Flywheel energy storage system control method based on data center photovoltaic power supply

Abstract

The invention discloses a flywheel energy storage system control method based on data center photovoltaic power supply, and belongs to the technical field of power system and new energy grid-connected control, and the method comprises the steps: obtaining the output prediction data of photovoltaic power generation, the real-time load power, the voltage fluctuation parameter of a commercial power grid, and the real-time rotating speed state of flywheel energy storage; generating a power gap predicted value by taking the real-time rotating speed state of flywheel energy storage as a dynamic constraint variable; generating a flywheel commercial power cooperative control strategy according to the commercial power voltage fluctuation parameter; the temperature change rate of the flywheel motor is detected in real time, and a control instruction is generated; and synchronously sending the control instruction to the photovoltaic inverter and the flywheel motor, adjusting the parameters of the photovoltaic inverter and the flywheel rotating speed response data, and dynamically updating the control parameter set. According to the invention, a dual adjustment mechanism of flywheel rotating speed dynamic constraint and voltage fluctuation ratio weight distribution is adopted, and a temperature gradient harmonic collaborative suppression strategy is combined, so that millisecond-level power compensation response can be realized, the power grid dependence degree is reduced, and the service life of an energy storage system is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of power systems and new energy grid-connected control, and particularly relates to a control method for a flywheel energy storage system based on photovoltaic power supply for a data center. Background Art

[0002] As a high-energy-consuming scenario, the data center relies on traditional mains power supply, which has problems such as high energy cost and large carbon emissions.

[0003] The existing technology combines photovoltaic power generation with battery energy storage for power supply, but the response delay is significant when dealing with photovoltaic intermittency and load mutation. The battery has defects such as short cycle life and temperature sensitivity. The static weight distribution method of mains power and energy storage cannot adapt to grid voltage fluctuations and high-frequency harmonic interference, resulting in a decline in power quality and even equipment damage. In addition, the conventional prediction model does not fully consider the physical state constraints of the flywheel energy storage, and it is easy to cause the energy storage system to overload and disconnect from the grid when the rotational speed approaches the limit value. Summary of the Invention

[0004] To solve the above problems, the present invention provides a control method for a flywheel energy storage system based on photovoltaic power supply for a data center. By adopting a dual regulation mechanism of dynamic constraint of flywheel rotational speed and weight distribution of voltage volatility, combined with a temperature gradient harmonic collaborative suppression strategy, it can achieve millisecond-level power compensation response, reduce the dependence on the power grid, and extend the life of the energy storage system.

[0005] The above object can be achieved through the following solutions: A control method for a flywheel energy storage system based on photovoltaic power supply for a data center, including obtaining the predicted output data of photovoltaic power generation, the real-time load power of the data center, the mains grid voltage fluctuation parameters, and the real-time rotational speed state of the flywheel energy storage; using the real-time rotational speed state of the flywheel energy storage as a dynamic constraint variable to correct the time series matching relationship between the predicted photovoltaic output data and the load power, and generating a predicted power gap value with rotational speed constraints; extracting the voltage volatility according to the mains voltage fluctuation parameters, generating a flywheel weight according to the magnitude of the voltage volatility, and combining the predicted power gap value with rotational speed constraints to generate a flywheel-mains collaborative control strategy; detecting the temperature change rate of the flywheel motor in real time, and performing gradient constraint on the predicted power gap value of the collaborative control strategy according to the temperature change rate to generate a control instruction; synchronously sending the control instruction to the photovoltaic inverter and the flywheel motor to adjust the parameters of the photovoltaic inverter and the flywheel rotational speed response data; dynamically updating the control parameter set based on the adjusted parameters of the photovoltaic inverter and the flywheel rotational speed response data.

[0006] Optionally, the obtaining of the output power prediction data of the photovoltaic power generation and the real-time load power of the data center includes: the output power prediction data is preliminarily dynamically corrected through a cloud movement speed prediction model and the monitored value of the photovoltaic module temperature; the obtaining of the real-time load power includes actively synchronizing the server cluster scheduling plan and the temperature control instruction of the refrigeration system of the data center.

[0007] Optionally, taking the real-time rotational speed state of the flywheel energy storage as a dynamic constraint variable to correct the time series matching relationship between the photovoltaic output power prediction data and the load power, and generating a power gap prediction value with rotational speed constraint includes: taking the real-time rotational speed state of the flywheel energy storage as a dynamic constraint variable, constructing a rotational speed constraint coefficient and a trend constraint condition; aligning the preliminarily dynamically corrected photovoltaic output power prediction data and the load power in time series, and calculating to obtain a theoretical power gap value; using the rotational speed constraint coefficient, the trend constraint condition, and the theoretical power gap value to calculate and obtain the power gap prediction value.

