Coordination control method of wind storage combined black-start power supply based on inertia response

Abstract

The invention relates to the technical field of power system operation monitoring and control, and provides a coordination control method of a wind storage combined black-start power supply based on inertia response. According to the method, a wind generating set and a flywheel energy storage system are configured to be black-start power supplies, a virtual inertia time constant and a power demand are calculated by monitoring power grid frequency deviation in real time, and the flywheel array action depth is determined. The power output of the flywheel array is controlled by adopting a hierarchical control strategy, the charging and discharging state of the wind storage system is adjusted according to the frequency change, the SOC state of the flywheel energy storage system is monitored in real time, and the SOC is balanced by utilizing a Nash bargaining solution negotiation algorithm. In the black-start process, the power output of the wind storage system is dynamically adjusted according to the frequency, and the two cooperate to maintain the power grid frequency stable. The method effectively solves the problems of power grid black start and frequency stability under the high proportion of new energy power generation, prevents energy storage from being overcharged and overdischarged, improves the power grid black start success rate and operation stability, and promotes the efficient utilization of new energy.

Description

Technical Field

[0001] The present invention relates to the technical field of power system operation monitoring and control, and specifically relates to a coordinated control method for a wind-storage combined black-start power source based on inertia response. Background Art

[0002] In the field of power systems, with the continuous increase in the proportion of new energy power generation, the frequency stability of the power grid faces severe challenges. New energy power generation (such as wind power generation) has strong randomness and volatility, and its output is unstable, which brings difficulties to the frequency control of the power grid. When a power grid failure causes a power outage, black start becomes a key link in restoring the operation of the power grid. However, traditional black-start power sources have exposed many problems in the power grid environment with a high proportion of new energy.

[0003] On the one hand, traditional black-start power sources in areas rich in wind and light are restricted by natural resources and are insufficient in quantity, making it difficult to meet the needs of the power grid for black start. On the other hand, when using a wind-solar-storage system as a black-start power source, due to the instability of the output of wind farms and photovoltaic power stations, during the black-start process, energy storage devices are prone to overcharging and over-discharging. As mentioned in the patent document CN117408164A, under the charge / discharge power constraint and energy constraint of the energy storage device, when the output of the wind farm and the photovoltaic power station is insufficient or fluctuates violently, it may cause the energy storage to be unable to continue to be utilized, thereby resulting in the failure of black start.

[0004] In terms of wind speed prediction, traditional wind speed prediction methods have problems with poor prediction accuracy. For example, the number of IMFs generated during the EMD decomposition of CEEMD is different, resulting in errors in the ensemble average, and the error further increases when summing at the end of the prediction; the VMD algorithm is sensitive to noise, and modal aliasing may occur during decomposition in the presence of noise; traditional particle swarm algorithms are prone to falling into local optima, with a fast convergence speed but low convergence accuracy. These problems make it difficult for wind speed prediction to accurately grasp the changes in wind power, unable to provide a reliable basis for the stable operation of the wind-storage system, and increasing the difficulty of power coordination control of the wind-storage system during the black-start process. Summary of the Invention

[0005] The present invention proposes a coordinated control method for a wind-storage combined black-start power source based on inertia response, aiming to solve various problems faced by the power grid under the operation with a high proportion of new energy power generation.

[0006] The coordinated control method for a wind-storage combined black-start power source based on inertia response includes the following steps:

[0007] S1. Configuration of black-start power source and setting of initial state, configure a wind turbine generator set and a flywheel energy storage system as black-start power sources, complete the initialization of both, and simultaneously obtain the real-time power and frequency at the connection points of the output ends of the wind turbine generator set and the flywheel energy storage system to the power grid;

[0008] S2. Real-time frequency monitoring and deviation calculation. Subtract the rated frequency from the real-time frequency signal at the grid connection point of the flywheel energy storage system to calculate the frequency deviation value, and determine whether the frequency deviation value exceeds the grid deadband constraint.

[0009] S3. Calculation of the power demand for inertia support. Calculate the virtual inertia time constant based on the frequency deviation value and the rated power of the flywheel energy storage system. Calculate the power demand of the flywheel energy storage system for participating in inertia support based on the virtual inertia time constant and the real-time frequency signal. At the same time, considering the power constraints of each array in the flywheel energy storage system, aggregate using the frequency response characteristics of the array to determine the action depth of each flywheel array participating in inertia support.

[0010] S4. Control of the action of the flywheel array. First, based on the action depth of each flywheel array, control the power output of each array in the flywheel energy storage system. Then, according to the actual situation of the power output, adjust the charge and discharge states of each array. At the same time, monitor the SOC state of the system in real time and dynamically adjust the action depth of each array according to the monitoring results.

[0011] S5. Coordinated control during the black start process. Dynamically adjust the output power of the wind turbine according to the change of the grid frequency. At the same time, dynamically adjust the power output of the flywheel energy storage system according to the change of the grid frequency.

[0012] Further, in S1, the wind turbine selects a doubly-fed induction generator (DFIG) as the main power generation unit, and its rated power is determined according to the target grid capacity, and a permanent magnet synchronous motor (PMSG) is configured as a backup.

[0013] Further, in S1, the wind turbine adopts a hierarchical starting strategy. First, the DC bus voltage of the converter is raised to 90% of the rated value at a slope of 0.5% / s through the pre-charge unit. During this process, the voltage fluctuation needs to be monitored in real time to ensure that it does not exceed the allowable deviation of ±2%. At the same time, the pitch angle control system starts a self-check program, verifies the initial angle of each blade through a laser rangefinder, and controls the error within the range of ±0.5°.

[0014] Further, in S1, the real-time power at the connection point between the output end of the wind turbine and the grid is obtained through the voltage sensor and current sensor at the connection point.

[0015] Further, in S1, the flywheel energy storage system adopts a modular design. The capacity of each flywheel array is 1 MWh, the speed range is 15,000 - 50,000 rpm, the response time < 100 ms, and the total capacity of the flywheel energy storage system needs to meet the inertia support demand for at least 30 minutes at the initial stage of the grid black start.

