Distributed wind set virtual inertia active support control method suitable for weak power grid

Abstract

The invention discloses a distributed wind set virtual inertia active support control method suitable for a weak power grid, and the method comprises the steps: dynamically adjusting the inertia support strength according to the frequency disturbance of a power grid through adaptive virtual inertia adjustment; multi-mode cooperative control is adopted, and an operation mode is switched according to the frequency deviation and the change rate so as to provide an optimal response; by means of dynamic damping compensation, frequency oscillation is suppressed in combination with multiple factors and an optimization algorithm, and system damping is improved; multi-time-scale cooperative control is implemented, and inertia support, power adjustment and rotating speed recovery are coordinated at a millisecond level, a second level and a minute level; and rotor kinetic energy graded release is adopted, kinetic energy release is accurately controlled according to the frequency drop degree, and sudden rotation speed drop is avoided. In addition, additional frequency damping control is introduced to enhance suppression of frequency fluctuation. The method can significantly improve the response speed and adaptability of the wind turbine generator to the frequency fluctuation of the weak power grid, ensures the frequency stability of the power grid, and provides key technical support for stable grid connection of wind power in the weak power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of power system stability control, and in particular to a virtual inertia active support control strategy for wind turbines applicable to weak power grids. Background Art

[0002] Traditional wind turbine control strategies usually adopt constant power or power following control, lacking inertia support ability and being unable to effectively suppress grid frequency fluctuations. Virtual inertia control technology enables wind turbines to have inertia support ability by simulating the inertia characteristics of synchronous generators, thereby improving the frequency stability of the power grid. However, the traditional virtual inertia control method has the following problems in a weak power grid environment:

[0003] Traditional wind turbines usually only focus on maximum power tracking to pursue the maximization of power generation efficiency, while ignoring the support for the frequency stability of the power grid. When the grid frequency fluctuates, due to the lack of an effective virtual inertia response mechanism, wind turbines cannot provide inertia support in time like traditional synchronous generators, resulting in insufficient grid frequency regulation ability, further expansion of frequency deviation, and even possible cascading failures, seriously threatening the safe and stable operation of the power grid.

[0004] To address this problem, research has successively proposed strategies such as virtual inertia control (VIC) and virtual synchronous generator (VSG), attempting to endow wind turbines with synchronous machine characteristics. However, in a weak power grid environment, these control methods still face many challenges: such as poor response adaptability due to fixed inertia gain; no hierarchical kinetic energy release, sudden drop in fan speed affecting continuous power generation; insufficient consideration of the grid feedback mechanism leading to secondary frequency oscillations; voltage-frequency coupling causing stability problems; and lack of coordinated response control from milliseconds to seconds. The inertia coefficient remains fixed and is difficult to adapt to different intensities of grid disturbances, easily causing over-response of the system under small disturbances and insufficient response under large disturbances; the control mode is single, unable to flexibly switch operating states according to the grid frequency deviation and its change rate, and unable to achieve the optimal control effect; the means of suppressing frequency oscillations are weak, lacking an effective damping compensation mechanism, unable to quickly eliminate oscillations, resulting in a low system damping ratio and difficult to guarantee stability. Moreover, there is a lack of overall consideration in dealing with the time scale of inertia response and the release of rotor kinetic energy, prone to problems such as untimely short-term response, unstable long-term operation, and out-of-control fan speed. In summary, there is an urgent need for a wind turbine control strategy that can comprehensively and effectively address the challenges of a weak power grid environment.

[0005] Therefore, there is an urgent need for a distributed wind turbine virtual inertia enhancement control strategy applicable to a weak power grid environment to improve the response speed of wind turbines to grid frequency fluctuations and system stability, and enhance the operation reliability and safety of weak power grids.

[0006] After retrieval, the invention patent with the patent publication number CN119171470A discloses a frequency and voltage support control method and system for an energy storage power station facing a weak power grid. Based on the deviation data of the three-phase voltage, current, positive and negative sequence power, frequency, etc. on the AC side of the power grid access point, the droop control strategy is used to adjust the active and reactive power output of the energy storage power station, so as to support the frequency and voltage of the main power grid during a fault.

[0007] The technical comparison between the above-mentioned comparative document and the present application is as follows:

[0008] For the active support control of the virtual inertia of distributed wind turbines in the present application, adaptive virtual inertia regulation, dynamic damping compensation and multi-modal cooperative control are adopted. The inertia support strength is adjusted in real time according to the power grid frequency disturbance, and inertia support, power regulation and speed recovery are cooperatively executed on multiple time scales such as milliseconds, seconds and minutes. In contrast, the comparative document focuses on ensuring the transient safety and stability of the power grid through the dynamic power regulation of the energy storage device. In comparison, the present invention utilizes the adjustment ability of the wind turbine itself, and precisely suppresses the frequency oscillation through the hierarchical release of the rotor kinetic energy and the additional frequency damping control, which is more suitable for the stable grid connection requirements after the access of wind power in a weak power grid.

[0009] After retrieval, the invention patent with the patent publication number CN119518698A discloses an online active support method and system for network-following new energy equipment under a weak power grid. The online real-time least squares method is used to identify the Thevenin equivalent circuit parameters of the equipment access point in the distribution network. Without relying on the parameters of multiple types of equipment, the critical short-circuit ratio and transmission capacity of the system are derived through the equivalent circuit, and the PQ control parameters of the new energy equipment are optimized under the equipment capacity constraint, so as to realize the active support for the system and improve the transmission capacity.

[0010] The technical comparison between the above-mentioned comparative document and the present application is as follows:

[0011] The present application focuses on the active control of the virtual inertia support of wind turbines. Through adaptive adjustment and multi-modal cooperative switching, the optimal operation mode is automatically selected based on the frequency deviation and change rate to realize dynamic damping compensation and hierarchical release of rotor kinetic energy, so as to quickly suppress the frequency fluctuation in a weak power grid. Compared with the above-mentioned comparative document which focuses on optimizing the PQ parameters by using the equipment capacity, the present invention conducts hierarchical and multi-time scale coordinated control for the real-time dynamic characteristics of the virtual inertia of wind turbines, can more precisely control the kinetic energy release, avoid the sudden drop of the speed, and has more significant advantages in suppressing the frequency oscillation.

[0012] After retrieval, the invention patent with the publication number CN119482501A discloses a transient voltage active support control method and device for the sending-end power grid of an LCC-HVDC system, which aims at providing transient voltage support for the sending-end power grid of the grid-forming photovoltaic converter in the LCC-HVDC system. By using the virtual synchronous generator control strategy, it modifies the output voltage by using the active power and voltage phase information output by the converter, realizes the effective adjustment of the grid transient voltage, and prevents the "reverse adjustment" phenomenon of reactive power.

[0013] The technical comparison between the above-mentioned comparative document and the present application is as follows:

[0014] The present application focuses on the virtual inertia support of wind turbines. By establishing a multi-time scale collaborative control strategy and a dynamic feedback mechanism, it realizes the synchronous coordination of virtual inertia support, power regulation and speed recovery of wind turbines at the millisecond level, second level and minute level. And by using the method of hierarchical release of rotor kinetic energy and additional frequency damping control, it accurately controls the release of wind turbine kinetic energy in response to grid frequency drops, avoiding sudden speed drops and effectively eliminating frequency oscillations. The comparative document focuses on the voltage amplitude and phase correction of photovoltaic converters to improve the stability of photovoltaic access to weak power grids. The present invention is applicable to the active frequency support of wind power grid connection in a weak power grid environment, focusing on transient regulation of frequency rather than voltage, and there are differences in technical focuses and application scenarios. Summary of the Invention

[0015] Object of the Invention: The object of the present invention is to provide a distributed wind turbine virtual inertia active support control method applicable to weak power grids. Aiming at problems such as severe frequency fluctuations and poor system stability caused by insufficient inertia in weak power grids, a composite control architecture including adaptive inertia regulation, multi-modal coordinated operation, dynamic damping compensation, multi-time scale coordinated control and hierarchical release of rotor kinetic energy is constructed. The present invention takes wind turbines as the main body of frequency auxiliary support. Without significantly affecting the main function power generation efficiency, it enhances the inertia response ability of wind turbines under frequency disturbance conditions, thereby improving the frequency stability and operation safety of the entire weak power grid, and providing an engineering feasible technical path for wind power grid connection operation in a complex power grid environment.

[0016] To achieve the above object, the solution of the present invention includes the following steps:

[0017] S1. System architecture construction

[0018] Embed the virtual inertia enhancement control strategy module of the present invention into the main control system of the wind turbine. This module is seamlessly connected to existing hardware devices such as wind turbine monitoring systems and power conversion systems to ensure that key parameter information such as grid frequency, wind speed, and unit output power can be obtained in real time and accurately.

[0019] S2, Real-time Monitoring and Data Acquisition

[0020] Through high-precision sensors, continuously and real-time monitor the power grid frequency and its rate of change. After converting the collected analog signals into digital signals, the sensors transmit them to the control strategy module in the main control system. At the same time, collect the operating state parameters of the wind turbine, such as the current wind speed, generator speed, output power, etc., to provide comprehensive data support for subsequent control decisions.

