Electric energy quality monitoring device of large-scale centralized energy storage station

Abstract

The invention discloses an electric energy quality monitoring device of a large-scale centralized energy storage station, and particularly relates to the technical field of electric energy quality monitoring, and the method comprises the steps: extracting a characteristic frequency component through extracting a time domain distribution characteristic of a direct-current low-frequency ring current and combining with grid-connected point voltage fundamental frequency correction; evaluating circulation interaction strength based on a coupling coefficient of the characteristic frequency and the inherent frequency of the power grid, and triggering a subsynchronous oscillation amplitude-frequency parameter extraction process; matching oscillation parameters and damping characteristics through a pre-stored power grid frequency response curve, determining a potential resonance risk level, and generating a dynamic power limit value adjustment strategy and a battery cluster switching instruction; full-link closed-loop control from feature extraction and risk quantification to active suppression is realized, and the capabilities of suppressing subsynchronous oscillation and maintaining power grid frequency stability of the energy storage system under complex working conditions are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power quality monitoring, and more specifically, to a power quality monitoring device for a large-scale centralized energy storage station. Background Art

[0002] A large-scale centralized energy storage station usually consists of thousands of battery clusters connected in parallel to achieve large-capacity energy throughput and flexible scheduling; current power quality monitoring technologies mainly focus on the voltage, frequency, and conventional harmonic parameters on the AC grid connection side, and obtain steady-state data through a centralized acquisition terminal. However, with the large-scale deployment of energy storage systems, the problem of dynamic circulating current caused by the discreteness of monomer parameters and the mismatch of DC-side impedance between battery clusters has become increasingly prominent.

[0003] Due to the single monitoring dimension of the prior art, it is unable to effectively identify the subsynchronous oscillation phenomenon caused by the coupling of DC-side low-frequency circulating current to the AC side during the parallel operation of multiple battery clusters. This blind spot leads to the voltage fluctuations and frequency offsets at the grid connection point of the energy storage station being often misjudged as conventional load disturbances, masking the potential resonance risk between the dynamic circulating current inside the battery system and the inherent frequency of the power grid, and seriously restricting the safe operation boundary of the energy storage system and the regulation accuracy of the power grid stability. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a power quality monitoring device for a large-scale centralized energy storage station to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] The power quality monitoring device for a large-scale centralized energy storage station includes a DC detection module, an AC acquisition module, a fusion analysis module, and a regulation generation module. The device executes the following steps:

[0007] Extract the time-domain distribution characteristics of the DC low-frequency circulating current through the DC detection module;

[0008] Input the time-domain distribution characteristics into the fusion analysis module, and combine with the fundamental wave frequency of the grid connection point obtained in real time by the AC acquisition module to extract the characteristic frequency components of the DC low-frequency circulating current;

[0009] Calculate the coupling coefficient between the characteristic frequency components and the inherent frequency of the power grid. The coupling coefficient is used to characterize the interaction intensity between the DC low-frequency circulating current and the power grid frequency;

[0010] When the coupling coefficient exceeds the preset coupling threshold, extract the amplitude-frequency parameters of the subsynchronous oscillation;

[0011] Determine the potential resonance risk level according to the amplitude-frequency parameters of the subsynchronous oscillation;

[0012] The generation module adjusts the power output limit of the energy storage converter according to the potential resonance risk level and generates a battery cluster switching instruction to suppress the DC circulating current.

[0013] In a preferred embodiment, the time-domain distribution characteristics are extracted based on the root-mean-square sequence of the current fluctuation amplitude by synchronously collecting the instantaneous DC voltage and instantaneous DC current values of each battery cluster.

[0014] In a preferred embodiment, the amplitude-frequency parameters of the subsynchronous oscillation are obtained by generating a time-frequency energy distribution map through short-time Fourier transform based on the voltage fluctuation sequence and frequency offset sequence acquired by the AC acquisition module.

[0015] In a preferred embodiment, the potential resonance risk level is mapped by calculating the ratio of the grid damping ratio to the subsynchronous oscillation damping ratio at the intersection point of the amplitude and center frequency parameters of the subsynchronous oscillation on the pre-stored grid frequency response curve.

[0016] In a preferred embodiment, the acquisition of the time-domain distribution characteristics specifically includes:

[0017] Synchronously collect the instantaneous DC voltage and instantaneous DC current values of each battery cluster, and calculate the current fluctuation amplitude of each battery cluster;

[0018] Based on a preset time window, perform sliding interception on the current fluctuation amplitude to generate a root-mean-square sequence of the current fluctuation amplitude;

[0019] According to the difference between the peak distribution and valley distribution of the root-mean-square sequence, extract the time-domain distribution characteristics representing the DC low-frequency circulating current intensity.

[0020] In a preferred embodiment, input the time-domain distribution characteristics into the fusion analysis module, and combine with the fundamental wave frequency of the grid connection point voltage obtained by the AC acquisition module in real time to extract the characteristic frequency components of the DC low-frequency circulating current, including:

[0021] The fundamental wave frequency of the grid connection point voltage obtained by the AC acquisition module in real time;

[0022] Align the phase of the time-domain distribution characteristics with the fundamental wave frequency of the grid connection point voltage to eliminate the interference component of the fundamental wave frequency on the time-domain distribution characteristics;

[0023] Perform Fourier transform on the time-domain distribution characteristics after eliminating interference to extract the candidate frequency components of the DC low-frequency circulating current;

[0024] Based on the pre-stored power frequency integer multiple frequency range, screen the candidate frequency components, eliminate the candidate components belonging to the power frequency integer multiple frequency, and retain the remaining candidate components as the characteristic frequency components.