[0008] Optionally, generating a flywheel weight according to the magnitude of the voltage volatility includes: when the voltage volatility is greater than or equal to a preset first threshold, increasing the flywheel weight to be greater than or equal to a preset first weight threshold; when the voltage volatility is between the first threshold and a preset second threshold, adjusting the flywheel weight linearly interpolated according to the duration of the fluctuation; when the voltage volatility is less than or equal to the second threshold, increasing the mains power weight to a preset second weight threshold; wherein, the first weight threshold is less than or equal to the second weight threshold.

[0009] Optionally, combining the power gap prediction value with rotational speed constraint to generate a flywheel-mains power collaborative control strategy includes: if the change rate of the power gap prediction value exceeds a preset change threshold, triggering a step adjustment mode of power distribution; if it is triggered for the first time, amplifying the flywheel weight according to a preset condition; if it is continuously triggered, decaying the flywheel weight according to an exponential curve.

[0010] Optionally, performing gradient constraint on the power gap prediction value of the collaborative control strategy according to the temperature change rate includes: preprocessing the temperature change rate; compensating the voltage amplitude of the photovoltaic inverter according to the magnitude of the temperature change rate; adjusting the upper limit of the trend constraint condition according to the magnitude of the temperature change rate; correcting the power gap prediction value according to the magnitude of the temperature change rate.

[0011] Optionally, the adjustment of photovoltaic inverter parameters and flywheel speed response data includes: real-time monitoring of each harmonic output by the photovoltaic inverter, and extracting a specific harmonic amplitude; when the specific harmonic amplitude exceeds a safety amplitude threshold, reducing the harmonic current response coefficient of the flywheel motor; calculating a harmonic severity index based on the specific harmonic amplitude, and adjusting the AC power weight in the flywheel-AC collaborative control strategy based on the harmonic severity index.

[0012] Optionally, the dynamically updated control parameter set includes: when the deviation between the actual response speed of the flywheel and the predicted response speed exceeds a preset response range, switching to an alternative residual compensation algorithm to correct the torque of the flywheel motor; performing stability rating according to the voltage fluctuation parameters of the mains power grid, and dynamically adjusting the update period of the control parameter set.

[0013] Optionally, the dynamically updated control parameter set also includes: generating a reverse feeding strategy for the mains power grid based on the relationship between the current remaining capacity of the flywheel energy storage and the output forecast data; and adjusting the upper limit of the flywheel weight based on the relationship between the current remaining capacity of the flywheel energy storage and a preset critical value.

[0014] Compared with the prior art, the present invention has the following advantages: 1. The present invention detects the flywheel speed in real time as a dynamic constraint variable, dynamically corrects the matching relationship between photovoltaic output and load, so that the power gap prediction value is always limited by the physical state of the flywheel, avoiding the control failure problem caused by the overcapacity of the energy storage system in the traditional method; combined with the coordinated control strategy generated by the voltage fluctuation rate, the millisecond-level power complementarity between the mains and the flywheel is realized, which significantly reduces the risk of voltage sag; 2. The present invention uses the temperature change rate to gradient constrain the power gap, actively reduces the output demand before the flywheel motor overheats, and solves the problem of increased mechanical wear caused by sudden temperature rise in traditional control; at the same time, the dynamic harmonic suppression mechanism reduces the loss of power electronic devices under high-frequency modulation, and prolongs the life of the photovoltaic inverter and the flywheel drive module; 3. The present invention adopts a residual compensation algorithm to dynamically correct the control parameter set, and performs dual adjustments on the flywheel response deviation and the grid stability rating, breaking through the limitations of traditional fixed parameter control in adapting to complex working conditions; through the dynamic association of the reverse feeding strategy and the weight upper limit, an intelligent balance of energy supply and demand is achieved.

[0015] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0017] Figure 1 It is a schematic flow chart of a control method for a flywheel energy storage system based on photovoltaic power supply in a data center according to an embodiment of the present invention.

[0018] Figure 2 It is a photovoltaic sudden drop curve graph according to an embodiment of the present invention.

[0019] Figure 3 It is a load mutation curve graph according to an embodiment of the present invention.

[0020] Figure 4 It is a harmonic suppression curve graph according to an embodiment of the present invention. Specific embodiments

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0022] Refer to Figure 1 , an embodiment of the present invention proposes a control method for a flywheel energy storage system based on photovoltaic power supply in a data center. By adopting a dual regulation mechanism of dynamic constraint of flywheel speed and weight distribution of voltage volatility, combined with a temperature gradient harmonic collaborative suppression strategy, it can achieve millisecond-level power compensation response, reduce the dependence on the power grid, and extend the service life of the energy storage system.