[0016] Further, in S1, for the initialization of the flywheel energy storage system, the preparation of the mechanical system and the activation of the electrical system need to be completed synchronously. Mechanically, the magnetic levitation bearing first passes 50% of the rated current to suspend the rotor. After the vibration sensor detects that the amplitude < 10μm, it is gradually increased to the operating current. Electrically, a three-stage SOC calibration is adopted: first, the open-circuit voltage is calibrated, then the internal resistance is measured by the pulse current method, and finally, the SOC estimation error is compressed to within 0.8% by combining the temperature compensation algorithm. The initial SOC of the flywheel energy storage system is 80%. The grid connection point frequency signal is collected by the synchronized phasor measurement unit PMU, and the sampling frequency ≥ 1kHz.

[0017] Further, in S1, during the black start power supply configuration and the initial state setting process, a high-speed communication network (delay < 1ms) is used to ensure that both enter the standby state synchronously, and the preparation progress of the wind turbine generator and the flywheel energy storage system is coordinated.

[0018] Further, in S2, in the standby monitoring channel of the real-time frequency monitoring system, the fast Fourier transform (FFT) technology is adopted to analyze the collected voltage signal. By decomposing the voltage signal into different frequency components, accurate frequency information is obtained for verifying the accuracy of the PMU measurement results. To improve the monitoring reliability, distributed monitoring nodes based on the Internet of Things (IoT) are also introduced. These nodes are distributed throughout the power grid, collect local frequency information and upload it to the central control system through wireless communication for fusion analysis with the PMU and FFT data.

[0019] Further, in S2, the calculation of the frequency deviation value: The real-time frequency signal collected and processed by the PMU is subtracted from the rated grid frequency to obtain the frequency deviation value. The grid dead zone constraint is dynamically adjusted according to the grid operating state, and the dead zone threshold is set to ±0.05Hz. When abnormal conditions such as grid faults or load mutations occur and the frequency change rate exceeds 1Hz / s, the dead zone threshold is relaxed to ±0.1Hz. By comparing the frequency deviation value with the dead zone threshold, if the frequency deviation value exceeds the dead zone threshold, the subsequent process of calculating the inertia support power demand is triggered; if not, the frequency is continuously monitored, and the frequency deviation calculation result is updated every 50ms to ensure real-time tracking of the grid frequency change.

[0020] Further, in S3, the calculation formula for the virtual inertia time constant:

[0021] T v =K p ·Δf+K i ·∫Δfdt;

[0022] Where, T v is the virtual inertia time constant; Δf is the frequency deviation value; K iis the integral coefficient, and its value range is 0.1 - 0.5. It is mainly used to suppress the steady-state error of frequency deviation, and its specific value needs to be adjusted in combination with the system dynamic response characteristics and stability requirements; K p is the proportional coefficient, and its value range is between 0.5 - 2.0. It needs to be dynamically calibrated according to the flywheel rated power, grid short-circuit capacity, and system inertia requirements. When the flywheel rated power is large, the grid short-circuit capacity is small, or the system inertia requirement is high, K p should take a larger value.

[0023] Furthermore, in S3, according to the calculated virtual inertia time constant T v and the real-time frequency signal, the formula for calculating the power demand of the flywheel energy storage system to participate in inertia support is:

[0024] P req = K·Δf·T v ;

[0025] K = 0.8 - 0.2·tanh(5|Δf);

[0026] P req is the power demand of the flywheel energy storage system to participate in inertia support; K is the non-linear correction coefficient. Introducing it is to more accurately simulate the inertia support characteristics of the system under different frequency deviations. It can be seen from the formula that when the frequency deviation Δf is small, tanh(5|Δf) is close to 0, and at this time K is close to 0.8, and the power demand is approximately proportional to the frequency deviation and virtual inertia, which enables the system to respond in a timely manner and provide appropriate power support when the frequency deviation is small. tanh represents the hyperbolic tangent function; when the frequency deviation Δf is large, tanh(5|Δf) is close to 1, and K is close to 0.6, realizing the saturation characteristic at large deviations, avoiding excessive power demand from impacting the system, and ensuring the stability of the system.

[0027] Furthermore, in S3, determine the action depth of each flywheel array participating in inertia support: When determining the action depth of each flywheel array participating in inertia support, in order to prevent a single array from overloading and causing equipment damage or affecting the overall stability of the system, it is necessary to set the power upper limit of each array, and the power upper limit is set to 0.2 times the rated power; the formula for calculating the action depth of each flywheel array participating in inertia support is:

[0028]

[0029] where, f n is the grid rated frequency; α is the action depth of the flywheel array. When α = 1, the flywheel array outputs full power; Δf is the frequency deviation value;

[0030] When determining the action depth, comprehensively consider the current SOC state, remaining life of each flywheel array, and the overall system stability to optimize and allocate the action depth; for flywheel arrays with a lower SOC, if full-power output according to the calculated action depth will cause over-discharge, affecting their service life or even damaging the equipment, the action depth should be appropriately reduced at this time; for arrays with a shorter remaining life, it is also necessary to control their action depth to avoid overuse. Therefore, the calculated action depth needs to be corrected:

[0031]

[0032] soc is the state of charge of the flywheel array; α′ is the corrected action depth; ω s is the soc weight of the wheel array; ω f is the weight of the remaining life; S h is the remaining life; S z is the total life. In this way, it can not only meet the power demand for system inertia support, but also optimize the use of each flywheel array to ensure the continuous and stable operation of the system.