[0021] S3, Adaptive Virtual Inertia Coefficient Calculation

[0022] Based on the fuzzy logic control algorithm, comprehensively analyze the collected power grid frequency and its rate of change, and the operating state parameters of the wind turbine in the control strategy module. According to the pre-set fuzzy rule base, which covers the optimized value range of the inertia coefficient under different power grid disturbance intensities and wind turbine operating conditions, calculate the most suitable adaptive inertia coefficient at the current moment through fuzzy reasoning. For example, when the power grid frequency deviation is greater than a certain threshold and the rate of change is fast, it is determined as a strong disturbance, and the inertia coefficient is increased accordingly; conversely, in the case of small deviation and slow change, the inertia coefficient is decreased.

[0023] S4, Multi-modal Cooperative Control Switching

[0024] The control system compares the frequency deviation and its rate of change calculated in real time with the preset mode switching threshold. When the frequency deviation and rate of change meet the conditions of the "Static Support Mode", the main control system sends an instruction to the power conversion system to switch the wind turbine to this mode and stably output inertia support; if the frequency fluctuation intensifies and reaches the trigger conditions of the "Dynamic Support Mode" or "Frequency Fast Response Mode", immediately switch to the corresponding mode and adjust the power output strategy to achieve fast and flexible response.

[0025] S5, Dynamic Damping Compensation Implementation

[0026] According to the monitoring results of the frequency oscillation trend, calculate the equivalent virtual damping coefficient using the built-in algorithm. In the control loop, introduce a damping force that matches the oscillation trend by adjusting the controller parameters or activating a specially designed compensator. For example, when it is detected that the frequency oscillation shows a divergent trend, quickly increase the damping gain to suppress the oscillation; if the oscillation tends to be stable, appropriately reduce the damping to avoid excessive damping affecting the system response speed.

[0027] S6, Multi-time Scale Cooperative Control Execution

[0028] During the short-term inertia support stage in milliseconds, once a sudden change in grid frequency is detected, the control strategy module immediately issues commands to the generator control system to quickly adjust the electromagnetic torque, realizing the rapid release or storage of rotor kinetic energy and preventing sharp fluctuations in frequency. During the medium-term power regulation period in seconds, based on the medium-term change trend of grid frequency and the power generation capacity of the unit, the power reference value is optimized, and the output of the converter is adjusted to maintain the stability of grid frequency. During the long-term speed recovery process in minutes, the control strategy module slowly adjusts the power generation according to the deviation between the current speed of the wind turbine and the rated speed, gradually restoring the speed of the wind turbine to the normal level and ensuring the continuous and stable operation of the wind turbine.

[0029] S7. Hierarchical release operation of rotor kinetic energy

[0030] Multiple hierarchical thresholds for grid frequency drop are preset, corresponding to different kinetic energy release levels. When it is monitored that the grid frequency drops to a certain threshold range, according to the preset hierarchical strategy, the intensity of the rotor kinetic energy release command is gradually increased. For example, when the frequency drop is small, a small amount of kinetic energy is released to provide a certain amount of inertia support while maintaining the wind turbine speed within a safe range; as the frequency drop intensifies, the kinetic energy release amount is further increased to ensure effective support for the grid frequency until the frequency returns to normal.

[0031] S8. Implementation of additional frequency damping control

[0032] Combined with the secondary control link, a special additional damping control unit is added to the main control system. This unit real-time monitors the frequency oscillation situation during the virtual inertia support process. By comparing with the preset oscillation suppression target, the additional damping parameters are dynamically adjusted. When it is detected that the frequency oscillation exceeds the allowable range, the additional damping is quickly increased to strongly suppress the oscillation, ensuring the stability of grid frequency and avoiding secondary frequency problems caused by virtual inertia support.

[0033] Furthermore, in step S1, the virtual inertia enhancement control strategy module adopts an adaptive virtual inertia adjustment mechanism, which real-time collects information on grid frequency and its change rate, and uses an adaptive inertia coefficient calculation method based on fuzzy logic control. Fuzzy logic control can effectively handle uncertainty and nonlinear problems. Through preset fuzzy rules, considering factors such as the intensity of grid disturbance and the operating state of the wind turbine generator set, such as wind speed and unit output power, the inertia support parameters are dynamically adjusted. This dynamic control strategy of "large inertia for high deviation and small inertia for low deviation" can not only quickly input a large amount of inertia support when there is a large deviation in grid frequency to accelerate the response speed, but also avoid over-regulation in the case of small deviations to prevent system overshoot and ensure the stable operation of the system.

[0034] The adaptive virtual inertia adjustment mechanism includes the following steps:

[0035] S101. Collect power grid frequency data and rate of change in real time, and calculate the frequency deviation;

[0036] S102. Use a fuzzy controller to calculate the inertia adjustment factor ΔK in ;

[0037] S103. Calculate the adaptive inertia coefficient;

[0038] S104. Calculate the inertia support power;

[0039] S105. Send a control command to the wind turbine converter to adjust the active power output and provide inertia support.

[0040] In traditional virtual inertia control, the inertia coefficient K in is usually a fixed value, resulting in a relatively large system rigidity and inability to adapt to grid disturbances of different intensities. The present invention proposes an adaptive virtual inertia regulation mechanism that can dynamically adjust the inertia coefficient according to the frequency change of the power grid (frequency deviation Δf and rate of change of frequency ), enabling the wind turbine to provide stronger inertia support when the power grid frequency fluctuates greatly, while reducing unnecessary power output under small disturbances, improving the dynamic response ability, and at the same time reducing the risk of system overshoot and secondary frequency oscillation.

[0041] The present invention uses fuzzy logic control (FLC) combined with a nonlinear mapping method to dynamically calculate the inertia coefficient K in :

[0042]

[0043] wherein, the function f(·) is mapped to the inertia coefficient through fuzzy control rules, making it have non-linear dynamic regulation characteristics.

[0044] The calculation formula of the inertia support power:

[0045]

[0046] Preferably, the inertia coefficient K in needs to be adjusted according to the power grid state, and is defined as follows:

[0047]

[0048] wherein:

[0049] K0 is the reference inertia coefficient;

[0050] α is the frequency deviation influence coefficient;

[0051] β is the rate of change of frequency influence coefficient;

[0052] When the power grid disturbance is large, Kin increases to provide stronger inertia support; when the power grid is relatively stable, K in decreases to reduce unnecessary power regulation and lower the dynamic load of the fan.

[0053] To further optimize the adaptive ability of inertia response, the present invention adopts a fuzzy control method to make the inertia coefficient K in have a more flexible adjustment ability. The input variables of the fuzzy controller are: frequency deviation and rate of change of frequency; the output variable of the fuzzy controller is: inertia adjustment factor ΔK in

[0054] The following fuzzy rules are set:

[0055] When Δf is very small and is very small, then ΔK in is close to zero.

[0056] When Δf is moderate and is large, then ΔK in increases moderately.

[0057] When Δf is large and is very large, then ΔK in increases significantly.

[0058] Fuzzy inference result:

[0059] K in = K0 + ΔK in

[0060] where ΔK in is calculated by the fuzzy controller.

[0061] To further improve the stability of inertia control, the present invention combines an inertia time-varying adjustment strategy to construct the following inertia response function:

[0062] K in (t) = K in,max ·e -γt + K in,min

[0063] where:

[0064] K in,max is the maximum inertia coefficient;

[0065] K in,min is the minimum inertia coefficient;

[0066] γ is the inertia decay factor;

[0067] t is the inertia response duration.

[0068] This formula shows that the inertia coefficient increases rapidly at the initial stage of the disturbance and gradually decays after a certain time to ensure the flexibility of inertia support and system stability.

[0069] Furthermore, the present invention proposes a multi-modal collaborative control mechanism. By designing three operating modes including "static support mode", "dynamic support mode" and "frequency fast response mode", the wind turbine control system can automatically switch different control modes according to the grid frequency deviation and its rate of change. This mechanism enables the wind turbine to provide optimal virtual inertia support under different grid disturbance scenarios, improves the adaptability and response ability of the wind turbine to frequency fluctuations, and ensures grid stability.

[0070] (1) Static support mode

[0071] The static support mode is applicable to the situation where the grid frequency is relatively stable, that is, there is no large fluctuation or disturbance in the grid. At this time, the inertia support of the wind turbine is mainly to maintain the stability of the grid, but does not need to consume too much rotor kinetic energy of the turbine.

[0072] The mode trigger condition is that the amplitude of grid frequency fluctuation |Δf| is small and the frequency rate of change is low.

[0073] In this mode, the inertia coefficient K in is set to a minimum value, and the control system maintains the stability of the grid through the lowest inertia support power without additional energy consumption.

[0074] K in,static = K0

[0075]

[0076] In this mode, the inertia support power P of the wind turbine in,static is mainly to provide basic support when the grid frequency is stable, minimizing the energy consumption of the turbine.

[0077] (2) Dynamic support mode

[0078] When the grid frequency fluctuates by a certain amplitude (i.e., the frequency deviation |Δf| is large, but not reaching the level of a fault or large disturbance), the wind turbine enters the dynamic support mode. In this mode, the wind turbine provides moderate inertia support to suppress the frequency deviation and prevent it from deteriorating further.

[0079] The mode trigger condition is that the grid frequency deviation |Δf| is large (but not exceeding the controllable range) and the frequency rate of change is small.

[0080] At this time, the inertia coefficient K inDynamic adjustment provides appropriate inertia support according to the magnitude of the frequency deviation. The purpose of this strategy is to prevent the frequency from continuing to deviate and ensure the stability of the power grid.