[0023] The method of this embodiment specifically includes: Obtain the output prediction data of photovoltaic power generation, the real-time load power of the data center, the voltage fluctuation parameters of the commercial power grid, and the real-time speed state of the flywheel energy storage; Specifically, the output prediction data within a future time interval is collected by the photovoltaic power generation prediction system, the total power consumption of the data center server cluster and the refrigeration system is monitored in real time, the voltage fluctuation range of the mains power grid and the rotational speed value of the flywheel energy storage device are measured synchronously. The output prediction data needs to be combined with the light intensity and temperature parameters in the meteorological data. The real-time load power includes the computing load of the server and the energy consumption of the refrigeration equipment. The mains voltage fluctuation parameter is calculated by the power monitoring module after sampling at the millisecond level to obtain the volatility, and the flywheel rotational speed state is collected by the rotational speed sensor. Among them, the output prediction data is the predicted value of the power generation capacity based on meteorological conditions and the performance of photovoltaic modules, the real-time load power is the current total power demand of the data center, the mains voltage fluctuation parameter is the change range of the grid voltage within a set time window, and the real-time rotational speed state of the flywheel energy storage is the instantaneous angular velocity of the energy storage flywheel rotor.

[0024] Taking the real-time rotational speed state of the flywheel energy storage as the dynamic constraint variable, correcting the time series matching relationship between the photovoltaic output prediction data and the load power, and generating a power gap prediction value with rotational speed constraint; The voltage volatility is extracted according to the mains voltage fluctuation parameter, and the flywheel weight is generated according to the magnitude of the voltage volatility. Combining with the power gap prediction value with rotational speed constraint, a flywheel-mains collaborative control strategy is generated; Specifically, in the flywheel-mains collaborative control strategy, for the power gap prediction value , there is: ; ; ; In the formula, is the flywheel weight, is the mains weight, is the flywheel power, is the mains power.

[0025] The temperature change rate of the flywheel motor is detected in real time, and the power gap prediction value of the collaborative control strategy is gradient-constrained according to the temperature change rate to generate a control command; The control command is synchronously sent to the photovoltaic inverter and the flywheel motor to adjust the parameters of the photovoltaic inverter and the flywheel speed response data; Specifically, when it is detected that the flywheel speed is insufficient, an instruction to increase the power factor is sent to the photovoltaic inverter to increase its output by 3% - 5% to compensate for the power gap; at the same time, a torque instruction is sent to the flywheel motor to control the flywheel speed increase rate within the threshold of 500 rpm per minute.

[0026] Based on the adjusted parameters of the photovoltaic inverter and the flywheel speed response data, the control parameter set is dynamically updated.

[0027] Specifically, according to the output characteristics modified based on the parameters of the PV inverter and the dynamic response results of the flywheel speed, the core parameter library of the control system is updated in real time. For the flywheel response speed deviation event, the residual compensation algorithm is switched to correct the flywheel torque; according to the stability level of the utility grid, the parameter update frequency is adaptively adjusted; at the same time, in combination with the matching degree between the remaining capacity of the flywheel and the predicted value of PV output, a reverse power feeding strategy and a weight upper limit constraint are dynamically generated.

[0028] This method constructs a collaborative control system for the PV-flywheel-grid through dynamic coupling of multi-source data. Its core lies in taking the flywheel speed as an immediate constraint condition and adaptively adjusting the power distribution strategy in combination with the grid state. Reducing the dependence on the utility grid, using the fast response characteristics of the flywheel to smooth PV fluctuations; introducing a temperature gradient constraint mechanism to prevent equipment overload risks; improving the power quality through harmonic dynamic suppression to ensure the stable operation of high-sensitivity loads in the data center. For example, in the composite scenario of sudden drop in PV output and unstable grid, it can automatically adjust the priority of flywheel output and restore power supply balance in a short time.

[0029] Optionally, the obtaining of the predicted output data of PV power generation and the real-time load power of the data center includes: The predicted output data is initially dynamically corrected through a cloud movement speed prediction model and the monitored values of PV module temperatures. Specifically, using the cloud trajectory data of meteorological satellites, inputting it into the cloud movement speed prediction model, calculating the time window of future shading of the PV panels, and at the same time combining the measured values of the backplane temperature sensors of the PV modules, based on the negative correlation between PV efficiency and temperature, reducing and correcting the original predicted output to obtain the predicted output data. . The correction formula is as follows: ; In the formula, is the original predicted output, is the PV temperature attenuation coefficient (usually taken as 0.004 - 0.005 / °C), is the difference between the current temperature and the standard test temperature, is the cloud shading attenuation factor, is the predicted shading duration ratio. Among them, the cloud movement speed prediction model is an algorithm for analyzing cloud movement vectors based on satellite cloud images, and the monitored values of PV module temperatures are the surface temperatures of the modules measured by infrared sensors or thermocouples.

[0030] The obtaining of the real-time load power includes actively synchronizing the server cluster scheduling plan and the temperature control instructions of the refrigeration system in the data center.