[0033] Further, in S4, the method for controlling the power output of each array in the flywheel energy storage system:

[0034] S411. Calculate the power output value of each array. According to the calculated corrected action depth α′ of each flywheel array, combined with the power upper limit of each array, through the formula: P ar =α′×0.2×P ra Calculate the actual power output value of each flywheel array; in the formula, P ar is the actual power output; P ra is the rated power of the flywheel energy storage system;

[0035] S412. Adopt hierarchical control to achieve precise tracking, and adopt a hierarchical control strategy to achieve precise control of the power output of the flywheel array;

[0036] S4121. Top-level model predictive control. At the top level, use model predictive control (MPC) technology to predict the power demand and system state in the future for a period of time. With the minimum power tracking error and SOC balance as the optimization goals, adopt a rolling optimization strategy. According to the current system state, calculate the power reference values of each flywheel array in the future for a period of time;

[0037] The rolling optimization strategy is: the top-level MPC performs an optimization calculation every 50 ms. During the prediction process, consider factors such as the charge and discharge dynamic characteristics of the flywheel array and the grid frequency response characteristics, and establish a dynamic model of the power output of the flywheel array and a model of SOC change; through rolling optimization, solve to make the power tracking error E pThe control input sequence with the minimum weighted sum of the SOC balance index JS is taken, and the first control input in the control input sequence with the minimum weighted sum is used as the control decision at the current moment, and the power reference values of each flywheel array in the next period of time are calculated;

[0038] S4122. Bottom - layer pulse width modulation control. At the bottom layer, the power output of the flywheel array is controlled by real - time tracking through a pulse width modulation (PWM) converter. The bottom - layer pulse width modulation converter adjusts the switching frequency and duty cycle in real time according to the power reference value calculated by the top - layer MPC. When the power reference value is greater than the current actual power output value, the duty cycle of the bottom - layer pulse width modulation converter is increased to increase the power output; if the power reference value is less than the current actual power output value, the duty cycle is decreased to reduce the power output, ensuring that the error between the actual power output and the reference value is less than 2%. At the same time, the power output, current, and voltage signals are monitored in real time. Once abnormal situations such as power mutations and over - current occur, the control parameters of the bottom - layer pulse width modulation converter are immediately adjusted or the circuit is cut off to prevent equipment damage.

[0039] Further, in S4, the steps of adjusting the charge - discharge states of each array according to the actual situation of power output and real - time monitoring of the SOC state of the system include:

[0040] Adjust the charge - discharge states of each array in the flywheel energy storage system to ensure that the flywheel energy storage system can quickly respond to frequency changes during the inertia support process; when the grid frequency drops, the flywheel array discharges to release energy to support the grid frequency; when the grid frequency rises, the flywheel array charges to absorb excess energy.

[0041] Real - time monitor the SOC state of the flywheel energy storage system, and adjust the action depth of the flywheel array according to the SOC state. When the SOC is lower than 30%, the power - limit mode is triggered, and the output power of the flywheel array is reduced to 50%, and a signal is sent to the wind turbine generator to request an increase in the generated power to maintain the grid power balance; when the SOC is higher than 90%, control the flywheel array to switch to the charging mode to absorb the excess power of the grid and prevent damage to the equipment caused by over - charging. To achieve the balance of the SOC of each flywheel array, when the SOC difference is greater than 15%, the Nash bargaining solution negotiation algorithm based on game theory is started. Each flywheel array negotiates and adjusts the charge - discharge power according to its own SOC state and the overall system requirements until the SOC difference is reduced to less than 5% to ensure that the flywheel energy storage system can operate continuously and stably during the inertia support process.

[0042] Further, in S5, the power adjustment strategy of the wind turbine generator:

[0043] During the black start process, the grid frequency fluctuates greatly. It is necessary to dynamically adjust the output power of the wind turbine according to the frequency change. The frequency droop control strategy is adopted. When the frequency deviation is greater than 0.1 Hz, the output power P of the wind turbine w The calculation formula is:

[0044] P w = P in + K dr ×Δf;

[0045] P in is the initial output power; K dr is the droop coefficient; Δf is the frequency deviation;

[0046] At the same time, to further optimize the power output and frequency response characteristics, the pitch angle is adjusted according to the frequency change. When the frequency deviation is greater than 0.1 Hz, by increasing the pitch angle, the wind energy captured by the wind turbine blades is reduced, and the output power is decreased;

[0047] The pitch angle adjustment formula is:

[0048]

[0049] θ is the pitch angle; θ0 is the initial pitch angle;

[0050] When the frequency deviation is less than -0.1 Hz, the pitch angle is decreased, the wind energy capture and output power are increased, ensuring that the wind turbine continuously and stably outputs power under different frequency conditions;

[0051] The pitch angle adjustment formula is:

[0052]

[0053] Furthermore, in S5, the method for dynamically adjusting the power output of the flywheel energy storage system according to the change of the grid frequency is as follows:

[0054] When the frequency deviation is greater than 0.2 Hz, the flywheel energy storage system, relying on its fast power regulation ability, quickly releases or absorbs energy to suppress the rapid frequency fluctuation. Each flywheel array performs charge and discharge operations according to the action depth and power output control strategy determined in S4. When the frequency deviation is less than 0.1 Hz, the wind turbine mainly undertakes the power adjustment task, and the flywheel energy storage system assists in the regulation. The flywheel energy storage system and the wind turbine exchange key information such as frequency deviation, power output, and SOC status in real time through a high-speed communication network (time delay < 1 ms), and dynamically adjust their respective power output strategies according to the actual situation of the grid to achieve efficient coordination and jointly maintain the grid frequency stability. For example, when the wind turbine cannot respond to low-frequency fluctuations in a timely manner due to wind speed changes or other reasons, the flywheel energy storage system can quickly supplement or absorb power to ensure the grid stability.

[0055] In summary, due to the adoption of the above technical solutions, the beneficial technical effects of the invention content are as follows:

[0056] Through a series of designs of the wind-storage combined black-start power source and its coordinated control, the present invention realizes comprehensive benefits in aspects such as enhancing the black-start capacity of the power grid, strengthening frequency stability, optimizing energy storage utilization, adapting to complex working conditions, and promoting new energy consumption, effectively ensuring the reliable operation of the power system under the condition of high proportion of new energy.

[0057] To ensure the smooth progress of black start, a wind turbine generator set and a flywheel energy storage system are configured as the black-start power source. The wind turbine generator set can convert wind energy into electrical energy, and the flywheel energy storage system can quickly respond to provide inertia support and power regulation. Its modular design, short response time, and sufficient capacity meet the inertia support requirements of the power grid in the initial stage of black start for at least 30 minutes. The two cooperate to make up for the deficiencies of traditional black-start power sources and improve the success rate of black start.