[0025] In a preferred embodiment, the coupling coefficient between the characteristic frequency component and the inherent frequency of the power grid is calculated. The coupling coefficient is used to characterize the interaction intensity between the DC low-frequency circulating current and the power grid frequency, including:

[0026] Calculate the absolute value of the frequency difference between the characteristic frequency component and the inherent frequency of the power grid;

[0027] Based on a preset frequency deviation threshold, normalize the absolute value of the frequency difference to generate a normalized frequency difference coefficient;

[0028] Generate a coupling coefficient according to the product of the normalized frequency difference coefficient and the pre-stored circulating current intensity weight value; the larger the coupling coefficient, the stronger the strong interaction between the DC low-frequency circulating current and the power grid frequency.

[0029] In a preferred embodiment, when the coupling coefficient exceeds the preset coupling threshold, the extraction of the amplitude-frequency parameters of the subsynchronous oscillation specifically includes:

[0030] Based on the AC acquisition module, continuously collect the voltage fluctuation sequence and frequency offset sequence for multiple power frequency cycles;

[0031] Perform time-axis synchronization alignment on the voltage fluctuation sequence and frequency offset sequence to generate a synchronized time-domain signal;

[0032] Perform a short-time Fourier transform on the synchronized time-domain signal to generate a time-frequency energy distribution map containing the three-dimensional distribution of time, frequency, and energy;

[0033] Extract the frequency components with amplitudes exceeding the preset energy threshold from the time-frequency energy distribution map as candidate oscillation components;

[0034] Based on the pre-stored subsynchronous oscillation frequency band range, screen the candidate oscillation components, and determine the components within the frequency band range and with a duration exceeding the preset duration threshold as the amplitude-frequency parameters of the subsynchronous oscillation.

[0035] In a preferred embodiment, the acquisition of the potential resonance risk level specifically includes:

[0036] According to the amplitude parameter and center frequency parameter in the amplitude-frequency parameters of the subsynchronous oscillation, locate the amplitude-frequency intersection position on the pre-stored power grid frequency response curve;

[0037] Calculate the power grid damping ratio and the subsynchronous oscillation damping ratio corresponding to the intersection position;

[0038] Divide the power grid damping ratio by the subsynchronous oscillation damping ratio to generate a damping ratio;

[0039] Based on the pre-stored risk level mapping table, match the damping ratio to the corresponding potential resonance risk level;

[0040] When the damping ratio ratio is within the preset high - risk threshold range, mark the potential resonance risk level as the high - risk level;

[0041] When the damping ratio ratio is within the preset medium - risk threshold range, mark the potential resonance risk level as the medium - risk level.

[0042] In a preferred embodiment, the regulation and generation module adjusts the power output limit of the energy storage converter according to the potential resonance risk level and generates a battery cluster switching instruction for suppressing the DC circulating current, including:

[0043] When the potential resonance risk level is the high - risk level, adjust the power output limit of the energy storage converter to the first preset proportion range;

[0044] When the potential resonance risk level is the medium - risk level, adjust the power output limit to the second preset proportion range;

[0045] Generate a battery cluster switching instruction for suppressing the DC circulating current based on the real - time obtained DC circulating current amplitude data, and preferentially switch the battery clusters whose circulating current amplitudes are ranked within the preset quantity;

[0046] After the switching instruction is executed, synchronously adjust the power output limits of adjacent battery clusters to the third preset proportion range.

[0047] Compared with the prior art, the present invention has the following beneficial effects:

[0048] 1. Through the cross - domain fusion analysis of the DC - side and AC - side parameters, the accurate capture of the coupling mechanism of the dynamic circulating current between battery clusters and the grid's subsynchronous oscillation is realized. By extracting the time - domain distribution characteristics of the DC low - frequency circulating current and combining with the dynamic correction of the fundamental wave frequency of the grid connection point voltage, the problem of misjudgment of subsynchronous oscillation caused by single - dimension of traditional monitoring technology is effectively solved. Based on the coupling coefficient between the characteristic frequency component and the grid's inherent frequency, the circulating current interaction intensity is dynamically quantified, and combined with the matching analysis of the amplitude - frequency parameters of subsynchronous oscillation and the pre - stored grid frequency response curve, the potential resonance risk source can be located in real time, significantly improving the accuracy and timeliness of the evaluation of the interaction stability between the energy storage system and the grid under complex working conditions.