[0031] Specifically, the data center management system sends the batch task startup schedule of the server cluster and the CPU utilization planning to the energy storage control unit, synchronously obtains the compressor frequency command generated by the refrigeration system according to the room temperature feedback, and comprehensively calculates the future minute-level load change trend.

[0032] Among them, the server cluster scheduling plan is the calculation task batch execution schedule preset for the data center, and the refrigeration system temperature control command is the equipment power regulation signal generated based on the thermodynamic model.

[0033] Optionally, using the real-time rotational speed state of the flywheel energy storage as a dynamic constraint variable to correct the time series matching relationship between the predicted photovoltaic output data and the load power, and generating a power gap prediction value with rotational speed constraints includes: Using the real-time rotational speed state of the flywheel energy storage as a dynamic constraint variable to construct a rotational speed constraint coefficient and a trend constraint condition; Specifically, the rotational speed constraint coefficient is a real-time constraint, and the trend constraint condition is a trend constraint. A dual rotational speed constraint model is constructed using the real-time constraint and the trend constraint; among them, for the rotational speed constraint coefficient , there is: ; In the formula, is the current flywheel rotational speed, is the maximum allowable rotational speed, is the threshold coefficient; the trend constraint condition is to perform ARIMA time series modeling on the historical rotational speed data to predict the rotational speed change rate within a preset future time period. The rotational speed constraint coefficient is the core parameter for dynamically adjusting the power distribution ratio according to the remaining stored energy of the flywheel, the trend constraint condition is the rule limit for predicting the rotational speed change rate based on historical data, and the theoretical power gap value is the original difference between the load power and the predicted photovoltaic output value.

[0034] Align the preliminarily dynamically corrected predicted photovoltaic output data with the load power in time series, and calculate the theoretical power gap value; Specifically, align the preliminarily dynamically corrected predicted photovoltaic output data and the load power according to the minute-level time stamp; calculate the theoretical power gap value for each time point , there is: ; In the formula, is the load power. Time series alignment is a preprocessing method for integrating multi-source data based on a unified time reference, and the theoretical power gap value is the calculated value of the power difference without considering the physical limitations of the flywheel.

[0035] Using the rotational speed constraint coefficient, the trend constraint condition, and the theoretical power gap value, calculate the power gap prediction value.

[0036] Specifically, considering the instantaneous output capacity of the flywheel, the predicted value of the power gap is calculated using the rotational speed constraint coefficient, the trend constraint condition, and the theoretical power gap value. For the predicted value of the power gap , there is: ; In the formula, is a normalization function used to suppress the overshoot risk, is the trend constraint condition.

[0037] Exemplarily, in a certain data center, the predicted photovoltaic output suddenly drops short-term. The original prediction of 400 kW suddenly drops to 220 kW, and the load power increases to 350 kW due to an emergency task. The flywheel rotational speed = 14500 rpm, = 16000 rpm, = 0.85, and the ARIMA model predicts that in the next 3 minutes = 200 rpm / min, which is in an upward trend; the rotational speed constraint coefficient is calculated to obtain , and then the theoretical power gap value is calculated, j is calculated to be approximately 1. Therefore, the predicted value of the power gap . Through the dual restrictions of real-time rotational speed constraint and trend prediction, the flywheel output only needs to bear a deficit of 121 kW instead of the theoretical 130 kW, actively reducing the flywheel load pressure under the condition that the rotational speed is close to the upper limit, and slowing down mechanical wear. After the trend prediction module detects the continuous upward trend of the rotational speed, it triggers the sigmoid function to suppress excessive power distribution, avoiding the risk of flywheel overspeed and disconnection from the grid, and ensuring the safe response of the energy storage system. This method dynamically fuses the instantaneous physical state (rotational speed) of the flywheel with the future change trend, constraining the matching process between the photovoltaic output prediction and the load demand, and ensuring that the generated power gap value is always within the safe operation boundary of the flywheel system.

[0038] Optionally, generating the flywheel weight according to the magnitude of the voltage volatility includes: When the voltage volatility is greater than or equal to a preset first threshold, the flywheel weight is increased to be greater than or equal to a preset first weight threshold; Specifically, when the voltage volatility is greater than or equal to the first threshold, the flywheel weight is increased to a value not lower than the first weight threshold. For the new flywheel weight , there is: ; In the formula, is the first weight threshold, and the first weight threshold is a relatively large value, and the value range can be set between 0.7 - 0.9, is the amplification factor, which is used to amplify the current flywheel weight, and its value range can be set between 1.1 and 1.5. is the current flywheel weight, and the first weight threshold and the amplification factor constraint condition are set, that is, the product of the first weight threshold and the amplification factor is less than or equal to 0.9. is the voltage volatility. is the first threshold.