[0058] To maintain the frequency stability of the power grid, the power grid frequency is monitored in real time. The virtual inertia time constant and power demand are calculated through complex algorithms, and the action depth and power output of the flywheel array are accurately controlled according to the frequency deviation. At the same time, the wind turbine generator set adjusts the output power and pitch angle according to the frequency change. The two cooperate to effectively suppress frequency fluctuations and solve the frequency control problem brought by the randomness and volatility of new energy power generation.

[0059] To optimize the management of the energy storage system, the SOC state of the flywheel energy storage system is monitored in real time. When the SOC is too low or too high, measures such as power limiting and switching charging modes are taken to prevent overcharging and over-discharging. The Nash bargaining solution negotiation algorithm is used to balance the SOC of each flywheel array, extending the service life of the energy storage system, improving its reliability and utilization efficiency, and ensuring the stable operation of the energy storage system.

[0060] To improve the overall performance of the system, all links from power source configuration, status monitoring to power control and coordinated operation are closely coordinated to form a complete system. The system can be dynamically adjusted according to the operating state of the power grid to adapt to different working conditions and abnormal situations. When the frequency change rate is abnormal, it can automatically adjust the control strategy to ensure the stable operation of the wind-storage system and improve the ability of the power system to cope with complex situations.

[0061] To promote the efficient utilization of new energy, through accurate wind speed prediction and efficient wind-storage coordinated control, the stability and reliability of wind power generation in black start and power grid operation are improved, which helps to increase the application proportion of new energy in the power system and promote the transformation of the energy structure towards cleaner and lower-carbon. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 It is a flowchart of a coordinated control method for a wind-storage combined black-start power source based on inertia response. Detailed implementation manners

[0063] In order to make the objectives, technical solutions and advantages of the invention content clearer and more understandable, the following further elaborates on the invention content in combination with embodiments. It should be understood that the specific embodiments described herein are only used to explain the invention content and are not used to limit the invention content.

[0064] As Figure 1 shown, the coordinated control method of the wind-storage combined black-start power source based on inertia response includes the following steps:

[0065] S1. Configuration of black-start power source and setting of initial state: Configure the wind turbine generator set and the flywheel energy storage system as the black-start power source, complete the initialization work of both, and at the same time obtain the real-time power and frequency of the connection points between the output ends of the wind turbine generator set and the flywheel energy storage system and the power grid.

[0066] S2. Real-time frequency monitoring and deviation calculation: Subtract the real-time frequency signal at the grid connection point of the flywheel energy storage system obtained from the rated frequency to calculate the frequency deviation value, and judge whether the frequency deviation value exceeds the grid dead zone constraint.

[0067] S3. Calculation of power demand for inertia support: Calculate the virtual inertia time constant according to the frequency deviation value and the rated power of the flywheel energy storage system, and calculate the power demand of the flywheel energy storage system participating in inertia support according to the virtual inertia time constant and the real-time frequency signal; at the same time, considering the power constraints of each array in the flywheel energy storage system, apply the frequency response characteristics aggregation of the array to determine the action depth of each flywheel array participating in inertia support.

[0068] S4. Control of flywheel array action: First, based on the action depth of each flywheel array, control the power output of each array in the flywheel energy storage system to meet the power required for inertia support. Then, according to the actual situation of the power output, adjust the charge and discharge states of each array so that the system can quickly respond to frequency changes during inertia support. At the same time, monitor the SOC state of the system in real time, and dynamically adjust the action depth of each array according to the monitoring results, so as to ensure that the system can operate continuously and stably during the entire inertia support process.

[0069] S5. Coordinated control during black-start process: During the black-start process, dynamically adjust the output power of the wind turbine generator set according to the change of the grid frequency to ensure that the wind turbine generator set can continuously and stably output power; at the same time, dynamically adjust the power output of the flywheel energy storage system according to the change of the grid frequency to ensure that the flywheel energy storage system can quickly respond to frequency changes.

[0070] The wind turbine generator set selects a doubly-fed wind turbine generator (DFIG) as the main power generation unit, whose rated power is determined according to the target grid capacity (e.g., 10MW / unit), and is equipped with a permanent magnet synchronous motor (PMSG) as a backup;

[0071] The wind turbine uses a staged startup strategy. First, the pre-charging unit ramps the converter DC bus voltage to 90% of its rated value at a rate of 0.5% / s. This process requires real-time monitoring of voltage fluctuations to ensure the tolerance does not exceed ±2%. Simultaneously, the pitch angle control system initiates a self-test routine, verifying the initial angle of each blade using a laser rangefinder, maintaining an error within ±0.5°.

[0072] The real-time power at the connection point between the output end of the wind turbine generator set and the grid is obtained through the voltage sensor and current sensor at the connection point.

[0073] The flywheel energy storage system adopts a modular design. Each flywheel array has a capacity of 1MWh, a speed range of 15,000-50,000rpm, and a response time of <100ms. The total capacity of the flywheel energy storage system must meet the inertia support requirements of at least 30 minutes during the initial black start of the power grid.

[0074] Initializing the flywheel energy storage system requires simultaneous mechanical system preparation and electrical system activation. Mechanically, the magnetic bearings initially levitate the rotor with 50% rated current. Once the vibration sensor detects an amplitude <10μm, the current is gradually increased to the operating level. Electrically, a three-stage SOC calibration is employed: first, open-circuit voltage calibration is performed, then internal resistance is measured using a pulsed current method, and finally, a temperature compensation algorithm is used to reduce the SOC estimation error to within 0.8%. The flywheel energy storage system's SOC is initialized to 80% (reserving 20% capacity for discharge response during frequency drops). The grid-connected frequency signal is collected via a synchronized phasor measurement unit (PMU) with a sampling frequency ≥1kHz.

[0075] During the black start power supply configuration and initial state setting process, a high-speed communication network (delay <1ms) is used to ensure that both enter the standby state synchronously and coordinate the preparation progress of the wind turbine generator set and flywheel energy storage system.

[0076] In S2, the fast Fourier transform (FFT) technology is used in the backup monitoring channel of the real-time frequency monitoring system to analyze the collected voltage signal. By decomposing the voltage signal into different frequency components, accurate frequency information is obtained to verify the accuracy of the PMU measurement results. To improve the monitoring reliability, distributed monitoring nodes based on the Internet of Things (IoT) are also introduced. These nodes are distributed throughout the power grid, collect local frequency information, and upload it to the central control system via wireless communication, and then perform fusion analysis with the PMU and FFT data.