[0049] 2. By means of a closed-loop control mechanism, the potential resonance risk level is dynamically associated with the power regulation strategy of the energy storage converter, breaking through the limitations of traditional passive monitoring and lagging regulation. According to the quantitative relationship between the amplitude-frequency characteristics of subsynchronous oscillation and the grid damping ratio, hierarchical regulation instructions are generated. This not only realizes the rapid suppression of power output under high-risk working conditions, but also blocks the diffusion path of circulating current energy through the priority sorting of the battery cluster switching strategy and the coordinated adjustment of the power of adjacent clusters. This active defense system based on multi-source parameter fusion can not only suppress the instantaneous deterioration of DC circulating current, but also optimize the pre-stored threshold and regulation logic through a dynamic learning mechanism, and finally form a double guarantee for the safe operation boundary of the energy storage system and the grid frequency stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 It is a schematic structural diagram of the power quality monitoring device of the large-scale centralized energy storage station of the present invention;

[0051] Figure 2 It is a flowchart of the steps executed by the power quality monitoring device of the large-scale centralized energy storage station of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0053] Embodiment: Figure 1 A schematic structural diagram of the power quality monitoring device of the large-scale centralized energy storage station of the present invention is given. The power quality monitoring device of the large-scale centralized energy storage station includes a DC detection module, an AC acquisition module, a fusion analysis module, and a regulation generation module.

[0054] Figure 2 A flowchart of the steps executed by the power quality monitoring device of the large-scale centralized energy storage station of the present invention is given. The device executes the following steps:

[0055] Extract the time-domain distribution characteristics of the DC low-frequency circulating current through the DC detection module;

[0056] Input the time-domain distribution characteristics into the fusion analysis module, and combine the fundamental wave frequency of the grid-connected point voltage obtained in real time by the AC acquisition module to extract the characteristic frequency components of the DC low-frequency circulating current;

[0057] Calculate the coupling coefficient between the characteristic frequency components and the grid natural frequency. The coupling coefficient is used to characterize the interaction intensity between the DC low-frequency circulating current and the grid frequency;

[0058] When the coupling coefficient exceeds a preset coupling threshold, extract the amplitude-frequency parameters of the subsynchronous oscillation;

[0059] Determine the potential resonance risk level according to the amplitude-frequency parameters of the subsynchronous oscillation;

[0060] Through the regulation generation module, adjust the power output limit of the energy storage converter according to the potential resonance risk level, and generate a switching instruction for the battery cluster to suppress the DC circulating current.

[0061] The acquisition of the time-domain distribution characteristics specifically includes:

[0062] First, by synchronously collecting the instantaneous DC voltage and instantaneous DC current of each battery cluster, record the voltage and current data of each battery cluster in real time. Among them, the time alignment accuracy of the synchronous acquisition is controlled within 1 microsecond to ensure that the instantaneous values of different battery clusters are aligned under the same time reference; Subsequently, for the instantaneous DC current value of each battery cluster, calculate its current fluctuation amplitude, and the current fluctuation amplitude is the absolute difference between the current instantaneous value at the current moment and the current instantaneous value at the previous moment. When the difference exceeds the preset fluctuation threshold, it is marked as an effective fluctuation amplitude; Then, perform a sliding intercept on the effective fluctuation amplitude based on a preset time window. The length of the preset time window is 1 second, and the sliding step is 0.1 second. Calculate the root mean square value of the effective fluctuation amplitude within each window to generate a root mean square sequence of the current fluctuation amplitude; Finally, according to the difference between the peak distribution and the valley distribution of the root mean square sequence, extract the time-domain distribution characteristics representing the intensity of the DC low-frequency circulating current.

[0063] Among them, the peak distribution is the set of maximum values in the root mean square sequence, the valley distribution is the set of minimum values in the root mean square sequence, and the difference is the absolute difference between the peak distribution mean and the valley distribution mean.

[0064] To ensure data integrity, the setting method of the preset fluctuation threshold includes: taking 10% of the maximum fluctuation range of the instantaneous DC current value in the historical operation data as the threshold; During the generation process of the root mean square sequence, if the number of effective fluctuation amplitudes within the window is less than the preset number, extend the window length until the number requirement is met. The specific application of the time-domain distribution characteristics includes: when the difference exceeds the preset circulating current intensity threshold, it is determined that there is significant DC low-frequency circulating current, and the subsequent extraction process of the characteristic frequency components is triggered.

[0065] Input the time-domain distribution characteristics into the fusion analysis module, and combine the fundamental wave frequency of the grid connection point voltage obtained by the AC acquisition module in real time to extract the characteristic frequency components of the DC low-frequency circulating current, including:

[0066] The grid connection point voltage signal is collected in real time through a voltage sensor, and the fundamental frequency is extracted from the voltage signal using the phase-locked loop technology to ensure the real-time performance and phase accuracy of the fundamental frequency; the time series of the time-domain distribution characteristics is phase-aligned with the periodic signal corresponding to the fundamental frequency. Specifically, the time axis of the time-domain distribution characteristics is segmented according to the period length of the fundamental frequency, and the starting point of each segment is aligned to the zero-crossing point of the fundamental period to eliminate the periodic interference component caused by the fundamental frequency; the Fourier transform is performed on the phase-aligned time-domain distribution characteristics to obtain its spectral distribution, and the candidate frequency components below the preset low-frequency threshold are extracted from the spectrum; the candidate frequency components are screened based on the pre-stored power frequency integer multiple frequency range. The pre-stored power frequency integer multiple frequency range includes the integer multiple frequencies of the fundamental frequency and their preset deviation ranges. When the candidate frequency component falls within the deviation range of any integer multiple frequency, it is determined as a power frequency harmonic interference component and removed, and the remaining candidate frequency components are the characteristic frequency components representing the DC low-frequency circulating current characteristics.