[0039] When the voltage volatility is between the first threshold and the preset second threshold, the flywheel weight is adjusted linearly according to the fluctuation duration. Specifically, when it is detected that the voltage volatility is between the first threshold and the second threshold, the timing starts, the fluctuation duration is calculated, and the flywheel weight is adjusted using the fluctuation duration. At this time, for the new flywheel weight , there is: ; In the formula, is the constant coefficient of the duration, is the second threshold. Among them, the linear interpolation adjustment strategy is a composite ratio adjustment method based on the degree of fluctuation and time.

[0040] When the voltage volatility is less than or equal to the second threshold, the mains power weight is increased to the preset second weight threshold. Among them, the first weight threshold is less than or equal to the second weight threshold.

[0041] Specifically, when the voltage volatility is less than or equal to the second threshold, the mains power dominates the weight value. At this time, the flywheel weight is greatly reduced. For the new flywheel weight , there is: ; In the formula, is the second weight threshold, is the second threshold. Here, the second weight threshold is greater than the first weight threshold to achieve the dominance of the mains power.

[0042] Optionally, generating the flywheel-mains coordinated control strategy by combining the power gap prediction value with the rotational speed constraint includes: If the change rate of the power gap prediction value exceeds the preset change threshold, then trigger the step adjustment mode of power distribution; If it is triggered for the first time, the flywheel weight is amplified according to the preset conditions; Specifically, if it is triggered for the first time, the flywheel needs to bear a large proportion of the power gap prediction value, and at the same time, the influence of the flywheel temperature needs to be considered. For the flywheel weight at this time , there is: ; In the formula, is the minimum value function, is the maximum value function, is the adjustment coefficient of the temperature constraint, is the safety temperature threshold of the flywheel, is the current temperature of the flywheel.

[0043] If it is continuously triggered, the flywheel weight is attenuated according to an exponential curve.

[0044] Specifically, an exponential decay distribution strategy is adopted to adjust the flywheel weight. For the flywheel weight , there is: ; In the formula, is the continuous trigger time.

[0045] Optionally, the gradient constraint on the power gap prediction value of the cooperative control strategy according to the temperature change rate includes: Preprocess the temperature change rate; Specifically, a sliding window check is adopted to calculate the temperature change rate 5 times continuously to eliminate abnormal mutations. For example, if it exceeds ±50°C / s instantaneously, the smoothed temperature change rate data is retained.

[0046] Compensate the voltage amplitude of the photovoltaic inverter according to the magnitude of the temperature change rate; Specifically, judge the magnitude of the temperature change rate. When the temperature change rate is greater than the preset first temperature threshold, the voltage amplitude of the photovoltaic inverter is compensated. For the voltage amplitude compensation value , there is: ; In the formula, is the compensation coefficient, is the temperature change rate. Through voltage amplitude compensation, the output prediction data of photovoltaic power generation is adjusted.

[0047] Adjust the upper limit of the trend constraint condition according to the magnitude of the temperature change rate; Specifically, adjust the upper limit of the trend constraint condition according to the magnitude of the temperature change rate and the preset second temperature threshold, where the second temperature threshold can be equal to the first temperature threshold. For the maximum value of the trend constraint condition, there is: ; In the formula, is the speed upper limit adjustment coefficient, is the second temperature threshold. By limiting the trend constraint condition, the power gap prediction value is limited.

[0048] Modify the predicted power gap value according to the magnitude of the temperature change rate.

[0049] Specifically, collect the duration during which the temperature change rate is greater than the first temperature threshold or the second temperature threshold. The duration is not calculated repeatedly. When it is greater than both the first temperature threshold and the second temperature threshold simultaneously, calculate the duration based on the longest duration. When the duration is greater than the preset load reduction time threshold, modify the current predicted power gap value. For the modified predicted power gap value , there is: .

[0050] Optionally, the adjustment of the photovoltaic inverter parameters and the flywheel speed response data includes: Real-time monitor the harmonics of the photovoltaic inverter output, and extract the specific harmonic amplitude; Specifically, collect the voltage and current waveforms at the output end of the inverter in real time, and extract the 3rd / 5th / 7th harmonic components through an improved fast Fourier transform (FFT) algorithm to obtain the specific harmonic amplitude.

[0051] When the specific harmonic amplitude exceeds the safety amplitude threshold, reduce the harmonic current response coefficient of the flywheel motor; Specifically, dynamically adjust the parameters of the current loop PI controller. For example, when it is detected that the 5th harmonic amplitude exceeds 110% of the IEC61000-3-2 standard limit, adjust the proportional gain coefficient in the PI controller. For the adjusted proportional gain coefficient , there is: ; In the formula, is the current proportional gain coefficient, is the th harmonic amplitude, is the th harmonic current limit, is the gain attenuation coefficient. Among them, the proportional gain coefficient is negatively correlated with the flywheel response delay; when the high-frequency harmonics are large, the proportional gain coefficient decreases, suppressing the fast response of the control loop and avoiding the oscillation risk of over-modulation of the flywheel motor; when the specific harmonic amplitude is less than the safety amplitude threshold, it is restored to a higher proportional gain coefficient to maintain the dynamic tracking ability of the control system.