[0077] In S2, the calculation of the frequency deviation value: The real-time frequency signal collected and processed by the PMU is subtracted from the rated frequency of the power grid to obtain the frequency deviation value; the dead zone constraint of the power grid is dynamically adjusted according to the operating state of the power grid, and the dead zone threshold is set to ±0.05 Hz; when abnormal conditions such as power grid faults or load mutations occur and the frequency change rate exceeds 1 Hz / s, the dead zone threshold is relaxed to ±0.1 Hz; by comparing the frequency deviation value with the dead zone threshold, if the frequency deviation value exceeds the dead zone threshold, the calculation process of the inertia support power demand is triggered (S3); if not, the frequency is continuously monitored, and the frequency deviation calculation result is updated every 50 ms to ensure real-time tracking of the power grid frequency change.

[0078] In S3, the calculation formula for the virtual inertia time constant:

[0079] T v = K p ·Δf + K i ·∫Δfdt;

[0080] Where, T v is the virtual inertia time constant; Δf is the frequency deviation value; K i is the integral coefficient, and its value range is 0.1 - 0.5, which is mainly used to suppress the steady-state error of the frequency deviation. Its specific value needs to be adjusted in combination with the system dynamic response characteristics and stability requirements; K p is the proportional coefficient, and its value range is between 0.5 - 2.0. It needs to be dynamically calibrated according to the flywheel rated power, the power grid short-circuit capacity, and the system inertia demand. When the flywheel rated power is large, the power grid short-circuit capacity is small, or the system inertia demand is high, K p should take a larger value;

[0081] ∫Δfdt is calculated using the numerical integration method. The time is discretized, and the time interval is Δt. The time interval Δt is consistent with the frequency deviation calculation update period to ensure the accuracy and real-time performance of the calculation. In the time period [t0, t n , the calculation formula is;

[0082]

[0083] Where, Δf(t i ) and Δf(t i+1 ) are the frequency deviation values at times t i and t i+1 respectively.

[0084] In S3, according to the calculated virtual inertia time constant T v and the real-time frequency signal, the formula for calculating the power demand of the flywheel energy storage system participating in inertia support is:

[0085] P req = K·Δf·T v ;

[0086] K = 0.8 - 0.2·tanh(5|Δf);

[0087] P req is the power demand for the flywheel energy storage system to participate in inertia support; K is the non - linear correction coefficient. Introducing it is to more accurately simulate the inertia support characteristics of the system under different frequency deviations. It can be seen from the formula that when the frequency deviation Δf is small, tanh(5|Δf) is close to 0, and at this time K is close to 0.8, and the power demand is approximately proportional to the frequency deviation and the virtual inertia. This enables the system to respond in a timely manner and provide appropriate power support under small frequency deviations. tanh represents the hyperbolic tangent function; when the frequency deviation Δf is large, tanh(5|Δf) is close to 1, and K is close to 0.6, realizing the saturation characteristic under large deviations, avoiding excessive power demand from impacting the system, and ensuring the stability of the system.

[0088] In S3, determine the action depth of each flywheel array participating in inertia support: When determining the action depth of each flywheel array participating in inertia support, in order to prevent a single array from over - exerting force, resulting in equipment damage or affecting the overall stability of the system, it is necessary to set the power upper limit for each array. The power upper limit is set to 0.2 times the rated power; The calculation formula for the action depth of each flywheel array participating in inertia support is:

[0089]

[0090] where, f n is the grid rated frequency; α is the action depth of the flywheel array. When α = 1, the flywheel array outputs full power;

[0091] When determining the action depth, comprehensively consider the current SOC state, remaining life of each flywheel array, and the overall stability of the system to optimize the distribution of the action depth; For a flywheel array with a lower SOC, if it outputs full power according to the calculated action depth, it will cause over - discharge, affecting its service life or even damaging the equipment. At this time, its action depth should be appropriately reduced; For an array with a shorter remaining life, it is also necessary to control its action depth to avoid over - use. Therefore, it is necessary to correct the calculated action depth:

[0092]

[0093] soc is the state of charge of the flywheel array; α′ is the corrected action depth; ω s is the soc weight of the wheel array; ω f is the weight of the remaining life; S h is the remaining life; S zis the total life. In this way, it can not only meet the power demand for system inertia support, but also optimize the use of each flywheel array to ensure the continuous and stable operation of the system.

[0094] In S4, the method for controlling the power output of each array in the flywheel energy storage system:

[0095] S411. Calculate the power output value of each array. According to the calculated action depth α′ of each corrected flywheel array and combined with the power upper limit of each array, through the formula: P ar = α′ × 0.2 × P ra calculate the actual power output value of each flywheel array; in the formula, P ar is the actual power output; P ra is the rated power of the flywheel energy storage system;

[0096] S412. Adopt hierarchical control to achieve precise tracking, and adopt a hierarchical control strategy to achieve precise control of the power output of the flywheel array;

[0097] S4121. Top-level model predictive control. At the top level, model predictive control (MPC) technology is used to predict the power demand and system state in the future period of time. With the minimum power tracking error and SOC balance as the optimization objectives, and adopting a rolling optimization strategy, according to the current system state (including frequency deviation, power output of each flywheel array, SOC value, etc.), calculate the power reference value of each flywheel array in the future period of time;

[0098] The power tracking error calculation formula is:

[0099]

[0100] E p represents the power tracking error; P ref (t) is the power reference value at time t; P ar (t) is the actual power output value at time t; t1 represents the starting point of time; t n represents the end point of time;

[0101] The SOC balance index is expressed as:

[0102]

[0103]

[0104] J S is the SOC balance index; m represents the number of flywheel arrays in the system; SOC a (t) represents the average SOC of all flywheel arrays at time t; SOC i (t) is the SOC of the i-th flywheel array at time t;