[0081] Formula representation:

[0082]

[0083] In this mode, the wind turbine moderately increases the inertia support power according to the grid frequency deviation, helping the grid gradually return to the normal frequency state.

[0084] (3) Frequency fast response mode

[0085] The frequency fast response mode mainly deals with situations such as large grid disturbances, such as grid faults and sudden load changes. In this mode, the wind turbine needs to quickly release the rotor kinetic energy to rapidly provide inertia support and restore the grid frequency.

[0086] The mode trigger condition is that the grid frequency changes violently (i.e., the frequency deviation |Δf| is large, and the frequency change rate is very high), indicating that the grid is affected by large disturbances or faults.

[0087] At this time, the wind turbine enters the fast response mode and quickly releases more rotor kinetic energy to provide strong inertia support. At this time, the inertia coefficient K of the wind turbine in will increase rapidly to achieve fast response.

[0088] Formula representation:

[0089] K in,fast =K in,max ·e -γt

[0090] In the fast response mode, the wind turbine not only provides instantaneous inertia support, but also can quickly absorb the frequency deviation by rapidly releasing the rotor kinetic energy, ensuring the frequency stability of the grid.

[0091] Through the design of the multi-modal collaborative control mechanism, the wind turbine can effectively respond to various changes in the grid frequency, ensure the best virtual inertia support under different grid states, and further improve the stability and reliability of the wind turbine in a weak grid environment.

[0092] Furthermore, during the virtual inertia support process of a wind turbine, due to the transient process caused by inertia release, it may trigger secondary oscillations of frequency, resulting in overshoot and continuous oscillation of the grid frequency, affecting the stability of the power grid. To solve this problem, the present invention proposes an improved adaptive adjustment method for the equivalent virtual damping coefficient, which effectively suppresses the frequency oscillation caused by inertia support by adjusting the damping gain in real time. This method introduces more dynamic adjustment mechanisms on the original basis and combines various feedback information to ensure that the system can provide more accurate and effective frequency suppression under more complex grid disturbance conditions.

[0093] The improved adaptive adjustment method for the virtual damping coefficient not only considers the rate of change of frequency and frequency deviation, but also introduces multi-factor information such as the change of grid load, the duration of grid frequency fluctuation, and the change of the power output of the wind turbine. The combined action of these factors improves the adaptability of the system, enabling it to make more reasonable adjustments in the face of complex grid disturbances.

[0094] During the inertia support process of a wind turbine, the suppression of frequency oscillation mainly depends on the role of virtual damping. The essence of virtual damping is to control the differential response to frequency changes, thereby reducing the transient oscillation of frequency. In this mechanism, we define an equivalent virtual damping coefficient K d , which is a dynamic parameter related to the rate of change of frequency and the system state. This parameter can be adaptively adjusted to meet the requirements of different grid frequency fluctuations.

[0095] Calculation formula for the multi-factor virtual damping coefficient:

[0096]

[0097] Where:

[0098] K d (t) is the virtual damping coefficient at the current moment;

[0099] K d0 is the initial virtual damping coefficient;

[0100] ΔP wind is the change in the power output of the wind turbine;

[0101] T f is the duration of the grid frequency deviation;

[0102] γ1, γ2, γ3, γ4 are coefficient adjustment parameters used to balance the influence of each factor.

[0103] This formula takes into account the following key factors:

[0104] Rate of change of frequency and deviation: When the frequency changes rapidly, a higher damping coefficient is required to suppress the frequency fluctuation as soon as possible; when the frequency deviation is large, stronger inertia support is needed to stabilize the system.

[0105] Power change of wind turbine: The change in the output power of the wind turbine reflects the fluctuation of wind energy. When the power change is large, a higher virtual damping coefficient is required to stabilize the frequency.

[0106] Duration of frequency disturbance: When the duration of the power grid frequency disturbance is long, it means that the intensity of the disturbance is large. Therefore, it is necessary to increase the virtual damping coefficient to ensure that the system can effectively suppress the frequency fluctuation for a long time.

[0107] In the process of further improving the adaptive adjustment of the virtual damping coefficient, the present invention also introduces a power grid mode recognition mechanism. This mechanism automatically recognizes the current operating mode of the power grid, such as normal mode, slight disturbance mode, moderate disturbance mode, severe disturbance mode, etc., according to the dynamic characteristics of the power grid frequency, and adjusts the adjustment strategy of the virtual damping coefficient according to the recognition result. Through different mode recognitions, the system can take targeted adjustment measures according to the nature and intensity of the disturbance, avoiding over-response or under-response.

[0108] In order to further improve the dynamic response performance of the system, the present invention introduces an adaptive weighting mechanism to dynamically adjust the weights of various influencing factors according to the real-time state of the power grid disturbance. Specifically, according to the degree of change of the power grid frequency, the frequency characteristics of the disturbance, and the operating state of the system, the weighting coefficient of each factor is automatically adjusted to achieve more refined adjustment of the virtual damping coefficient.

[0109]

[0110] Wherein:

[0111] γ i0 is the initial weighting coefficient;

[0112] δ i is the adjustment factor, which changes dynamically according to the disturbance degree and system state.

[0113] In order to further improve the frequency suppression effect, a differential response and compensator are introduced into the control system. The purpose of these terms is to respond to the frequency fluctuation with high frequency and high precision to achieve rapid suppression. Specifically, an adjustment strategy based on differential control can be introduced into the inertia support control algorithm to reduce frequency overshoot and continuous oscillation. The differential control term will adjust the output power of the inertia support in real time according to the rate of change of the frequency This differential response helps to provide stronger support when the frequency changes rapidly, thereby reducing the excessive fluctuation of the frequency.

[0114] Formula representation:

[0115]

[0116] Among them, P damping (t) represents the control power provided by dynamic damping compensation.

[0117] In order to accurately suppress frequency fluctuations, a PID controller is used in the design of the compensator. The derivative term D in it can adjust the inertia support of the wind turbine in real time according to the rate of change of frequency. The compensator increases the damping effect of the system and improves the frequency stability by adjusting the control output of the wind turbine.

[0118] Compensator output formula:

[0119]

[0120] Among them:

[0121] u(t) is the control signal, which determines the output of the system;

[0122] e(t) is the error, representing the frequency deviation;

[0123] By introducing the above derivative control term and compensator in the inertia support control, we can more accurately suppress frequency oscillations and avoid excessive deviation of the frequency from the target value.

[0124] In the process of virtual inertia enhancement control of the wind turbine, the adaptive adjustment of the virtual damping coefficient can significantly improve the damping ratio of the system. The damping ratio ζ of the system is a key index to measure the attenuation speed of the oscillation amplitude. The higher the damping ratio of the system, the better the suppression effect of the frequency fluctuation.

[0125] Damping ratio calculation formula:

[0126]

[0127] Among them:

[0128] b is the damping coefficient, which is related to the virtual damping coefficient Kd;

[0129] m is the mass of the system (here it is the kinetic energy of the wind turbine);

[0130] k is the stiffness of the system (here it is the response strength of the inertia support).

[0131] By adjusting the virtual damping coefficient K d , the damping ratio ζ of the system can be dynamically changed, thereby increasing the attenuation speed of the frequency fluctuation and ensuring that the grid frequency can quickly stabilize.

[0132] Through the design of a dynamic damping compensation mechanism, a wind turbine can quickly suppress frequency oscillations during the virtual inertia support process, reduce overshoot and sustained oscillation phenomena, thereby improving the frequency stability of the power grid, especially in a weak grid environment.

[0133] To further improve the control effect of the virtual inertia support of the wind turbine, the present invention proposes to use a particle swarm optimization algorithm to optimize the adjustment strategy of the virtual damping coefficient. The particle swarm optimization algorithm is a global optimization algorithm based on swarm intelligence. By simulating the flight process of particles in the solution space, it can efficiently search for the optimal solution. The optimization goal is to minimize the response of the wind turbine to frequency fluctuations, reduce system overshoot, and improve frequency stability. Specifically, the following objectives are achieved by optimizing the virtual damping coefficient:

[0134] 1) Minimize the amplitude of frequency fluctuations, that is, reduce the power grid frequency deviation;

[0135] 2) Reduce system overshoot, that is, avoid excessive frequency fluctuations during the virtual inertia support process;

[0136] 3) Improve the frequency stability of the system, that is, enhance the adaptability of the power grid to sudden disturbances.

[0137] Define a comprehensive objective function J, which combines the indicators of frequency fluctuations, overshoot, and system stability. Assume that the frequency deviation of the system is Δf and the amplitude of frequency fluctuations is A f , the overshoot of the system is Δf peak , and the frequency stability is reflected by the damping ratio ζ of the system. The comprehensive objective function can be expressed as:

[0138] J = w1·A f + w2·Δf peak + w3·(1 - ζ)

[0139] The adaptive adjustment mechanism that automatically adjusts the virtual damping coefficient through the optimization algorithm can significantly improve the response ability of the wind turbine to complex disturbances and ensure that the power grid frequency is always within a stable range.

[0140] Through the particle swarm optimization algorithm, the present invention can automatically select the optimal virtual damping coefficient and weighting coefficient under various power grid disturbance conditions, thereby improving the effect of virtual inertia support and enabling the wind turbine to exhibit higher frequency stability and faster response speed in a weak grid environment.