[0067] During the phase alignment process, if the length of the time series of the time-domain distribution characteristics does not meet an integer multiple of the complete fundamental period, the missing data points are complemented by the linear interpolation method; the setting basis of the preset low-frequency threshold includes the maximum observed value of the DC circulating current frequency in the historical operation data of the energy storage station; the pre-stored power frequency integer multiple frequency range is determined by the allowable range of power frequency harmonics specified by the grid standard, and the specific deviation range is ±5% of the fundamental frequency. The candidate frequency components retained during the screening process are used for subsequent coupling coefficient calculation. When there are no remaining candidate components after screening, it is determined that there is no significant DC low-frequency circulating current characteristic at present, and the subsequent analysis steps are skipped.

[0068] Calculate the coupling coefficient between the characteristic frequency component and the grid natural frequency. The coupling coefficient is used to characterize the interaction intensity between the DC low-frequency circulating current and the grid frequency, including:

[0069] Obtain the values of the characteristic frequency component and the grid natural frequency. The characteristic frequency component is the specific frequency value of the candidate frequency component retained after screening, and the grid natural frequency is the fundamental frequency during the steady-state operation of the grid; calculate the absolute value of the frequency difference between the characteristic frequency component and the grid natural frequency. Among them, the calculation formula of the absolute value of the frequency difference is as follows:

[0070] ΔF=|f c -f g |

[0071] Where f c represents the characteristic frequency component, f g represents the grid natural frequency, and ΔF represents the absolute value of the frequency difference.

[0072] The calculation result of the absolute value of the frequency difference is used to quantify the frequency deviation degree between the characteristic frequency component and the grid natural frequency.

[0073] Normalize the absolute value of the frequency difference based on a preset frequency deviation threshold. The setting basis of the preset frequency deviation threshold includes the allowable maximum frequency deviation range in the historical operation data of the power grid. Specifically, the normalized frequency difference coefficient is generated through the following formula:

[0074]

[0075] where D max is the preset frequency deviation threshold, K n is the normalized frequency difference coefficient, and ΔF is the absolute value of the frequency difference.

[0076] When the absolute value of the frequency difference exceeds the preset frequency deviation threshold, the normalized frequency difference coefficient will be greater than 1, indicating that the degree of deviation of the characteristic frequency component from the inherent frequency of the power grid exceeds the allowable range.

[0077] Generate a coupling coefficient based on the product of the normalized frequency difference coefficient and the pre-stored circulating current intensity weight value. The setting method of the pre-stored circulating current intensity weight value includes: based on the statistical analysis of the correlation between the DC low-frequency circulating current amplitude and the power grid frequency offset in the historical operation data of the energy storage station, dynamically adjust the weight value. The specific calculation formula is as follows:

[0078] C c = K n × W r

[0079] where W r is the pre-stored circulating current intensity weight value, C c is the coupling coefficient, and K n is the normalized frequency difference coefficient.

[0080] The value range of the weight value is from 0.5 to 2.0. When the influence degree of the DC low-frequency circulating current amplitude on the power grid frequency offset in the historical data is relatively high, the weight value increases towards 2.0; when the influence degree is relatively low, the weight value decreases towards 0.5.

[0081] When the coupling coefficient exceeds the preset coupling threshold, it is determined that there is a strong interaction between the DC low-frequency circulating current and the power grid frequency. The setting rules of the preset coupling threshold include: according to the power grid stability requirements, when the coupling coefficient reaches 1.0, it is determined as a strong interaction; when the coupling coefficient is between 0.5 and 1.0, it is determined as a medium interaction; when the coupling coefficient is less than 0.5, it is determined as a weak interaction.

[0082] During the normalization process, if the absolute value of the frequency difference is less than or equal to the preset frequency deviation threshold, the normalized frequency difference coefficient is within the range of 0 to 1. At this time, the product of the coupling coefficient and the weight value directly reflects the interaction strength. If the absolute value of the frequency difference exceeds the preset frequency deviation threshold, the warning mechanism is triggered to dynamically correct the preset frequency deviation threshold. The correction method includes: according to the maximum allowable frequency deviation of the current power grid operation state, updating the threshold to 1.2 times the absolute value of the current frequency difference.

[0083] The dynamic adjustment process of the pre-stored circulating current strength weight value further includes: when detecting the subsynchronous oscillation caused by the DC low-frequency circulating current, according to the correlation between the oscillation amplitude and the frequency offset, update the weight value according to the following rules:

[0084] If the frequency offset increases linearly with the circulating current amplitude, the weight value increases by 0.1;

[0085] If the frequency offset fluctuates non-linearly with the circulating current amplitude, the weight value decreases by 0.1;

[0086] If the frequency offset has no significant correlation with the circulating current amplitude, the weight value remains unchanged.

[0087] When the coupling coefficient exceeds the preset coupling threshold, the extraction of the amplitude-frequency parameters of the subsynchronous oscillation specifically includes:

[0088] First, continuously collect the voltage fluctuation sequence and the frequency offset sequence of multiple complete power frequency cycles through the voltage sensor and the frequency measurement device. The number of multiple power frequency cycles is determined according to the minimum duration of the subsynchronous oscillation. For example, when the preset subsynchronous oscillation frequency band range is 0.1 - 10 Hz, the acquisition duration covers at least 3 power frequency cycles to ensure signal integrity; synchronize the time axes of the voltage fluctuation sequence and the frequency offset sequence. The specific method is to unify the timestamps of the two sequences to the same reference clock, and align the sampling points through the interpolation method to eliminate the time-domain deviation caused by the acquisition device delay or communication jitter, and generate a synchronous time-domain signal with consistent phases.