[0105] The rolling optimization strategy is as follows: The top-level MPC performs an optimization calculation every 50 ms (which can be adjusted according to the system response speed). During the prediction process, factors such as the charge and discharge dynamic characteristics of the flywheel array and the grid frequency response characteristics are considered to establish a dynamic power output model and an SOC change model of the flywheel array; through rolling optimization, the control input sequence that minimizes the weighted sum of the power tracking error E p and the SOC balance index J S is obtained. The first control input in the control input sequence with the minimum weighted sum is taken as the control decision at the current moment, and the power reference values of each flywheel array in the future period are calculated;

[0106] Among them, the dynamic power output model of the flywheel array is expressed as:

[0107] P ar (t + 1) = P ar (t) + ΔP ar (t)

[0108] ΔP ar (t) is the change in power output at time t; P ar (t + 1) is the actual power output value at time t + 1;

[0109] The SOC change model is expressed as:

[0110]

[0111] Δt is the time interval (50 ms); E c is the capacity of the flywheel array;

[0112] S4122. Bottom-layer pulse width modulation control. At the bottom layer, the power output of the flywheel array is controlled in real time through a pulse width modulation (PWM) converter. The bottom-layer pulse width modulation converter adjusts the switching frequency and duty cycle in real time according to the power reference value calculated by the top-level MPC. When the power reference value is greater than the current actual power output value, the duty cycle of the bottom-layer pulse width modulation converter is increased to increase the power output; if the power reference value is less than the current actual power output value, the duty cycle is decreased to reduce the power output, ensuring that the error between the actual power output and the reference value is less than 2%. At the same time, the power output, current, and voltage signals are monitored in real time. Once abnormal situations such as power mutation (such as the power change rate exceeds the set threshold, such as changing more than 10% of the rated power per second) and overcurrent (the current exceeds 120% of the rated current) occur, the control parameters of the bottom-layer pulse width modulation converter (such as reducing the switching frequency and decreasing the duty cycle) are immediately adjusted or the circuit is cut off to prevent equipment damage.

[0113] In S4, according to the actual situation of power output, the steps of adjusting the charge and discharge states of each array and real-time monitoring the SOC state of the system include:

[0114] Adjust the charge and discharge states of each array in the flywheel energy storage system to ensure that the flywheel energy storage system can quickly respond to frequency changes during the inertia support process; when the grid frequency drops, the flywheel array discharges to release energy to support the grid frequency; when the grid frequency rises, the flywheel array charges to absorb the excess energy.

[0115] Real-time monitor the SOC state of the flywheel energy storage system, and adjust the action depth of the flywheel array according to the SOC state. When the SOC is lower than 30%, trigger the power limit mode, reduce the output power of the flywheel array to 50%, and send a signal to the wind turbine generator to request an increase in the power generation power to maintain the grid power balance; when the SOC is higher than 90%, control the flywheel array to switch to the charging mode to absorb the excess grid power and prevent damage to the equipment caused by overcharging. To achieve the balance of the SOC of each flywheel array, when the SOC difference is greater than 15%, start the Nash bargaining solution negotiation algorithm based on game theory. Each flywheel array negotiates and adjusts the charge and discharge power according to its own SOC state and the overall system requirements until the SOC difference is reduced to a range less than 5%, ensuring that the flywheel energy storage system can operate continuously and stably during the inertia support process.

[0116] In S5, the power adjustment strategy of the wind turbine generator:

[0117] During the black start process, the grid frequency fluctuates greatly. It is necessary to dynamically adjust the output power of the wind turbine generator according to the frequency change. Adopt the frequency droop control strategy. When the frequency deviation is greater than 0.1Hz, the output power P of the wind turbine generator w The calculation formula is:

[0118] P w =P in +K dr ×Δf;

[0119] P in is the initial output power; K dr is the droop coefficient; Δf is the frequency deviation;

[0120] At the same time, to further optimize the power output and frequency response characteristics, adjust the pitch angle according to the frequency change. When the frequency deviation is greater than 0.1Hz, increase the pitch angle to reduce the wind energy captured by the wind turbine blades and reduce the output power;

[0121] The pitch angle adjustment formula is:

[0122]

[0123] θ pitch angle; θ0 is the initial pitch angle;

[0124] When the frequency deviation is less than -0.1 Hz, reduce the pitch angle, increase the wind energy capture and output power, and ensure that the wind turbine continuously and stably outputs power under different frequency conditions;

[0125] The pitch angle adjustment formula is:

[0126]

[0127] In S5, the method for dynamically adjusting the power output of the flywheel energy storage system according to the change of the grid frequency is as follows:

[0128] When the frequency deviation is greater than 0.2 Hz (high-frequency disturbance), the flywheel energy storage system, relying on its fast power regulation ability, quickly releases or absorbs energy to suppress the rapid fluctuation of the frequency. Each flywheel array performs charge and discharge operations according to the action depth and power output control strategy determined in S4. When the frequency deviation is less than 0.1 Hz (low-frequency fluctuation), the wind turbine mainly undertakes the power adjustment task, and the flywheel energy storage system assists in the regulation. The flywheel energy storage system and the wind turbine exchange key information such as frequency deviation, power output, and SOC status in real time through a high-speed communication network (time delay < 1 ms), and dynamically adjust their respective power output strategies according to the actual situation of the power grid to achieve efficient coordination and jointly maintain the stability of the grid frequency. For example, when the wind turbine cannot respond to low-frequency fluctuations in time due to wind speed changes or other reasons, the flywheel energy storage system can quickly supplement or absorb power to ensure the stability of the power grid.

[0129] The above is a preferred embodiment of the invention content, and it is not intended to limit the invention content. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the invention content shall be included within the protection scope of the invention content.