[0141] In order to further enhance the stability of wind turbines in a weak grid environment, the present invention proposes a multi-time scale collaborative control strategy. The core idea is to adopt a three-stage control strategy of short-term inertia support (millisecond level), medium-term power regulation (second level), and long-term speed recovery (minute level) for different time scale requirements of the inertia response of wind turbines. This control strategy optimizes the response ability of wind turbines to grid disturbances by finely controlling the inertia response, power regulation, and speed recovery at different time scales, ensuring the stable operation of the system under various frequency fluctuation conditions.

[0142] The overall control process is as follows:

[0143] S601. Short-term inertia support (millisecond level): Provide instantaneous inertia support for the grid by releasing the rotor kinetic energy.

[0144] S602. Medium-term power regulation (second level): Adjust the active power of the wind turbine according to the grid frequency deviation to quickly restore the frequency.

[0145] S603. Long-term speed recovery (minute level): Restore the speed of the wind turbine to ensure the long-term stable operation of the unit.

[0146] Specifically as follows:

[0147] (1) Short-term inertia support at the millisecond level

[0148] When the grid encounters transient disturbances, instant virtual inertia support is provided for the grid by quickly releasing the rotor kinetic energy, thereby improving the transient stability of the grid.

[0149] Short-term inertia support is mainly used to cope with sudden changes and large disturbances in grid frequency, especially to suppress frequency fluctuations within a short time (millisecond level).

[0150] In the virtual inertia control of wind turbines, the response speed of short-term inertia support is crucial for quickly suppressing grid frequency fluctuations. For rapid frequency changes in a weak grid or high-disturbance environment, relying solely on inertia and other conventional control methods often cannot provide a fast enough response. To cope with millisecond-level frequency fluctuations, the present invention proposes to introduce a phase-locked loop technology to improve the response accuracy and speed of short-term inertia support. The PLL technology can provide precise frequency regulation and inertia support in an extremely short time by continuously tracking the phase change of the grid frequency and synchronously adjusting it with the generator set of the wind turbine, thus effectively coping with short-term frequency fluctuations.

[0151] To improve the response speed of wind turbines to frequency fluctuations, especially within the millisecond-level time scale, the PLL technology is used to optimize the control system of wind turbines. The specific implementation steps are as follows:

[0152] S(1-1). Construct a phase-locked loop based on the grid frequency signal and the output frequency signal of the wind turbine generator set to detect and track the phase change of the grid frequency in real time.

[0153] S(1-2). The phase-locked loop control system adjusts the output frequency of the wind turbine generator set according to the phase deviation of the grid frequency to synchronize it with the grid frequency. The frequency deviation is eliminated by adjusting the frequency regulation rate of the wind turbine generator set.

[0154] S(1-3). When the grid frequency suddenly changes or fluctuates violently, the phase-locked loop adjusts the output power of the wind turbine generator set through rapid response to provide short-term inertia support. The inertia response speed of the wind turbine generator set is increased to the millisecond level so that it can participate in frequency regulation in a timely manner and provide inertia support in the case of rapid fluctuations in the grid frequency.

[0155] Furthermore, in order to achieve precise control during short-term inertia support, the output of the phase-locked loop is combined with the virtual inertia response of the wind turbine generator set. The specific control strategy is as follows:

[0156] a. Calculate the phase error of the phase-locked loop

[0157] Set the phase error between the grid frequency and the output frequency of the wind turbine generator set This phase error can be calculated by the following formula:

[0158]

[0159] where θ grid (t) and θ wind (t) are the phases of the grid frequency and the output frequency of the wind turbine generator set respectively.

[0160] b. PLL feedback control

[0161] Adjust the control system according to the phase error The feedback control of the phase-locked loop can be realized by the following formula:

[0162]

[0163] where ω PLL is the control gain of the phase-locked loop. By adjusting this gain, the synchronization ability of the wind turbine generator set to the grid frequency change can be controlled.

[0164] c. After achieving frequency synchronization through the phase-locked loop, the output power P wind (t) of the wind turbine generator set will be adjusted along with the grid frequency change. The specific control strategy is:

[0165] P wind (t) = ΔP sync (t) + ΔP inertia (t)

[0166] Among them, ΔP sync (t) is the synchronous adjustment power, and ΔP inertia (t) is the inertia support power.

[0167] Synchronous power adjustment: According to the grid frequency deviation Δf(t), calculate the synchronous power adjustment amount ΔP sync (t), and the formula is as follows:

[0168] ΔP sync (t) = k sync ·Δf(t)

[0169] Among them, k sync is the synchronous adjustment coefficient, which is adjusted according to the response ability of the wind turbine.

[0170] Inertia support power regulation: According to the virtual inertia parameter of the wind turbine, calculate the inertia support power regulation amount, and the formula is:

[0171]

[0172] Among them, I wind is the rotor inertia of the wind turbine, is the rotational speed change rate of the wind turbine.

[0173] In order to ensure that the wind turbine can achieve millisecond-level response when the grid frequency fluctuates, the control gain ω of the phase-locked loop PLL must be accurately adjusted. The gain of the phase-locked loop determines the speed and accuracy of the system response. Excessive gain may cause system oscillation, while too low gain may cause response delay. Therefore, by optimizing the phase-locked loop gain, it is ensured that the system can quickly provide sufficient inertia support when the grid frequency suddenly changes, while avoiding unnecessary frequency oscillation.

[0174] (2) Second-level medium-term power adjustment period

[0175] After the short-term inertia support, according to the change trend of the grid frequency, adjust the active power output of the wind turbine to gradually restore the grid frequency to the set value. Medium-term power adjustment is to effectively suppress frequency fluctuations and provide continuous power support for the grid to ensure the smooth transition of the grid frequency in a short time.

[0176] When the grid frequency fluctuates too much, by controlling the active power output of the wind turbine, making its output power proportional to the deviation Δf(t) of the grid frequency, gradually stabilizing the grid frequency. Assuming that the active power output of the wind turbine is P(t), the active power adjustment rule is:

[0177] P(t) = P nom·(1 - β·Δf(t))

[0178] where P nom is the rated power of the wind turbine, β is the frequency response coefficient, and the power output of the wind turbine is controlled by adjusting β.

[0179] By dynamically adjusting the power output of the wind turbine, the grid frequency is quickly restored to the normal value. The specific adjustment method can be adjusted through a PI controller, and the control strategy is:

[0180]

[0181] where

[0182] K p and K i are the proportional and integral gains, which are used to adjust the speed of the system response.

[0183] Within seconds, by adjusting the power output, the grid frequency can be quickly restored to the set value, effectively reducing the amplitude of frequency fluctuations and providing continuous guarantee for grid stability.

[0184] (3) Long-term speed recovery process in minutes

[0185] Objective: After the grid frequency is restored to a stable state, according to the wind energy resource conditions, gradually restore the speed of the wind turbine to ensure the long-term stable operation of the unit.

[0186] The long-term speed recovery is mainly used to gradually restore the speed of the wind turbine to the normal operating value after the grid frequency fluctuates, avoiding excessive mechanical load on the unit due to over-operation.

[0187] When the frequency is restored to the set range, the wind turbine gradually restores its speed according to the recovery speed of the grid frequency. The recovery rate depends on the smoothness of the grid frequency recovery, and the specific recovery strategy is:

[0188]

[0189] where ω nom is the rated speed of the wind turbine, γ is the adjustment factor, f set is the grid set frequency, and Δf(t) is the current frequency deviation.

[0190] The wind turbine gradually restores its speed over a long period of time to ensure a smooth transition of the mechanical load of the fan and avoid excessive stress on mechanical components.

[0191] By gradually restoring the speed of the wind turbine, ensure stable operation after frequency recovery, and at the same time avoid load impact caused by too fast recovery, ensuring the long-term stability of the wind turbine.

[0192] Through dynamic switching, the control strategies in the above three stages achieve coordinated control from short-term inertia support to medium-term power regulation and then to long-term speed recovery. During actual operation, the control system of the wind turbine will judge the currently required control strategy in real time according to the change of the grid frequency and operate according to the corresponding time scale.

[0193] Furthermore, the core objective of the rotor kinetic energy hierarchical release mechanism is to precisely control the release speed of the rotor kinetic energy of the wind turbine when the grid frequency is disturbed, avoiding excessive release of the rotor kinetic energy and resulting in too rapid a decrease in the speed of the wind turbine, thus maintaining the stability and safety of the unit. Specifically, this mechanism designs a dynamic release strategy based on the degree of grid frequency drop, and realizes precise regulation of the wind turbine in the face of frequency fluctuations through refined control strategies.

[0194] In a weak grid environment, frequent grid fluctuations may cause the wind turbine to frequently participate in virtual inertia support. If the rotor kinetic energy is released too quickly during this process, it may lead to a sharp drop in the speed of the wind turbine, further affecting the normal operation of the unit and even causing equipment failures. Therefore, adopting the kinetic energy hierarchical release mechanism and dynamically adjusting the release rate of the rotor kinetic energy according to the amplitude of frequency change, fluctuation characteristics, etc. can effectively avoid the negative impacts brought by excessive release and improve the stability and safety of the system.

[0195] The design of this mechanism is based on the degree of grid frequency drop and adopts a step-by-step release strategy. The specific steps are as follows:

[0196] Frequency drop threshold: Set a frequency drop threshold Δf th , which is used to determine whether the grid needs to release kinetic energy. If the grid frequency f(t) deviates from the rated frequency Δf(t)≥Δf th , the kinetic energy release strategy is triggered.