[0089] Perform the short-time Fourier transform on the synchronous time-domain signal to generate the time-frequency energy distribution map. The window length of the short-time Fourier transform is set to an integer multiple of the power frequency cycle. For example, the window length is 20 milliseconds (corresponding to one cycle of 50 Hz power frequency), and the window overlap rate is set to 50% to ensure the balance between time resolution and frequency resolution; the time-frequency energy distribution map obtained through the transformation contains three-dimensional information of time, frequency, and energy, where the energy value is obtained by calculating the square of the amplitude of the signal within each time-frequency unit.

[0090] Extract the frequency components with amplitudes exceeding the preset energy threshold from the time-frequency energy distribution map as candidate oscillation components. The setting basis of the preset energy threshold includes the historical statistical values of the grid background noise level. The specific method is as follows: Take 3 times the peak value of the background noise energy during the steady-state operation of the grid as the threshold. When the energy value of a certain frequency component exceeds this threshold, it is determined as a valid candidate oscillation component.

[0091] Screen the candidate oscillation components based on the pre-stored sub-synchronous oscillation frequency band range, which is set according to the operation experience of the energy storage station and the grid stability requirements. For example, it is set to 0.1 Hz to 10 Hz to exclude high-frequency harmonic and power frequency fundamental wave interferences; Determine the amplitude-frequency parameters of the sub-synchronous oscillation for the candidate oscillation components within the frequency band range and with a duration exceeding the preset duration threshold. The preset duration threshold is set according to the minimum maintenance period of the sub-synchronous oscillation. For example, it is set to 1 second to avoid transient interferences or noise pulses being misjudged as sub-synchronous oscillations.

[0092] During the time-axis synchronization alignment process, if the sampling rates of the voltage fluctuation sequence and the frequency offset sequence are inconsistent, the sampling rate of the low-frequency sampling sequence is increased to be the same as that of the high-frequency sampling sequence through linear interpolation to ensure that the time stamp alignment accuracy is within 1 millisecond; During the generation process of the time-frequency energy distribution map, if there are sudden amplitude changes or data missing in the window signal, data repair is performed through weighted averaging of the energy values of adjacent windows to prevent energy distribution distortion caused by spectrum leakage.

[0093] The dynamic adjustment method of the preset energy threshold includes: When the grid background noise level temporarily increases due to environmental factors (such as lightning strikes, equipment start-stop), the threshold is increased to 5 times the noise peak value according to the real-time noise monitoring data; When the noise level returns to normal, the threshold is adjusted back to 3 times. The update rule of the pre-stored sub-synchronous oscillation frequency band range includes: Dynamically adjust the upper and lower limits of the frequency band according to the frequency stability requirements issued by the grid dispatching system. For example, during the vulnerable operation period of the grid, the upper limit of the frequency band is compressed from 10 Hz to 5 Hz to improve the monitoring sensitivity.

[0094] The verification process of the preset duration threshold includes: Mark the start and end time points of the candidate oscillation components, calculate their time span. If the time span is greater than or equal to the preset duration threshold, it is determined as a valid sub-synchronous oscillation; If the time span is insufficient, it is classified as a transient interference and excluded. To ensure the screening accuracy, the setting of the preset duration threshold needs to cover at least 3 complete cycles of the sub-synchronous oscillation. For example, when the oscillation frequency is 2 Hz, the duration threshold is set to 1.5 seconds (3 / 2 Hz).

[0095] The acquisition of the potential resonance risk level specifically includes:

[0096] First, based on the amplitude parameter and the center frequency parameter in the amplitude-frequency parameters of subsynchronous oscillation, locate the amplitude-frequency intersection point on the pre-stored grid frequency response curve. The pre-stored grid frequency response curve is constructed from the frequency-amplitude characteristic data under typical operating conditions provided by the grid operator, specifically including the grid impedance amplitude information corresponding to different frequency points. The method for locating the amplitude-frequency intersection point is as follows: taking the center frequency parameter as the abscissa, find the corresponding frequency point on the pre-stored curve, and match the impedance amplitude of this frequency point with the amplitude parameter as the ordinate. When the difference between the two is less than the preset matching tolerance, determine this point as the amplitude-frequency intersection point. If there are multiple candidate intersection points, select the intersection point with the smallest impedance amplitude as the final positioning point.

[0097] Calculate the grid damping ratio and the subsynchronous oscillation damping ratio corresponding to the intersection point. The grid damping ratio is obtained by calculating the impedance phase angle at this intersection point on the pre-stored grid frequency response curve. The specific method is as follows: according to the damping ratio calculation formula defined in the IEEE standard, take half of the tangent value of the impedance phase angle as the grid damping ratio; the subsynchronous oscillation damping ratio is calculated through the subsynchronous oscillation decay characteristic curve collected in real time. The specific method is as follows: perform exponential fitting on the amplitude parameter of subsynchronous oscillation in time series, and take the ratio of the decay coefficient of the fitting curve to the oscillation frequency as the subsynchronous oscillation damping ratio. The units of both damping ratios are unified in percentage form.