Claims

1. A coordinated control method for wind-storage combined black start power supply based on inertia response, characterized in that: Including the following steps: S1. Black start power supply configuration and initial state setting: Configure a wind turbine generator set and a flywheel energy storage system as black start power supplies, and complete their initialization work. At the same time, obtain the real-time power and frequency of the connection points between the output ends of the wind turbine generator set and the flywheel energy storage system and the power grid. S2. Real-time frequency monitoring and deviation calculation: Calculate the frequency deviation value by subtracting the rated frequency from the real-time frequency signal at the grid connection point of the flywheel energy storage system obtained, and determine whether the frequency deviation value exceeds the grid dead zone constraint. S3. Inertia support power demand calculation: Calculate the virtual inertia time constant according to the frequency deviation value and the rated power of the flywheel energy storage system. Calculate the power demand of the flywheel energy storage system participating in inertia support according to the virtual inertia time constant and the real-time frequency signal. At the same time, considering the power constraints of each array in the flywheel energy storage system, aggregate using the frequency response characteristics of the array to determine the action depth of each flywheel array participating in inertia support. S4. Flywheel array action control: First, control the power output of each array in the flywheel energy storage system based on the action depth of each flywheel array. Then, adjust the charge and discharge states of each array according to the actual situation of the power output. At the same time, monitor the SOC state of the system in real time and dynamically adjust the action depth of each array according to the monitoring results. S5. Coordinated control during black start: Dynamically adjust the output power of the wind turbine generator set according to the change of the grid frequency. At the same time, dynamically adjust the power output of the flywheel energy storage system according to the change of the grid frequency.

2. The coordinated control method for wind-storage combined black start power supply based on inertia response according to claim 1 is characterized in that: In S1, a doubly-fed induction generator (DFIG) is selected as the main power generation unit of the wind turbine generator set, and its rated power is determined according to the target grid capacity, and a permanent magnet synchronous motor (PMSG) is configured as a backup. The wind turbine generator set adopts a hierarchical starting strategy. First, the DC bus voltage of the converter is raised to 90% of the rated value at a slope of 0.5% / s through a pre-charge unit. During this process, the voltage fluctuation needs to be monitored in real time to ensure that it does not exceed the allowable deviation of ±2%. At the same time, the pitch angle control system starts a self-check program, verifies the initial angle of each blade through a laser rangefinder, and controls the error within the range of ±0.5°. The flywheel energy storage system adopts a modular design. The capacity of each flywheel array is 1 MWh, the speed range is 15,000 - 50,000 rpm, the response time < 100 ms, and the total capacity of the flywheel energy storage system needs to meet the inertia support demand of at least 30 minutes at the initial stage of the grid black start. Initialization of the flywheel energy storage system requires synchronously completing the preparation of the mechanical system and the activation of the electrical system. Mechanically, the magnetic levitation bearing first passes 50% of the rated current to make the rotor levitate. After the vibration sensor detects that the amplitude < 10 μm, it is gradually increased to the working current. Electrically, a three-stage SOC calibration is adopted: first, the open-circuit voltage is calibrated, then the internal resistance is measured by the pulse current method, and finally, the SOC estimation error is compressed to within 0.8% by combining the temperature compensation algorithm. The real-time power at the connection point between the output end of the wind turbine generator set and the power grid is obtained through the voltage sensor and current sensor at the connection point. The SOC of the flywheel energy storage system is initialized to 80%. The grid connection point frequency signal is collected by a synchronized phasor measurement unit (PMU), and the sampling frequency is ≥1 kHz. During the configuration of the black start power supply and the setting of the initial state, a high-speed communication network is used to ensure that both enter the standby state synchronously.

3. The coordinated control method of the inertia response-based wind-storage combined black-start power supply according to claim 1, characterized in that, In S2, in the standby monitoring channel of the real-time frequency monitoring system, fast Fourier transform (FFT) is used to analyze the collected voltage signal. By decomposing the voltage signal into different frequency components, accurate frequency information is obtained to verify the accuracy of the PMU measurement results. To improve the monitoring reliability, distributed monitoring nodes based on the Internet of Things (IoT) are introduced. These nodes are distributed throughout the power grid, collect local frequency information, and upload it to the central control system through wireless communication for fusion analysis with PMU and FFT data. Calculation of the frequency deviation value: The real-time frequency signal collected and processed by the PMU is subtracted from the rated grid frequency to obtain the frequency deviation value. The grid dead zone constraint is dynamically adjusted according to the grid operation state, and the dead zone threshold is set to ±0.05 Hz. When a grid fault or abnormal load mutation occurs and the frequency change rate exceeds 1 Hz / s, the dead zone threshold is relaxed to ±0.1 Hz. By comparing the frequency deviation value with the dead zone threshold, if the frequency deviation value exceeds the dead zone threshold, the subsequent process of calculating the inertial support power demand is triggered; if not, the frequency is continuously monitored, and the frequency deviation calculation result is updated every 50 ms to ensure real-time tracking of the grid frequency change.

4. The coordinated control method for wind-storage combined black start power supply based on inertia response according to claim 1 is characterized in that: In S3, Calculation formula for the virtual inertia time constant: T v = K p ·Δf + K i ·∫Δfdt; Among them, T v is the virtual inertia time constant; Δf is the frequency deviation value; K i is the integral coefficient, the value range is 0.1-0.5; K p is the proportionality coefficient; According to the calculated virtual inertia time constant T v and the real-time frequency signal, the power demand formula for the flywheel energy storage system to participate in inertia support is: P req = K·Δf·T v ; K = 0.8 - 0.2·tanh(5|Δf|); P req is the power requirement of the flywheel energy storage system participating in inertia support; K is the nonlinear correction coefficient; tanh represents the hyperbolic tangent function.

5. The coordinated control method of the inertia response-based wind-storage combined black-start power supply according to claim 1, characterized in that In S3, determining the action depth of each flywheel array participating in inertial support: When determining the action depth of each flywheel array participating in inertial support, in order to prevent a single array from overloading and causing equipment damage or affecting the overall system stability, the power upper limit of each array needs to be set, and the power upper limit is set to 0.2 times the rated power. The calculation formula for the action depth of each flywheel array participating in inertial support is: where f n is the rated frequency of the power grid; α is the operating depth of the flywheel array. When α = 1, the flywheel array outputs at full power; Δf is the frequency deviation value. When determining the action depth, the current SOC state, remaining life, and overall system stability of each flywheel array are comprehensively considered to optimize the distribution of the action depth. For flywheel arrays with a lower SOC, if full-power output according to the calculated action depth will cause over-discharge, affecting their service life or even damaging the equipment, the action depth should be appropriately reduced at this time; for arrays with a shorter remaining life, the action depth also needs to be controlled to avoid overuse, so the calculated action depth needs to be corrected: The SOC of the flywheel array; α′ is the corrected depth of action; ω s is the SOC weight of the wheel array; ω f is the weight of the remaining life; S h is the remaining life; S z is the total life.