[0197] According to the severity of the grid frequency drop, multiple kinetic energy release levels are divided. Assume that the range of the grid frequency deviation Δf(t) is [Δf min , Δf max , then the kinetic energy release amount can be graded according to the following rules:

[0198]

[0199] Among them, k1 and k2 are coefficients dynamically adjusted according to the drop amplitude of the grid frequency. The release amount ΔP release1 (t) is proportional to the frequency change amount and its change rate, and the release amount will gradually increase as the frequency fluctuation intensifies.

[0200] If the frequency deviation is greater than the set upper limit, a larger amount of kinetic energy is released and a higher-level control mode is entered. At this time, the amount of kinetic energy released is further amplified according to the following formula:

[0201]

[0202] where k3 and k4 are adjustment parameters for the higher control level.

[0203] During the kinetic energy release process, the rotational speed will change accordingly. To prevent the rotational speed from dropping too fast, it is necessary to control the kinetic energy release rate of the rotor to match the amplitude of the grid frequency change. The kinetic energy release rate and the rotational speed change rate are related as follows:

[0204]

[0205] Therefore, during the kinetic energy release process, it is necessary to avoid excessive drop in the rotational speed of the wind turbine by precisely controlling to avoid excessive drop in the rotational speed of the wind turbine.

[0206] By stepwise adjusting the kinetic energy release amount of the wind turbine, the strength of the inertia support can be flexibly adjusted according to the change of the grid frequency. The implementation steps of this strategy in the actual control system are as follows:

[0207] S701. Monitor the grid frequency in real time

[0208] Collect the grid frequency and the frequency change rate to calculate the frequency deviation.

[0209] S702. Judge the frequency deviation range

[0210] Judge the intensity of the grid frequency fluctuation according to the frequency deviation and the change rate, and determine the kinetic energy release level to be taken. If the deviation exceeds the set threshold, start the kinetic energy release.

[0211] S703. Release the rotor kinetic energy in stages

[0212] Release different levels of rotor kinetic energy in sequence according to the change degree of the grid frequency. A smaller frequency deviation only releases less kinetic energy, while a larger frequency deviation releases more kinetic energy.

[0213] S704. Adjust the rotational speed change

[0214] During the kinetic energy release process, by controlling the rotational speed change rate ensure that the rotational speed does not drop sharply and avoid the wind turbine from stalling or overloading.

[0215] S705. Real-time feedback adjustment

[0216] According to the recovery of the grid frequency, the kinetic energy release rate is dynamically adjusted to avoid excessive kinetic energy release and ensure the smooth operation of the wind turbine.

[0217] This invention precisely controls inertia release, releasing rotor kinetic energy in stages based on the degree of grid frequency fluctuation. This prevents a sharp drop in speed caused by excessive inertia release and ensures the stability of the wind turbine. Furthermore, this mechanism can flexibly respond to grid disturbances of varying intensities, ensuring smooth system operation in both strong and weak disturbances. Furthermore, by limiting the excessive release of kinetic energy, it effectively protects the mechanical components of the wind turbine and improves its operational safety.

[0218] To further enhance the frequency stability of the inertia support strategy, this paper introduces an additional frequency damping control mechanism. This mechanism optimizes damping compensation to enhance the wind turbine's ability to suppress grid frequency fluctuations, particularly in the event of large frequency disturbances or frequency oscillations. Specifically, additional frequency damping control monitors and dynamically adjusts frequency oscillations in real time, adjusting the wind turbine's damping coefficient appropriately to optimize its response to frequency fluctuations. This allows it to provide inertia support while effectively avoiding excessive oscillations and frequency instability.

[0219] In practical applications, frequency fluctuations can cause grid system instability, especially when the frequency fluctuations are large. Inertial support systems can cause frequency oscillations due to excessive inertia release or slow frequency response. To address this issue, this invention introduces additional frequency damping control, combining virtual inertia with additional damping to achieve a smoother and more stable system response to frequency fluctuations. By adjusting the wind turbine's damping gain and compensation strategy, the ability to quickly suppress frequency oscillations is enhanced, thereby improving system stability and reliability.

[0220] Furthermore, the present invention combines the secondary control link and adds additional damping control. This additional damping control targets the frequency oscillation problem that may be caused during the virtual inertia support process, monitors the oscillation in real time, and dynamically adjusts the damping parameters. Let the additional damping coefficient be D a , according to the frequency oscillation amplitude A and the rate of change For example, an adaptive algorithm is used: l is a pre-designed function. In the control loop, the adjusted control signal Suppress oscillations at their root. Cooperating with the previous dynamic damping compensation mechanism, it forms a double guarantee, further improving system stability and ensuring that the grid frequency remains stable after the virtual inertia support is introduced.

[0221] Beneficial effects: The present invention has the following advantages:

[0222] By introducing an adaptive virtual inertia regulation mechanism, a multi-modal collaborative control mechanism, and a dynamic damping compensation method, the present invention significantly improves the response speed of wind turbines to grid frequency fluctuations and system stability, especially in weak grid environments. The adaptive virtual inertia regulation mechanism precisely controls the inertia release, avoiding a sharp drop in rotational speed caused by excessive inertia release and effectively protecting the mechanical components and operation safety of wind turbines. At the same time, additional frequency damping control further suppresses frequency oscillations and enhances system stability. The multi-modal collaborative control mechanism automatically switches control modes according to the intensity and change trend of grid frequency disturbances, ensuring that wind turbines can provide optimal responses in different grid states and improving system adaptability. In addition, the combination of a multi-time scale collaborative control strategy and a rotor kinetic energy hierarchical release strategy ensures the timely response of wind turbines in the short, medium, and long terms, maximizing energy output and reducing energy waste. These mechanisms not only improve the adaptability of wind turbines to grid frequency fluctuations but also enhance their operation reliability and safety in weak grid environments, providing technical support for the stable grid connection and sustainable development of wind turbines. While improving system response speed, stability, and adaptability, the present invention greatly enhances the collaborative operation safety and reliability of wind turbines and the grid, laying a solid technical foundation for the smooth and safe access of large-scale wind power to weak grids, and having important pioneering and practical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0223] Figure 1 is a framework diagram of the virtual inertia enhancement control strategy for wind turbines;

[0224] Figure 2 is a schematic structural diagram of the main control system of a wind turbine after embedding a control strategy module;

[0225] Figure 3 is a schematic diagram of the principle of the adaptive virtual inertia regulation mechanism;

[0226] Figure 4 is a working state diagram of the multi-modal collaborative control mechanism;

[0227] Figure 5 is a timing diagram of the multi-time scale collaborative control; DETAILED DESCRIPTION OF THE EMBODIMENTS

[0228] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0229] Embodiment

[0230] This embodiment is a distributed wind turbine virtual inertia active support control method applicable to weak grids, as Figure 1 shown, and includes the following steps:

[0231] S1. System architecture construction:

[0232] Embed a virtual inertia enhancement control strategy module in the main control system of the wind turbine. This module is connected to existing hardware devices such as the wind turbine monitoring system and the power conversion system to obtain key parameter information in real time and accurately;

[0233] S2. Real-time monitoring and data acquisition:

[0234] Continuously and real-time monitor the grid frequency and its rate of change. After the sensor converts the collected analog signal into a digital signal, it is transmitted to the control strategy module in the main control system; at the same time, collect the operating state parameters of the wind turbine to provide comprehensive data support for subsequent control decisions;

[0235] S3. Adaptive virtual inertia coefficient calculation:

[0236] Based on the fuzzy logic control algorithm, comprehensively analyze the collected operating state parameters in the virtual inertia enhancement control strategy module; according to the pre-set fuzzy rule base, calculate the most suitable adaptive inertia coefficient at the current moment through fuzzy inference;

[0237] S4. Multi-modal collaborative control switching:

[0238] The main control system compares the frequency deviation and its rate of change calculated in real time with the preset modal switching threshold to switch the operating mode of the wind turbine;

[0239] The operating modes of the wind turbine include:

[0240] (1) Static support mode

[0241] The triggering condition for this mode is that the frequency deviation |Δf| of the power grid is small and the frequency rate of change is low;

[0242] In this mode, both the adaptive inertia coefficient and the inertia support power are at their minimum values;

[0243] K in,static = K0

[0244]

[0245] (2) Dynamic support mode

[0246] The triggering condition for this mode is that the frequency deviation |Δf| of the power grid is large and the frequency rate of change is small;

[0247] In this mode, the adaptive inertia coefficient K in is dynamically adjusted as follows:

[0248]

[0249] (3) Fast frequency response mode

[0250] The triggering condition of this mode is that the frequency deviation |Δf| of the power grid is large, and the frequency change rate is high;

[0251] In this mode, the adaptive inertia coefficient K in increases rapidly as follows:

[0252] K in,fast = K in,max ·e -γt ;

[0253] S5. Implementation of dynamic damping compensation:

[0254] According to the monitoring results of the frequency oscillation trend, calculate the equivalent virtual damping coefficient using the built-in algorithm; in the control loop, introduce a damping force matching the oscillation trend by adjusting the controller parameters or activating the compensator;

[0255] The virtual damping coefficient is calculated as follows:

[0256]

[0257] Where:

[0258] K d (t) is the virtual damping coefficient at the current moment;

[0259] K d0 is the initial virtual damping coefficient;

[0260] ΔP wind is the change in the power output of the wind turbine;

[0261] T f is the duration of the power grid frequency deviation;

[0262] γ1, γ2, γ3, γ4 are coefficient adjustment parameters, which are adjusted according to the adaptive weighting mechanism as follows:

[0263]

[0264] Where:

[0265] γ i0 is the initial weighting coefficient;

[0266] δ i is the adjustment factor;

[0267] The controller parameters are adjusted according to the following differential response:

[0268]

[0269] Among them, P damping (t) represents the control power provided by dynamic damping compensation;

[0270] The compensator outputs as follows:

[0271]

[0272] Among them:

[0273] u(t) is the control signal;

[0274] e(t) is the error.