[0098] Divide the grid damping ratio by the subsynchronous oscillation damping ratio to generate the damping ratio ratio. The calculation process of the damping ratio ratio includes: when the subsynchronous oscillation damping ratio is zero, directly determine that the damping ratio ratio is infinite; when the subsynchronous oscillation damping ratio is non-zero, perform the division operation and retain two decimal places for the result. Based on the pre-stored risk level mapping table, match the damping ratio ratio to the corresponding potential resonance risk level. The pre-stored risk level mapping table is constructed through statistical analysis of historical accident data, specifically including three risk level intervals: the damping ratio ratio corresponding to the high risk level is less than 1.0, the damping ratio ratio corresponding to the medium risk level is between 1.0 and 2.0, and the damping ratio ratio corresponding to the low risk level is greater than 2.0. The matching process includes: compare the calculated damping ratio ratio with the interval thresholds in the mapping table level by level, and select the risk level corresponding to the first interval that completely contains this ratio as the matching result.

[0099] When the damping ratio ratio is within the preset high-risk threshold range (i.e., less than 1.0), the potential resonance risk level is marked as the high-risk level, indicating that the grid damping capacity is insufficient to suppress the subsynchronous oscillation energy and there is an extremely high resonance risk; when the damping ratio ratio is within the preset medium-risk threshold range (i.e., between 1.0 and 2.0), it is marked as the medium-risk level, indicating that the grid and the oscillation damping capacity are in a critical balance state and continuous monitoring is required; when the damping ratio ratio exceeds 2.0, it is marked as the low-risk level, indicating that the grid has sufficient damping capacity to suppress the oscillation. If the damping ratio ratio is infinite, it is directly marked as the highest risk level and an emergency regulation is triggered.

[0100] The update rules of the pre-stored risk level mapping table include: regularly re-statistical historical data and adjust the threshold interval according to the change of the grid topology structure or the operation of newly added energy storage devices. For example, when the newly added battery cluster enhances the system damping characteristics, the high-risk threshold is adjusted from 1.0 to 0.8 to improve the sensitivity; the setting basis of the preset matching tolerance includes the accuracy error range of the measuring device. For example, when the amplitude measurement error of the voltage sensor is ±1%, the matching tolerance is set to 2%; the acquisition requirements of the subsynchronous oscillation attenuation characteristic curve include: continuously collecting the amplitude data of at least 5 oscillation cycles under the condition of no external regulation intervention to ensure the reliability of the fitting curve.

[0101] To verify the accuracy of the damping ratio calculation, the real-time calculation results of the grid damping ratio and the subsynchronous oscillation damping ratio need to be cross-checked with the off-line simulation results. When the deviation exceeds 10%, the data re-acquisition process is triggered; the cross-check method includes: inputting the real-time collected amplitude-frequency parameters into the pre-stored simulation model under the same working conditions, comparing the damping ratio output by the model with the real-time calculation results. If the deviation of three consecutive checks exceeds the standard, it is determined that the sensor data is abnormal and the self-check program is started. The self-check program includes: zero calibration of the voltage sensor and the frequency measurement device, re-collecting the reference signal and updating the damping ratio calculation parameters.

[0102] The marking process of the high-risk level further includes: when the risk level is high-risk, automatically associate the resonance accident records under similar working conditions in the historical database, and generate a reference report containing the accident time, regulation measures and consequences; the reference report is used to assist the operation and maintenance personnel in formulating regulation strategies. For example, when the historical records show that switching a specific battery cluster can effectively suppress the oscillation under similar working conditions, the same switching scheme is preferentially recommended.

[0103] The marking process of the medium-risk level includes: continuously recording the change trend of the oscillation amplitude and the frequency offset. When the trend index exceeds the preset warning line, even if the damping ratio ratio is still within the medium-risk interval, the risk level is temporarily raised to the high-risk level until the trend returns to stability.

[0104] The generation module adjusts the power output limit of the energy storage converter according to the potential resonance risk level and generates a switching instruction for the battery clusters to suppress the DC circulating current, including:

[0105] When the potential resonance risk level is determined to be a high-risk level, according to the dynamic power constraint rules issued by the power grid dispatching system, the power output limit of the energy storage converter is adjusted to the first preset proportion range, and the first preset proportion range is 50% to 70% of the current operating power value. The specific proportion is dynamically calculated and determined according to the weighted coefficient of the current power grid frequency offset and the circulating current amplitude; when the potential resonance risk level is a medium-risk level, based on the safe power threshold under similar working conditions in the historical operation data, the power output limit is adjusted to the second preset proportion range, and the second preset proportion range is 70% to 90% of the current operating power value. And the linear interpolation method is used to smoothly transition the limit adjustment process between the high and low risk levels to prevent power mutation.

[0106] When generating the switching instruction for the battery clusters to suppress the DC circulating current, based on the circulating current amplitude data of each battery cluster collected in real time by the DC detection module, a circulating current amplitude sorting list is generated in descending order of amplitude, and the battery clusters within the preset quantity in the circulating current amplitude sorting are preferentially switched. The preset quantity is dynamically set according to the total number of parallel battery clusters in the energy storage station topology. For example, when the total number of clusters is 100, the preset quantity is 3, that is, the top three battery clusters with the largest circulating current amplitude are switched; after the switching instruction is executed, according to the circulating current diffusion path model, the power output limits of adjacent battery clusters are synchronously adjusted to the third preset proportion range, and the third preset proportion range is 80% to 95% of the adjusted power limit to prevent the circulating current energy of the switched battery cluster from transferring to adjacent clusters.