6. The coordinated control method of the inertia response-based wind-storage combined black-start power supply according to claim 1, characterized in that In S4, the method for controlling the power output of each array in the flywheel energy storage system: S411. Calculate the power output values of each array. Based on the calculated action depth α′ of each flywheel array after correction, combined with the power upper limit of each array, through the formula: P ar = α′ × 0.2 × P ra Calculate the actual power output values of each flywheel array; in the formula, P ar is the actual power output; P ra is the rated power of the flywheel energy storage system; S412. Achieve precise tracking through hierarchical control, and adopt a hierarchical control strategy to achieve precise control of the power output of the flywheel array.

7. The coordinated control method of the inertia response-based wind-storage combined black-start power supply according to claim 6, characterized in that The method for achieving precise control of the power output of the flywheel array by adopting a hierarchical control strategy: S4121. Top-level model predictive control. The top-level model predictive control (MPC) technology is used to predict power demand and system status over a period of time. With minimizing power tracking error and balancing the SOC as optimization goals, a rolling optimization strategy is employed to calculate the power reference value for each flywheel array over a period of time based on the current system status. S4122. Low-level pulse width modulation control: The low-level pulse width modulation (PWM) converter performs real-time tracking and control of the flywheel array's power output. The low-level PWM converter adjusts the switching frequency and duty cycle in real time based on the power reference value calculated by the top-level MPC. When the power reference value is greater than the current actual power output value, the duty cycle of the low-level PWM converter is increased to improve power output. If the power reference value is less than the current actual power output value, the duty cycle is reduced and the power output is lowered to ensure that the error between the actual power output and the reference value is less than 2%. At the same time, the power output, current and voltage signals are monitored in real time. Once a power mutation or overcurrent abnormality occurs, the control parameters of the underlying pulse width modulation converter are immediately adjusted or the circuit is cut off to prevent equipment damage.

8. The coordinated control method of the inertia response-based wind-storage combined black-start power supply according to claim 7, characterized in that, The rolling optimization strategy is as follows: the top-level MPC performs an optimization calculation every 50ms. During the prediction process, the dynamic charging and discharging characteristics of the flywheel array and the grid frequency response characteristics are taken into account to establish a dynamic power output model and a state-of-charge (SOC) variation model for the flywheel array. Through rolling optimization, a control input sequence that minimizes the weighted sum of the power tracking error and the SOC balance index is solved. The first control input in the control input sequence with the minimum weighted sum is taken as the control decision at the current moment, and the power reference value of each flywheel array in the future period is calculated.

9. The coordinated control method for wind-storage combined black start power supply based on inertia response according to claim 1 is characterized in that: In S4, the steps of adjusting the charge and discharge state of each array according to the actual power output and monitoring the SOC state of the system in real time include: Adjust the charge and discharge status of each array in the flywheel energy storage system to ensure that the flywheel energy storage system can quickly respond to frequency changes during the inertia support process; when the grid frequency drops, the flywheel array discharges and releases energy to support the grid frequency; when the grid frequency rises, the flywheel array charges and absorbs excess energy; The SOC status of the flywheel energy storage system is monitored in real time, and the action depth of the flywheel array is adjusted according to the SOC status; when the SOC is lower than 30%, the power limit mode is triggered, the output power of the flywheel array is reduced to 50%, and a signal is sent to the wind turbine generator set to request an increase in power generation to maintain the power balance of the grid; when the SOC is higher than 90%, the flywheel array is controlled to switch to charging mode to absorb excess power from the grid and prevent overcharging from damaging the equipment. In order to achieve the balance of the SOC of each flywheel array, when the SOC difference is greater than 15%, the Nash bargaining solution negotiation algorithm based on game theory is started; each flywheel array negotiates to adjust the charging and discharging power according to its own SOC status and the overall system needs until the SOC difference is reduced to less than 5%, ensuring that the flywheel energy storage system can continue to operate stably during the inertia support process.

10. The coordinated control method of the inertia response-based wind-storage combined black-start power supply according to claim 1, wherein In S5, the wind turbine power adjustment strategy is: During the black start process, the grid frequency fluctuates greatly, and the output power of the wind turbine generator set needs to be dynamically adjusted according to the frequency change. The frequency droop control strategy is adopted. When the frequency deviation is greater than 0.1Hz, the output power P of the wind turbine generator set is adjusted. w The calculation formula is: P w =P in +K dr ×Δf; P in is the initial output power; K dr is the droop coefficient; Δf is the frequency deviation; At the same time, in order to further optimize the power output and frequency response characteristics, the pitch angle is adjusted according to the frequency change. When the frequency deviation is greater than 0.1Hz, the pitch angle is increased to reduce the wind energy captured by the wind turbine blades and reduce the output power. The pitch angle adjustment formula is: θ is the pitch angle; θ0 is the initial pitch angle; When the frequency deviation is less than -0.1Hz, the pitch angle is reduced to increase wind energy capture and output power, ensuring that the wind turbine generator set continuously and stably outputs power under different frequency conditions; The pitch angle adjustment formula is:

11. The coordinated control method of the inertia response-based wind-storage combined black-start power supply according to claim 1, wherein In S5, the method for dynamically adjusting the power output of the flywheel energy storage system according to the change of the grid frequency is: When the frequency deviation is greater than 0.2Hz, the flywheel energy storage system, with its fast power regulation capability, quickly releases or absorbs energy and suppresses the rapid fluctuation of frequency; each flywheel array performs charging and discharging operations according to the action depth and power output control strategy determined in S4; When the frequency deviation is less than 0.1Hz, the wind turbine generator set is mainly responsible for power adjustment, and the flywheel energy storage system assists in the adjustment; the flywheel energy storage system and the wind turbine generator set exchange key information such as frequency deviation, power output, and SOC status in real time through a high-speed communication network, and dynamically adjust their respective power output strategies according to the actual situation of the power grid, achieving efficient collaboration and jointly maintaining the stability of the power grid frequency.

Images