[0275] S6. Multi-time scale collaborative control execution:

[0276] In the short-term inertia support stage in milliseconds, once a sudden change in the grid frequency is detected, the virtual inertia enhancement control strategy module immediately issues an instruction to the generator control system to quickly adjust the electromagnetic torque; in the medium-term power regulation period in seconds, according to the medium-term change trend of the grid frequency and the power generation capacity of the unit, the power reference value is optimized, and the grid frequency stability is maintained by adjusting the output of the converter; in the long-term speed recovery process in minutes, the virtual inertia enhancement control strategy module slowly adjusts the power generation power according to the deviation between the current speed of the wind turbine and the rated speed, so that the speed of the wind turbine gradually recovers to the normal level to ensure the continuous and stable operation of the wind turbine; specifically as follows:

[0277] (1) Short-term inertia support stage in milliseconds

[0278] S(1-1). Construct a phase-locked loop through the grid frequency signal and the output frequency signal of the wind turbine generator set to detect and track the phase change of the grid frequency in real time;

[0279] S(1-2). The phase-locked loop control system adjusts the output frequency of the wind turbine generator set according to the phase deviation of the grid frequency to synchronize it with the grid frequency to eliminate the frequency deviation;

[0280] S(1-3). When the grid frequency undergoes a sudden change or violent fluctuation, the phase-locked loop quickly adjusts the output power of the wind turbine generator set through rapid response to provide short-term inertia support;

[0281] The output of the phase-locked loop is combined with the virtual inertia response of the wind turbine generator set. The combination method is as follows:

[0282] a. Calculate the phase error of the phase-locked loop

[0283] Set the phase error between the grid frequency and the output frequency of the wind turbine generator set This phase error is calculated by the following formula:

[0284]

[0285] Among them, θ grid (t) and θ wind (t) are the phases of the grid frequency and the output frequency of the wind turbine respectively;

[0286] b. PLL feedback control

[0287] According to the phase error to guide the control system to adjust, the feedback control of the phase-locked loop is achieved through the following formula:

[0288]

[0289] Among them, ω PLL is the control gain of the phase-locked loop;

[0290] c. Adjust the output power of the wind turbine according to the change of the grid frequency

[0291] After the phase-locked loop achieves frequency synchronization, the output power P wind (t) of the wind turbine is adjusted according to the change of the grid frequency, and the adjustment method is as follows:

[0292] P wind (t) = ΔP sync (t) + ΔP inertia (t)

[0293] Among them, ΔP sync (t) is the synchronous power adjustment amount, and ΔP inertia (t) is the inertia support power adjustment amount;

[0294] The synchronous power adjustment amount ΔP sync (t) is calculated as follows:

[0295] ΔP sync (t) = k sync ·Δf(t)

[0296] Among them, k sync is the synchronous adjustment coefficient;

[0297] The inertia support power adjustment amount ΔP inertia (t) is calculated as follows:

[0298]

[0299] Among them, I wind is the rotor inertia of the wind turbine, is the rotational speed change rate of the wind turbine;

[0300] (2) Second-level medium-term power adjustment period

[0301] Adjust the active power output of the wind turbine through a PI controller according to the changing trend of the grid frequency as follows:

[0302] P(t) = P nom ·(1 - β·Δf(t))

[0303] where P(t) is the active power output of the wind turbine, P nom is the rated power of the wind turbine, β is the frequency response coefficient, and Δf(t) is the current frequency deviation;

[0304] The adjustment by the PI controller is carried out as follows:

[0305]

[0306] where K p and K i are the proportional and integral gains;

[0307] (3) Long - term speed recovery process in minutes

[0308] When the frequency recovers to the set range, the wind turbine gradually recovers its speed according to the recovery speed of the grid frequency as follows:

[0309]

[0310] where ω nom is the rated speed of the wind turbine, γ is the adjustment factor, f set is the grid - set frequency, and Δf(t) is the current frequency deviation.

[0311] S7. Rotor kinetic energy hierarchical release operation:

[0312] Preset multiple hierarchical thresholds for the grid frequency drop, corresponding to different kinetic energy release levels; when it is monitored that the grid frequency drops to a certain threshold range, gradually increase the intensity of the rotor kinetic energy release instruction according to the preset hierarchical strategy;

[0313] The specific method is as follows:

[0314] S701. Monitor the grid frequency in real - time

[0315] Collect the grid frequency and the frequency change rate, and calculate the frequency deviation;

[0316] S702. Judge the frequency deviation range

[0317] Judge the intensity of the grid frequency fluctuation according to the frequency deviation and the change rate, and determine the kinetic energy release level that needs to be taken. If the deviation exceeds the set threshold, start the kinetic energy release;

[0318] Set the threshold value Δf for frequency drop th , if the grid frequency f(t) deviates from the rated frequency Δf(t) ≥ Δf th , then trigger the kinetic energy release;

[0319] If the range of the grid frequency deviation Δf(t) is [Δf min , Δf max , then the kinetic energy release amount is calculated as follows:

[0320]

[0321] where k1 and k2 are coefficients dynamically adjusted according to the drop amplitude of the grid frequency;

[0322] If the grid frequency deviation Δf(t) is greater than the set upper limit, a greater amplitude of kinetic energy release is carried out, and the calculation is as follows:

[0323]

[0324] where k3 and k4 are adjustment parameters at a higher control level;

[0325] During the kinetic energy release process, in order to prevent the rotational speed from dropping too fast, control the kinetic energy release rate of the rotor to match the amplitude of the grid frequency change, as follows:

[0326]

[0327] where, is the kinetic energy release rate, is the rotational speed change rate;

[0328] S703. Gradually release the rotor kinetic energy

[0329] Release different levels of rotor kinetic energy in sequence according to the change degree of the grid frequency;

[0330] S704. Adjust the rotational speed change

[0331] During the kinetic energy release process, by controlling the rotational speed change rate ensure that the rotational speed does not drop sharply and avoid the wind turbine from stalling or overloading;

[0332] S705. Real-time feedback adjustment

[0333] According to the recovery situation of the grid frequency, dynamically adjust the kinetic energy release rate, avoid excessive kinetic energy release, and ensure the stable operation of the wind turbine.

[0334] S8. Realization of additional frequency damping control:

[0335] Combined with the secondary control link, an additional damping control unit is added to the generator control system to monitor the frequency oscillation during the virtual inertia support process in real time. By comparing with the preset oscillation suppression target, the additional damping parameters are dynamically adjusted.

[0336] The virtual inertia enhancement control strategy module adopted therein uses an adaptive virtual inertia adjustment mechanism, including the following steps:

[0337] S101. Collect the grid frequency data and the rate of change in real time, and calculate the frequency deviation;

[0338] S102. Use a fuzzy controller to calculate the inertia adjustment factor ΔK in ;

[0339] S103. Calculate the adaptive inertia coefficient;

[0340] The adaptive inertia coefficient is dynamically calculated by using fuzzy logic control combined with a non - linear mapping method, as follows:

[0341]

[0342] Among them, the function f(·) is mapped to the inertia coefficient through fuzzy control rules, Δf is the frequency deviation, is the rate of change of frequency;

[0343] S104. Calculate the inertia support power, as follows:

[0344]

[0345] S105. Send a control instruction to the wind turbine converter to adjust the active power output and provide inertia support.

[0346] The above - mentioned adaptive inertia coefficient K in is adjusted according to the grid state, as follows:

[0347]

[0348] Among them:

[0349] K0 is the reference inertia coefficient;

[0350] α is the frequency deviation influence coefficient;

[0351] β is the rate of change of frequency influence coefficient;

[0352] The inertia adjustment factor ΔK in follows the following fuzzy rules:

[0353] When Δf is very small and is very small, then ΔK in is close to zero;

[0354] When Δf is moderate and relatively large, then ΔK in increases moderately;

[0355] When Δf is large and very large, then ΔK in increases significantly;

[0356] And it satisfies:

[0357] K in = K0 + ΔK in .