[0052] Calculate the harmonic severity index according to the specific harmonic amplitude, and adjust the mains power weight in the flywheel-mains collaborative control strategy according to the harmonic severity index.

[0053] Specifically, for the harmonic severity index , there is: ; In the formula, is the amplitude of the th harmonic, is the current limit of the th harmonic. The weight of the mains power is corrected using the harmonic severity index. For the corrected mains power weight , there is: ; In the formula, is a constant coefficient, is the current weight of the mains power.

[0054] Optionally, the dynamically updated control parameter set includes: When the deviation between the actual response speed and the predicted response speed of the flywheel exceeds the preset response range, switch to a backup residual compensation algorithm to correct the torque of the flywheel motor; Specifically, by collecting the flywheel speed change rate in real time and the expected speed change rate calculated by the preset flywheel dynamics model, calculate the absolute value of the deviation of the speed change rate; determine whether the absolute value of the deviation is greater than the preset response range. If so, determine that the response is abnormal and switch to a backup residual compensation algorithm; the residual compensation algorithm calculates the mean value of the absolute value of the deviation by recording the historical data of the actual speed change rate in recent multiple cycles to obtain the residual mean value; use the residual mean value to correct the torque command. For the corrected torque command , there is: ; In the formula, is the current torque command, is the correction gain, is the residual mean value. In the first 10 ms after triggering, a linear superposition of the current torque command and the corrected torque command is adopted. For the superposition command , there is: ; In the formula, is the smoothing coefficient, which increases from 0 to 1 with a step of 0.1 / ms to achieve smooth transition.

[0055] Perform a stability rating based on the mains power grid voltage fluctuation parameters and dynamically adjust the update period of the control parameter set.

[0056] Specifically, extract the voltage volatility rate, frequency deviation degree, and harmonic distortion rate from the mains power grid voltage fluctuation parameters, calculate the stability evaluation value. For the stability evaluation value , there is: ; In the formula, is the normalized value of the voltage volatility rate, is the normalized value of the frequency offset degree, is the harmonic distortion rate, 、 、 are weight coefficients. Divide the levels of the stability rating, such as high stability, medium stability, and low stability. According to the levels, divide the range of the stability evaluation value. For example, 0 - 0.4 is high stability, 0.4 - 0.7 is medium stability, and 0.7 - 1 is low stability. Set an update period for each interval. Update the corresponding update period according to the interval where the stability evaluation value is located.

[0057] Optionally, the dynamic update control parameter set further includes: Generate a reverse power feeding strategy for the mains power grid according to the relationship between the remaining capacity of the current flywheel energy storage and the output prediction data; Specifically, define the relationship threshold between the predicted photovoltaic output and the remaining capacity of the flywheel : ; In the formula, is the remaining capacity of the current flywheel energy storage, is the maximum capacity of the flywheel energy storage, is the rated value of the output prediction data. When the relationship threshold is greater than the preset standard value, power feeding to the grid is allowed, and the priority is set according to the time-of-use electricity price strategy. When the relationship threshold x is less than or equal to the preset standard value, power feeding is prohibited, and local loads are preferentially guaranteed.

[0058] Adjust the upper limit of the flywheel weight according to the magnitude relationship between the remaining capacity of the current flywheel energy storage and the preset critical value.

[0059] Specifically, the critical value can be set to 20% of the maximum capacity of the flywheel energy storage. Therefore, for the upper limit of the flywheel weight , there is: ; In the formula, is the preset critical value.

[0060] To verify the feasibility of the present invention in implementation, the present invention is applied to a large-scale cloud computing data center. This data center is located in the northwest region with rich sunlight resources, and a megawatt-level distributed photovoltaic power generation system is deployed, and a flywheel energy storage array is equipped to suppress photovoltaic fluctuations and load mutations. The data center load includes a server cluster, a liquid cooling system, and a backup power supply, and the peak total power demand is 800 kW, and power supply continuity and voltage stability need to be guaranteed. Since the photovoltaic output fluctuates frequently due to cloud shading, the traditional mode of mains power plus energy storage cannot achieve millisecond-level power compensation, resulting in a significant risk of voltage sag for critical loads.

[0061] In this embodiment, the data center adopts the photovoltaic flywheel collaborative control method proposed by the present invention, and real-time collects photovoltaic output prediction data (based on cloud trajectory model and component temperature correction), server load scheduling plan, flywheel speed, and mains voltage fluctuation parameters. The system corrects the predicted value of the power gap through a dynamic speed constraint model, generates a flywheel weight allocation strategy in combination with the voltage volatility rate, and simultaneously introduces a temperature gradient constraint and a harmonic suppression mechanism to optimize the control instruction.