[0107] The setting basis of the dynamic power constraint rules includes: the real-time frequency stability margin of the power grid, the upper limit of the heat dissipation capacity of the energy storage converter, and the state of charge balance of the battery clusters. When the frequency stability margin is lower than the preset safety threshold, the first preset proportion range shifts towards a lower limit; the calculation method of the weighted coefficient includes: after normalizing the frequency offset to the 0-1 interval, it is weighted and summed with the normalized result of the circulating current amplitude according to a 6:4 weight, and the corresponding proportion interval is mapped according to the summation result. The construction method of the circulating current diffusion path model includes: according to the DC bus impedance parameters between battery clusters and the historical circulating current propagation data, a transfer matrix of the circulating current amplitude and the correlation degree of adjacent clusters is established, and the range of adjacent clusters that need to synchronously adjust the power is determined based on the transfer matrix.

[0108] The preset quantity dynamic setting rules are as follows: when the total number of parallel battery clusters is less than 50, the preset quantity is 1; when the total number is between 50 and 200, the preset quantity is 2 to 5; when the total number exceeds 200, the preset quantity is 5 to 10, and the specific values are determined by grouping the distribution characteristics of the circulating current amplitude through a clustering algorithm. During the adjustment process of the third preset ratio range, if the adjustment of the power of adjacent clusters causes a new abnormal increase in the circulating current amplitude, the secondary switching instruction generation process is triggered to perform additional switching on the newly added high-circulating-current-amplitude battery clusters until the circulating current amplitude drops below the preset safety threshold.

[0109] To ensure the smoothness of power adjustment, the switching process between the first and second preset ratio ranges needs to meet the following conditions: when the risk level is downgraded from high risk to medium risk, the power limit gradually recovers from the first preset ratio range to the second preset ratio range at a rate of 1% per second; when the risk level is upgraded, the power limit drops from the second preset ratio range to the first preset ratio range at a rate of 2% per second. The execution interval of the switching instruction is dynamically adjusted according to the circulating current suppression effect. If the decline rate of the circulating current amplitude within 30 seconds after switching does not meet the expectation, the generation interval of the next switching instruction is shortened to 50% of the original interval.

[0110] During the generation process of the circulating current amplitude sorting list, if the difference in the circulating current amplitudes of multiple battery clusters is less than the preset tolerance threshold, secondary sorting is performed according to the battery cell voltage consistency index within the cluster, and the battery cluster with poorer voltage consistency is preferentially switched; the voltage consistency index is obtained by calculating the standard deviation of the cell voltages within the cluster, and the larger the standard deviation, the poorer the consistency. The priority rules for adjacent cluster power adjustment include: preferentially adjusting the adjacent cluster with the highest DC impedance matching degree with the switched cluster, and the impedance matching degree is determined by calculating the similarity of the busbar resistance and inductance parameters. The higher the similarity, the higher the adjustment priority.

[0111] The power adjustment and switching instructions generated through the above steps will be sent to the energy storage converter and battery cluster controller in real time. When the circulating current amplitude of the system does not exceed the preset safety threshold for 2 minutes after the instruction is executed, it is determined that the regulation is effective and the temporary power limit is lifted; if the circulating current amplitude exceeds the standard again after the limit is lifted, the risk assessment and regulation instruction generation process is triggered again.

[0112] The above formulas are all dimensionless and take their numerical values for calculation. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation.

[0113] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more collections of available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0114] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and modules described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0115] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules is only a logical function division, and there can be other division methods in actual implementation. For example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces, and the indirect couplings or communication connections of the devices or modules can be in electrical, mechanical, or other forms.

[0116] The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules. They can be located in one place or distributed to multiple network modules. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0117] In addition, in each embodiment of the present application, each functional module may be integrated into one processing module, may exist physically alone for each module, or two or more modules may be integrated into one module.

[0118] If the above-mentioned function is implemented in the form of a software functional module and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0119] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

[0120] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. The power quality monitoring device of a large centralized energy storage station is characterized by: The device includes a DC detection module, an AC acquisition module, a fusion analysis module and a control generation module. The device performs the following steps: The time domain distribution characteristics of DC low-frequency circulating current are extracted through the DC detection module; The time domain distribution characteristics are input into the fusion analysis module, combined with the grid connection point voltage fundamental frequency obtained in real time by the AC acquisition module, to extract the characteristic frequency components of the DC low-frequency circulating current; Calculate the coupling coefficient between the characteristic frequency component and the natural frequency of the power grid. The coupling coefficient is used to characterize the interaction strength between the DC low-frequency circulating current and the power grid frequency. When the coupling coefficient exceeds a preset coupling threshold, the amplitude-frequency parameters of the subsynchronous oscillation are extracted; Determine the potential resonance risk level according to the amplitude-frequency parameters of the subsynchronous oscillation; The control generation module adjusts the power output limit of the energy storage inverter according to the potential resonance risk level, and generates battery cluster switching instructions to suppress DC circulating current.

2. The power quality monitoring device for a large centralized energy storage station according to claim 1, characterized in that: The time domain distribution characteristics are extracted based on the root mean square sequence of the current fluctuation amplitude by synchronously collecting the instantaneous values ​​of the DC voltage and DC current of each battery cluster.