[0358] As Figure 1 shown is the framework diagram of the virtual inertia enhancement control strategy module, which shows the overall control system architecture of the present invention. The core idea of this control strategy framework is to improve the response ability of the wind turbine to the grid frequency fluctuation through the virtual inertia enhancement mechanism, thereby enhancing the system stability. The entire framework includes multiple key control modules, each module works independently but collaborates with each other to ensure that the wind turbine can flexibly adjust its response strategy according to different characteristics of the grid frequency change. The system first monitors the grid frequency and the frequency change rate in real time to evaluate the intensity of the grid disturbance. Then, according to the change of the grid frequency, the system automatically switches to different control modes, including the static support mode, the dynamic support mode and the frequency fast response mode, to ensure the best response under different disturbance conditions. In each control mode, the control system dynamically adjusts the virtual inertia coefficient and the damping compensation parameter according to the grid frequency fluctuation, avoids excessive inertia release and suppresses the frequency oscillation in time. At the same time, the system adopts a hierarchical kinetic energy release strategy according to the operating state of the wind turbine and the intensity of the grid disturbance to ensure the smooth release of the rotor kinetic energy and prevent the problem of rapid speed drop of the wind turbine. The entire control strategy framework is implemented in an embedded manner in the main control system of the wind turbine, can adaptively adjust the control parameters, maximize the performance of the wind turbine, and effectively improve the system stability in a weak grid environment.

[0359] As Figure 2As shown in the figure, in the overall architecture after the control strategy module of the present invention is embedded in the main control system of the wind turbine generator set, the core part is the newly added virtual inertia enhancement control strategy module, which is closely integrated with the main control system and becomes the key brain for the operation of the entire system. On the one hand, this module is connected to multiple high-precision sensors through high-speed data transmission lines. These sensors are widely distributed around the wind turbine generator set and the grid access point, and are used to collect various key parameters such as grid frequency, frequency change rate, wind speed, generator speed, and output power in real time. The collected grid frequency and change rate information are directly sent to the adaptive virtual inertia regulation sub-module to provide a data basis for accurately calculating the inertia coefficient; the operation state parameters of the wind turbine generator set such as wind speed, speed, and power are synchronously transmitted to the multi-modal collaborative control sub-module, the multi-time scale collaborative control sub-module, and the rotor kinetic energy hierarchical release sub-module, etc., to ensure that each control strategy can be flexibly adjusted according to the real-time working conditions of the unit.

[0360] On the other hand, a direct command interaction channel is also established between the virtual inertia enhancement control strategy module and the power conversion system and the electromagnetic torque control system of the wind turbine generator set. When the adaptive virtual inertia regulation sub-module calculates the appropriate inertia support parameters, it quickly sends commands to the power conversion system to accurately adjust the output power and achieve inertia support; the multi-modal collaborative control sub-module judges the mode according to the frequency deviation and change rate and issues corresponding mode power output commands to the power conversion system; the dynamic damping compensation sub-module and the additional frequency damping control sub-module continuously monitor the frequency oscillation situation. Once an abnormality is found, a damping adjustment command is immediately sent to the electromagnetic torque control system to quickly suppress the oscillation by adjusting the electromagnetic torque and ensure the stable operation of the system. Each sub-module has a clear division of labor and cooperation, and jointly relies on the main control system platform to achieve efficient and intelligent control of the wind turbine generator set in a weak grid environment.

[0361] As Figure 3The figure shows a schematic diagram of the adaptive virtual inertia regulation mechanism, demonstrating how a wind turbine dynamically adjusts the virtual inertia coefficient based on grid frequency fluctuations and rate-of-change information to enhance its response ability to grid frequency disturbances. The core of this mechanism lies in the real-time acquisition of the grid frequency and its rate of change, and the calculation of an adaptive inertia coefficient through a fuzzy logic control algorithm. When the grid disturbance is strong, the inertia coefficient increases to provide greater inertia support and help smooth out grid frequency fluctuations; when the disturbance is small or the frequency fluctuation is relatively stable, the inertia coefficient decreases to avoid unnecessary excessive release of inertia. The system evaluates the disturbance intensity in real time based on the rate of change (i.e., the speed of frequency change) and amplitude of the grid frequency, and adjusts the response strategy of the virtual inertia to achieve dynamic control of "large inertia for high deviation and small inertia for low deviation". In addition, this mechanism also takes into account the operating state of the wind turbine to ensure reasonable adjustment of the virtual inertia under different working conditions, being able to quickly respond to large-frequency fluctuations and avoid over-regulation of the system, thus ensuring the stability and safety of the wind turbine under different grid conditions. Through this mechanism, the wind turbine can more flexibly adapt to changes in grid frequency, improve its adaptability to grid disturbances, ensure the stability of the grid, and reduce the losses of the mechanical components of the wind turbine.

[0362] As Figure 4 The figure shows the working state diagram of the multi-modal cooperative control mechanism, demonstrating how a wind turbine achieves more efficient virtual inertia enhancement control by different control modes in response to different fluctuations of the grid frequency. Based on the grid frequency and its rate of change, this mechanism designs three main control modes: static support mode, dynamic support mode, and frequency rapid response mode. The working principle and applicable situation of each mode are different, aiming to ensure that the system can provide the optimal response strategy under different disturbance conditions. In the static support mode, when the grid frequency fluctuates slightly and is stable, the wind turbine mainly provides stable inertia support to maintain the basic stability of the grid; in this mode, the inertia coefficient remains at a relatively low level to avoid excessive release of inertia. In the dynamic support mode, when the grid frequency fluctuates greatly, the system increases the inertia coefficient to provide stronger support and combines a dynamic damping compensation mechanism to suppress frequency fluctuations. In the frequency rapid response mode, in response to sudden rapid frequency fluctuations, the system can quickly adjust the virtual inertia coefficient to rapidly respond to frequency changes and ensure the rapid recovery of the grid frequency. The switching of each mode is automatic, based on real-time grid frequency change data and the operating state of the wind turbine. This multi-modal cooperative control mechanism can flexibly adapt to various grid conditions, ensure the stable operation of the wind turbine under various grid disturbance conditions, and improve the adaptability and reliability of the wind turbine in a weak grid environment through precise inertia regulation and dynamic response.

[0363] As Figure 5The following is the timing diagram of multi-time scale coordinated control of this embodiment. The time axis of multi-time scale coordinated control and the control actions at each stage closely revolve around the different stages and time characteristics of power grid frequency fluctuations. At the starting end of the time axis, once a frequency disturbance occurs in the power grid, the short-term inertia support stage in milliseconds immediately starts. At this time, the high-precision sensors on the wind turbine quickly sense the instantaneous change of the power grid frequency and transmit the signal to the short-term inertia support module in the main control system. Based on the pre-set fast response strategy, this module quickly adjusts the electromagnetic torque of the wind turbine, prompting the rotor to release or store kinetic energy in an extremely short time. Its control action focuses on instantly preventing the rapid expansion of frequency deviation and ensuring that the power grid frequency will not get "out of control" instantly.

[0364] As time goes on and enters the medium-term power regulation stage in seconds, the medium-term power regulation module in the main control system takes over the control right. It continuously pays attention to the medium-term change trend of the power grid frequency. By analyzing the collected frequency data and using strategies such as droop control, it accurately adjusts the power generation of the unit according to the frequency deviation and outputs the power adjustment amount. The key to this stage is to stabilize the power grid frequency and make it gradually return to a reasonable range.

[0365] When time further extends to the long-term speed recovery stage in minutes, the long-term speed recovery module starts to play a role. It adjusts the electromagnetic torque slowly according to the current speed of the wind turbine and the impact on the speed caused by the previous inertia support stage, and gradually restores the wind turbine speed to the normal level, ensuring that the wind turbine can operate continuously and stably in the future and making preparations for coping with possible power grid disturbances next time. The whole process advances orderly along the time axis, and the control actions at each stage cooperate with each other and complement each other, realizing the all-round and whole-process regulation of the power grid frequency.

Claims

1. A virtual inertia active support control method for distributed wind turbines applicable to weak power grids, characterized in that, It includes the following steps: S1. System architecture construction: Embed a virtual inertia enhancement control strategy module in the main control system of the wind turbine. This module is connected to existing hardware devices such as the wind turbine monitoring system and the power conversion system to obtain key parameter information in real time and accurately; S2. Real-time monitoring and data acquisition: Continuously and real-time monitor the grid frequency and its rate of change. After the sensor converts the collected analog signal into a digital signal, it is transmitted to the control strategy module in the main control system; at the same time, collect the operating state parameters of the wind turbine to provide comprehensive data support for subsequent control decisions; S3. Adaptive virtual inertia coefficient calculation: Based on the fuzzy logic control algorithm, comprehensively analyze the collected operating state parameters in the virtual inertia enhancement control strategy module; according to the pre-set fuzzy rule base, calculate the most suitable adaptive inertia coefficient at the current moment through fuzzy inference; S4. Multi-modal collaborative control switching: The main control system compares the frequency deviation and its rate of change calculated in real time with the preset modal switching threshold, and switches the operating mode of the wind turbine; S5. Dynamic damping compensation implementation: According to the monitoring results of the frequency oscillation trend, calculate the equivalent virtual damping coefficient using the built-in algorithm; in the control loop, introduce a damping force matching the oscillation trend by adjusting the controller parameters or activating the compensator; S6. Multi-time scale collaborative control execution: In the short-term inertia support stage at the millisecond level, once a sudden change in the grid frequency is detected, the virtual inertia enhancement control strategy module immediately issues an instruction to the generator control system to quickly adjust the electromagnetic torque; in the medium-term power regulation period at the second level, optimize the power reference value according to the medium-term change trend of the grid frequency and the power generation capacity of the unit, and maintain the grid frequency stability by adjusting the output of the converter; in the long-term speed recovery process at the minute level, the virtual inertia enhancement control strategy module slowly adjusts the power generation power according to the deviation between the current speed of the wind turbine and the rated speed, so that the speed of the wind turbine gradually recovers to the normal level to ensure the continuous and stable operation of the wind turbine; S7. Rotor kinetic energy hierarchical release operation: Preset multiple hierarchical thresholds for grid frequency drop, corresponding to different kinetic energy release levels; when it is detected that the grid frequency drops to a certain threshold range, increase the intensity of the rotor kinetic energy release instruction step by step according to the preset hierarchical strategy; S8. Additional frequency damping control implementation: Combined with the secondary control link, add an additional damping control unit in the generator control system to real-time monitor the frequency oscillation situation during the virtual inertia support process, and dynamically adjust the additional damping parameters by comparing with the preset oscillation suppression target.