[0062] To verify the effectiveness, the period from March to September 2024 is selected as the test cycle, and the performance of the traditional energy storage control (lithium iron phosphate battery plus static weight allocation) and the method of the present invention is compared. The experimental group and the control group are both connected to the same photovoltaic array and load, the data acquisition frequency is millisecond-level, and the analysis scenarios include three typical events: sudden drop in photovoltaic power, sudden change in load, and mains voltage fluctuation.

[0063] Test scenario 1: Short-term sudden drop in photovoltaic output At 11:20 on July 15th, as Figure 2 shown, the cloud cover causes the photovoltaic output to drop from 650 kW to 320 kW within 30 seconds, and the load stabilizes at 720 kW. The system of the present invention detects the speed state in real time (currently 12,500 rpm, maximum 16,000 rpm), calculates the speed constraint coefficient to be 0.88, and generates a predicted value of the power gap of 148 kW (theoretical value 170 kW). Combining the mains voltage volatility rate (3.2%, exceeding the threshold of 2.5%), the flywheel weight is increased to 0.75, the instantaneous output of the flywheel is 111 kW, and the mains supply supplements 37 kW, with a response time of 1.2 seconds. Due to the battery response delay (4.5 seconds) in the control group, there is a voltage sag of 280 ms (dropping to 198 V), while the voltage fluctuation range of the experimental group is always controlled within 205 V ± 2 V.

[0064] Test scenario 2: Sudden burst of load in the server cluster At 14:05 on August 3rd, as Figure 3 shown, the start of a batch calculation task causes the load to jump from 600 kW to 780 kW, and the photovoltaic output remains at 480 kW. The system anticipates the load change by synchronizing the server scheduling plan and triggers a step adjustment mode: when first triggered, the flywheel weight is amplified to 0.85 and decays exponentially to 0.6 within 10 seconds. The flywheel speed rises from 14,000 rpm to 15,200 rpm, and the gap compensation amount reaches 238 kW. The mains supply supplements the remaining 42 kW after a delay of 5 seconds. The voltage recovery time of the experimental group is 0.8 seconds, while in the control group, due to the static weight allocation (flywheel:mains = 0.5:0.5), the compensation is lagged, and the voltage recovery takes 3.6 seconds.

[0065] Test scenario 3: High harmonic interference in the mains At 10:45 on September 12th, asFigure 4 As shown, the 5th harmonic is introduced on the grid side (amplitude exceeding the limit by 15%), the system detects the harmonic severity index H = 1.32 in real time, reduces the flywheel harmonic response coefficient (PI gain decays by 32%), and at the same time increases the mains weight to 0.65. The THD of the output voltage of the experimental group drops from 7.8% to 3.1%, while the THD of the control group continues to be as high as 6.9% because the harmonic suppression strategy is not dynamically adjusted, resulting in some sensitive servers triggering protection shutdowns.

[0066] Experiments show that through dynamic speed constraint and flywheel-mains collaborative weight distribution, the present invention improves the power gap compensation response speed to 1.2 seconds, which is 275% faster than the traditional method. The temperature gradient constraint mechanism enables the flywheel to operate in a safe speed range, reducing the average annual overspeed events by 92%, and the harmonic adaptive suppression reduces the THD by 56%. In terms of economy, the annual comprehensive cost savings reach 378,000 yuan, verifying the significant advantages of this method in improving power supply quality, extending equipment life, and reducing operating costs.

[0067] It should be noted that the electrical connections between the above-mentioned various units do not necessarily mean direct line connections. Indirect connection methods, as long as they achieve the purpose of the present invention, are applicable to the embodiments of the present invention. The above are only exemplary embodiments of the present invention and cannot be used to limit the scope of the present invention.

[0068] That is, any equivalent changes and modifications made in accordance with the teachings of the present invention still fall within the scope covered by the present invention. After considering the specification and the disclosure of the practical truth, those skilled in the art will easily think of other implementation schemes of the present invention. This application aims to cover any variations, uses, or adaptive changes of the present invention, which follow the general principles of the present invention and include common general knowledge or conventional technical means in the technical field not recorded in the present invention.

