchronous drive items.

3. The method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage according to claim 1, characterized in that: The process of calculating the inertia parameters of the wind turbine rotor and calculating the rotor kinetic energy value includes sampling the angular velocity signal output by the wind turbine rotor shaft speed sensor at fixed time intervals, calculating the change in angular acceleration using the angular velocity signal of three consecutive sampling cycles, and correcting the inertia parameters by combining the preset rated inertia coefficient of the wind turbine rotor to obtain the dynamic inertia parameters.

4. The method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage according to claim 1, characterized in that: The process of calculating the rotor kinetic energy value includes performing corresponding calculations based on the dynamic inertia parameters and the angular velocity signal output by the wind turbine rotor shaft speed sensor.

5. The method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage according to claim 1, characterized in that, The specific steps of S1 are as follows: S101: Obtain the angular velocity signal output by the fan rotor shaft speed sensor, call the rotor inertia parameter, perform kinetic energy numerical calculation on the angular velocity signal and perform dimensionless processing to generate the rotor kinetic energy value. S102: Obtain the current state of charge (SOC) and the SOC of the previous sampling period from the energy storage battery management system, perform differential calculation on the two sets of SOC data, and normalize them in combination with the sampling time interval to obtain the rate of change of SOC. S103: Based on the rotor kinetic energy value and the rate of change of charge, perform time identifier alignment processing, synchronize and rearrange according to the sampling period index, and perform pairing and integration of the two types of sequences to form a unified structure data frame set and generate a kinetic energy-charge coupling data group.

6. The method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage according to claim 1, characterized in that, The specific steps of S2 are as follows: S201: Based on the kinetic energy load coupling data group, call the wind farm grid connection point frequency measurement device to obtain the grid frequency signal, extract the sampling time identifier to construct the frequency sequence, and perform frequency deviation value calculation with the rated frequency reference. Perform consistency processing on the obtained sequence to obtain the frequency deviation sequence. S202: Based on the frequency deviation sequence, call the first frequency modulation droop coefficient, perform proportional mapping calculation on the frequency deviation sequence to form the target frequency modulation power sequence, and simultaneously call the wind turbine rotor inertial control unit to output the kinetic energy release power sequence, and perform participation quantity calculation processing on the power sequence to obtain the frequency modulation power demand difference. S203: Based on the frequency modulation power demand difference, perform matching processing, align the target frequency modulation power sequence with the kinetic energy release power participation amount at the corresponding positions, perform constraint condition judgment and structural reconstruction processing on the difference sequence, and generate the kinetic energy frequency modulation demand amount.

7. The method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage according to claim 1, characterized in that, The specific steps for S3 are as follows: S301: Based on the required kinetic energy frequency modulation, call the rate of change of state of charge in the kinetic energy charge coupling data group, perform adjustment coefficient calculation on the rate of change of state of charge, and combine it with the sampling period identifier to sort the sequence and obtain the charge adjustment coefficient. S302: Call the state of charge interval threshold table according to the charge regulation coefficient, retrieve the interval where the current state of charge is located and extract the corresponding reference power, perform energy storage participation power calculation processing on the reference power, and at the same time perform proportional allocation calculation by combining the kinetic energy frequency regulation demand and the rotor kinetic energy value to obtain the kinetic energy release power command. S303: Based on the combined processing of the kinetic energy release power command and the energy storage participation power execution, the two types of power sequences are aligned by time index and superimposed to generate a wind-storage coordinated power command.

8. The method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage according to claim 1, characterized in that, The specific steps of S4 are as follows: S401: Based on the wind-storage coordinated power command, the current state of charge is invoked, the current state of charge is extracted and the predicted value is calculated to form a data sequence of the state of charge change trend. At the same time, the upper limit threshold of the energy storage state of charge and the lower limit threshold of the energy storage state of charge are invoked, and the interval comparison judgment is performed on the predicted sequence to obtain the state of charge constraint judgment result. S402: Based on the state of charge constraint determination result, call the kinetic energy release power command, extract the fan rotor operating state parameters and retrieve the minimum speed limit value of the fan, perform constraint condition judgment on the kinetic energy release power command, identify the power component that meets the limit trigger condition and perform marking processing to obtain the restricted power component identifier set; S403: Based on the constrained power component identifier set, perform adjustment processing, perform amplitude correction and structural rearrangement on the marked power components according to the constraint judgment results, and perform unified integration processing with the unconstrained power components to form a power sequence set that satisfies multiple constraint conditions, and generate a coordinated constraint power command.

9. The method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage according to claim 1, characterized in that, The specific steps of S5 are as follows: S501: According to the coordinated constraint power command, the energy storage DC bus voltage detection unit is called to obtain the DC bus voltage measurement sequence and construct the voltage state dataset. The coordinated constraint power command and the voltage measurement sequence are processed to perform current command calculation to obtain the energy storage current command. S502: Call the kinetic energy release power command according to the energy storage current command, perform torque mapping calculation on the kinetic energy release power command, and combine it with the wind turbine rotor operating status parameters for consistency processing to obtain the electromagnetic torque command; S503: Based on the electromagnetic torque command and the energy storage current command, the energy storage converter PWM drive unit and the wind turbine rotor-side converter control unit are invoked to perform synchronous modulation and timing alignment processing on the two types of control sequences, and generate a wind-storage joint frequency modulation control signal.

10. The method for coordinated frequency regulation of wind turbine rotor kinetic energy and energy storage according to claim 1, characterized in that: The process of calculating the rate of change of state of charge includes performing a difference calculation between the current state of charge output by the energy storage battery management system and the state of charge of the previous sampling period, and obtaining the rate of change of state of charge by combining the calculation with a preset fixed sampling time interval.

Images