2. The distributed wind turbine virtual inertia active support control method applicable to a weak power grid according to claim 1, wherein The virtual inertia enhancement control strategy module adopts an adaptive virtual inertia adjustment mechanism, including the following steps: S101. Real-time collect grid frequency data and its rate of change, and calculate the frequency deviation; S102. Calculate the inertia adjustment factor ΔK using a fuzzy controller in ; S103. Calculate the adaptive inertia coefficient; Dynamically calculate the adaptive inertia coefficient using the fuzzy logic control combined with the non-linear mapping method as follows: Among them, the function f(·) is mapped to the inertia coefficient through fuzzy control rules, Δf is the frequency deviation, and is the rate of change of frequency; S104. Calculate the inertia support power as follows: S105. Send a control instruction to the wind turbine converter to adjust the active power output and provide inertia support.

3. The virtual inertia active support control method for a distributed wind turbine applicable to a weak power grid according to claim 2, characterized in that, The adaptive inertia coefficient K in is adjusted according to the grid state as follows: Wherein: $K_0$ is the reference inertia coefficient; $\alpha$ is the frequency deviation influence coefficient; $\beta$ is the frequency change rate influence coefficient; The inertia adjustment factor ΔK in According to the following fuzzy rules: When Δf is very small and is very small, then ΔK in is close to zero; When Δf is moderate and relatively large, then ΔK in increases moderately; When Δf is large and is very large, then ΔK in increases significantly; And it satisfies: K in = K0 + ΔK in ; To further improve the stability of inertia control, combined with the time-varying inertia adjustment strategy, construct the following inertia response function: K in (t) = K in,max ·e -γt + K in,min Where: K in,max is the maximum inertia coefficient; K in,min is the minimum inertia coefficient; $\gamma$ is the inertia decay factor; $t$ is the inertia response duration; This formula shows that the inertia coefficient rapidly increases at the initial stage of the disturbance and gradually decays after a certain time to ensure the flexibility of inertia support and the stability of the system.

4. The distributed wind turbine virtual inertia active support control method applicable to a weak power grid according to claim 1, wherein, The operating modes of the wind turbine in step S4 include: (1) Static support mode The triggering condition of this mode is that the frequency deviation |Δf| of the power grid is small and the frequency change rate is low. In this mode, both the adaptive inertia coefficient and the inertia support power are at the minimum values; K in,static = K0 (2) Dynamic support mode The triggering condition of this mode is that the frequency deviation |Δf| of the power grid is large and the frequency change rate is small; In this mode, the adaptive inertia coefficient K in is dynamically adjusted as follows: (3) Frequency fast response mode The triggering condition of this mode is that the frequency deviation |Δf| of the power grid is large and the rate of change of frequency is high; In this mode, the adaptive inertia coefficient K in rapidly increases as follows: K in,fast = K in,max ·e -γt 。 5. The active support control method for virtual inertia of a distributed wind turbine applicable to a weak power grid according to claim 1, characterized in that The virtual damping coefficient in step S5 is calculated as follows: Where: K d (t) is the virtual damping coefficient at the current moment; K d0 is the initial virtual damping coefficient; ΔP wind is the change in the power output of the wind turbine; T f is the duration of the power grid frequency deviation; $\gamma_1$, $\gamma_2$, $\gamma_3$, $\gamma_4$ are coefficient adjustment parameters, which are adjusted according to the adaptive weighting mechanism, as follows: Where: γ i0 is the initial weighting coefficient; δ i is the adjustment factor.

6. The active support control method for virtual inertia of a distributed wind turbine applicable to a weak power grid according to claim 5, characterized in that The controller parameters are adjusted according to the following differential response: where P damping (t) represents the control power provided by dynamic damping compensation; The compensator outputs as follows: Where: $u(t)$ is the control signal; $e(t)$ is the error.

7. The distributed wind turbine virtual inertia active support control method applicable to a weak power grid according to claim 1, wherein The specific method of step S6 is as follows: (1) Millisecond-level short-term inertia support stage S(1 - 1), Construct a phase-locked loop through the grid frequency signal and the output frequency signal of the wind turbine to detect and track the phase change of the grid frequency in real time; S(1 - 2), The phase-locked loop control system adjusts the output frequency of the wind turbine according to the phase deviation of the grid frequency to make it synchronized with the grid frequency to eliminate the frequency deviation; S(1 - 3), When the grid frequency suddenly changes or fluctuates violently, the phase-locked loop adjusts the output power of the wind turbine through rapid response to provide short-term inertia support; (2) Second-level medium-term power adjustment period According to the change trend of the grid frequency, adjust the active power output of the wind turbine through a PI controller, as follows: P(t) = P nom ·(1 - β·Δf(t)) Among them, P(t) is the active power output of the wind turbine, P nom is the rated power of the wind turbine, β is the frequency response coefficient, and Δf(t) is the current frequency deviation; The adjustment through the PI controller is carried out as follows: where K p and K i are the proportional and integral gains; (3) Minute-level long-term speed recovery process When the frequency recovers to the set range, the wind turbine gradually recovers its speed according to the recovery speed of the grid frequency, as follows: where ω nom is the rated speed of the wind turbine, γ is the adjustment factor, f set is the grid-set frequency, and Δf(t) is the current frequency deviation.

8. The distributed wind turbine virtual inertia active support control method applicable to a weak power grid according to claim 7, wherein The output of the phase-locked loop is combined with the virtual inertia response of the wind turbine, and the combination method is as follows: a. Calculate the phase error of the phase-locked loop Set the phase error between the grid frequency and the output frequency of the wind turbine This phase error is calculated by the following formula: where, θ grid (t) and θ wind (t) are the phases of the grid frequency and the output frequency of the wind turbine, respectively; b. PLL feedback control According to the phase error Guide the control system to adjust, and the feedback control of the phase-locked loop is achieved through the following formula: where ω PLL is the control gain of the phase-locked loop; c. Adjust the output power of the wind turbine according to the change of the grid frequency After the phase-locked loop achieves frequency synchronization, the output power P of the wind turbine wind (t) is adjusted according to the change of the grid frequency, and the adjustment method is as follows: P wind P(t) = ΔP sync + ΔP(t) inertia (t) Among them, ΔP sync (t) is the synchronous power adjustment amount, and ΔP inertia (t) is the inertia support power adjustment amount; Synchronous power adjustment amount ΔP sync (t) is calculated as follows: ΔP sync (t) = k sync ·Δf(t) where k sync is the synchronization adjustment coefficient; Inertia support power adjustment amount ΔP inertia (t) is calculated as follows: Among them, I wind is the rotor inertia of the wind turbine generator set, and is the rate of change of the rotational speed of the wind turbine generator set.

9. The distributed wind turbine virtual inertia active support control method applicable to a weak power grid according to claim 1, characterized in that The specific method of step S7 is as follows: S701, Monitor the grid frequency in real time Collect the grid frequency and the frequency change rate, and calculate the frequency deviation; S702, Judge the frequency deviation range Judge the intensity of the grid frequency fluctuation according to the frequency deviation and the change rate, and determine the kinetic energy release level that needs to be taken. If the deviation exceeds the set threshold, start the kinetic energy release; S703, Release the rotor kinetic energy in stages Release different levels of rotor kinetic energy in sequence according to the change degree of the grid frequency; S704, Adjust the speed change During the kinetic energy release process, by controlling the rate of change of rotational speed ensure that the rotational speed does not drop sharply, and avoid stall or overload conditions of the wind turbine; S705, Real-time feedback adjustment According to the recovery situation of the grid frequency, dynamically adjust the kinetic energy release rate to avoid excessive release of kinetic energy and ensure the stable operation of the wind turbine.

10. The distributed wind turbine virtual inertia active support control method applicable to a weak power grid according to claim 9, characterized in that The specific method of step S702 is as follows: Set the threshold value Δf for frequency drop th , if the grid frequency f(t) deviates from the rated frequency by Δf(t) ≥ Δf th , then trigger the kinetic energy release; If the range of the power grid frequency deviation Δf(t) is [Δf min , Δf max , then the kinetic energy release amount is calculated as follows: Where, $k_1$ and $k_2$ are coefficients dynamically adjusted according to the drop amplitude of the grid frequency; If the power grid frequency deviation Δf(t) is greater than the set upper limit, a greater amount of kinetic energy is released, calculated as follows: where k3 and k4 are regulation parameters at a higher control level; During the kinetic energy release process, to prevent the rotational speed from dropping too fast, the kinetic energy release rate of the rotor is controlled to match the amplitude of the power grid frequency change, as follows: Among them, is the kinetic energy release rate, is the rotational speed change rate.

Images