Claims

1. A flywheel energy storage system control method based on photovoltaic power supply in a data center, characterized in that: The method comprises: Obtain output forecast data of photovoltaic power generation, real-time load power of data centers, voltage fluctuation parameters of the mains grid, and real-time speed status of flywheel energy storage; Taking the real-time speed state of flywheel energy storage as the dynamic constraint variable, the time series matching relationship between photovoltaic output forecast data and load power is corrected to generate a power gap forecast value with speed constraint. The voltage fluctuation rate is obtained according to the mains voltage fluctuation parameter, and the flywheel weight is generated according to the magnitude of the voltage fluctuation rate, and the flywheel-mains coordinated control strategy is generated in combination with the power gap prediction value constrained by the speed; Detecting the temperature change rate of the flywheel motor in real time, and performing gradient constraints on the power gap prediction value of the collaborative control strategy according to the temperature change rate to generate a control instruction; The control command is synchronously sent to the photovoltaic inverter and the flywheel motor to adjust the photovoltaic inverter parameters and the flywheel speed response data; The control parameter set is dynamically updated based on the adjusted PV inverter parameters and flywheel speed response data.

2. A flywheel energy storage system control method based on photovoltaic power supply for data center according to claim 1, characterized in that: The acquisition of photovoltaic power generation output forecast data and real-time load power of the data center includes: The output prediction data is initially and dynamically corrected by using a cloud movement speed prediction model and photovoltaic module temperature monitoring values; The acquisition of the real-time load power includes actively synchronizing the server cluster scheduling plan of the data center and the temperature control instructions of the refrigeration system.

3. A flywheel energy storage system control method based on photovoltaic power supply for data center according to claim 2, characterized in that: The method of using the real-time speed state of the flywheel energy storage as a dynamic constraint variable, correcting the time series matching relationship between the photovoltaic output prediction data and the load power, and generating a power gap prediction value with a speed constraint includes: The real-time speed state of the flywheel energy storage is used as the dynamic constraint variable to construct the speed constraint coefficient and trend constraint condition; Align the initial dynamically corrected photovoltaic output forecast data with the load power in time series, and calculate the theoretical power gap value; The power gap prediction value is calculated using the speed constraint coefficient, the trend constraint condition, and the theoretical power gap value.

4. A flywheel energy storage system control method based on photovoltaic power supply for data center according to claim 1, characterized in that: Generating the flywheel weight according to the magnitude of the voltage fluctuation rate comprises: When the voltage fluctuation rate is greater than or equal to a preset first threshold, the flywheel weight is increased to be greater than or equal to the preset first weight threshold; When the voltage fluctuation rate is between the first threshold and a preset second threshold, adjusting the flywheel weight by linear interpolation according to the duration of the fluctuation; When the voltage fluctuation rate is less than or equal to the second threshold, the mains power weight is increased to a preset second weight threshold; The first weight threshold is less than or equal to the second weight threshold.

5. A flywheel energy storage system control method based on photovoltaic power supply for data center according to claim 3, characterized in that: The generating of the flywheel-mains coordinated control strategy by combining the power gap prediction value with the speed constraint includes: If the change rate of the power gap prediction value exceeds a preset change threshold, a step adjustment mode of power allocation is triggered; If it is the first trigger, the flywheel weight will be amplified according to the preset conditions; If it is continuously triggered, the flywheel weight will be attenuated according to the exponential curve.

6. A flywheel energy storage system control method based on photovoltaic power supply for data center according to claim 3, characterized in that: The step of performing a gradient constraint on the power gap prediction value of the collaborative control strategy according to the temperature change rate includes: Preprocessing the temperature change rate; According to the magnitude of the temperature change rate, the voltage amplitude of the photovoltaic inverter is compensated; According to the temperature change rate, adjusting the upper limit of the trend constraint condition; The power gap prediction value is corrected according to the temperature change rate.

7. A flywheel energy storage system control method based on photovoltaic power supply for data center according to claim 1, characterized in that: The adjustment of photovoltaic inverter parameters and flywheel speed response data includes: Monitor the harmonics output by the photovoltaic inverter in real time and extract the amplitude of specific harmonics; When the amplitude of the specific harmonic exceeds the safety amplitude threshold, reducing the harmonic current response coefficient of the flywheel motor; A harmonic severity index is calculated according to the specific harmonic amplitude, and a mains power weight in the flywheel-mains power coordinated control strategy is adjusted according to the harmonic severity index.

8. A flywheel energy storage system control method based on photovoltaic power supply for data center according to claim 1, characterized in that: The dynamic update control parameter set includes: When the deviation between the actual response speed of the flywheel and the predicted response speed exceeds a preset response range, switching to a backup residual compensation algorithm is used to correct the torque of the flywheel motor; Stability rating is performed according to the voltage fluctuation parameters of the mains power grid, and the update period of the control parameter set is dynamically adjusted.

9. A flywheel energy storage system control method based on photovoltaic power supply for data center according to claim 1, characterized in that: The dynamic update control parameter set also includes: Generating a reverse feeding strategy for the mains power grid according to the relationship between the current remaining capacity of the flywheel energy storage and the output prediction data; The upper limit of the flywheel weight is adjusted according to the relationship between the current remaining capacity of the flywheel energy storage and the preset critical value.

Images