3. The power quality monitoring device for a large centralized energy storage station according to claim 1, characterized in that: The amplitude-frequency parameters of subsynchronous oscillation are obtained by generating a time-frequency energy distribution spectrum through short-time Fourier transform based on the voltage fluctuation sequence and frequency offset sequence obtained by the AC acquisition module.

4. The power quality monitoring device for a large centralized energy storage station according to claim 1, characterized in that: The potential resonance risk level is obtained by calculating and mapping the ratio of the grid damping ratio to the subsynchronous oscillation damping ratio at the intersection point based on the intersection position of the amplitude and center frequency parameters of the subsynchronous oscillation on the pre-stored grid frequency response curve.

5. The power quality monitoring device for a large centralized energy storage station according to claim 2, characterized in that: The acquisition of time domain distribution features specifically includes: Synchronously collect the instantaneous value of DC voltage and DC current of each battery cluster, and calculate the current fluctuation amplitude of each battery cluster; Sliding interception of the current fluctuation amplitude is performed based on a preset time window to generate a root mean square sequence of the current fluctuation amplitude; According to the difference between the peak distribution and the valley distribution of the RMS sequence, the time domain distribution characteristics characterizing the intensity of the DC low-frequency circulating current are extracted.

6. The power quality monitoring device for a large centralized energy storage station according to claim 1, characterized in that: The time domain distribution characteristics are input into the fusion analysis module, combined with the grid connection point voltage fundamental frequency obtained in real time by the AC acquisition module, to extract the characteristic frequency components of the DC low-frequency circulating current, including: The fundamental frequency of the grid connection point voltage obtained in real time by the AC acquisition module; Phase-align the time domain distribution characteristics with the fundamental frequency of the grid connection point voltage to eliminate the interference component of the fundamental frequency on the time domain distribution characteristics; Performing Fourier transform on the time domain distribution characteristics after eliminating interference, and extracting candidate frequency components of the DC low-frequency circulating current; The candidate frequency components are screened based on the pre-stored frequency range of integer multiples of the power frequency, the candidate components belonging to the frequency range of integer multiples of the power frequency are eliminated, and the remaining candidate components are retained as characteristic frequency components.

7. The power quality monitoring device for a large centralized energy storage station according to claim 1, characterized in that: Calculate the coupling coefficient between the characteristic frequency component and the natural frequency of the power grid. The coupling coefficient is used to characterize the interaction strength between the DC low-frequency circulating current and the power grid frequency, including: Calculate the absolute value of the frequency difference between the characteristic frequency component and the natural frequency of the power grid; Normalizing the frequency difference absolute value based on a preset frequency deviation threshold to generate a normalized frequency difference coefficient; The coupling coefficient is generated according to the product of the normalized frequency difference coefficient and the pre-stored circulating current intensity weight value; the larger the coupling coefficient, the stronger the strong interaction between the DC low-frequency circulating current and the grid frequency.

8. The power quality monitoring device for a large centralized energy storage station according to claim 3, characterized in that: When the coupling coefficient exceeds a preset coupling threshold, the extraction of the amplitude-frequency parameters of the subsynchronous oscillation specifically includes: Based on the AC acquisition module, the voltage fluctuation sequence and frequency offset sequence of multiple power frequency cycles are continuously collected; Perform time axis synchronization alignment on the voltage fluctuation sequence and the frequency offset sequence to generate a synchronized time domain signal; Performing short-time Fourier transform on the synchronized time domain signal to generate a time-frequency energy distribution spectrum including a three-dimensional distribution of time, frequency and energy; Extracting frequency components whose amplitudes exceed a preset energy threshold from the time-frequency energy distribution spectrum as candidate oscillation components; The candidate oscillation components are screened based on the pre-stored sub-synchronous oscillation frequency band range, and the components within the frequency band range and whose duration exceeds a preset duration threshold are determined as the amplitude-frequency parameters of the sub-synchronous oscillation.

9. The power quality monitoring device for a large centralized energy storage station according to claim 4, characterized in that: The acquisition of potential resonance risk level specifically includes: According to the amplitude parameter and the center frequency parameter in the amplitude-frequency parameter of the subsynchronous oscillation, the amplitude-frequency intersection position is located on the pre-stored power grid frequency response curve; Calculate the grid damping ratio and subsynchronous oscillation damping ratio corresponding to the intersection position; Dividing the grid damping ratio by the subsynchronous oscillation damping ratio generates a damping ratio ratio value; Based on a pre-stored risk level mapping table, the damping ratio value is matched to the corresponding potential resonance risk level; When the damping ratio value is within a preset high risk threshold range, the potential resonance risk level is marked as a high risk level; When the damping ratio value is within a preset medium risk threshold range, the potential resonance risk level is marked as a medium risk level.

10. The power quality monitoring device for a large centralized energy storage station according to claim 1, characterized in that: The control generation module adjusts the power output limit of the energy storage converter according to the potential resonance risk level, and generates a battery cluster switching instruction to suppress the DC circulating current, including: When the potential resonance risk level is a high risk level, adjusting the power output limit of the energy storage converter to a first preset ratio range; When the potential resonance risk level is a medium risk level, adjusting the power output limit to a second preset ratio range; Generate a battery cluster switching instruction for suppressing the DC circulating current based on the DC circulating current amplitude data acquired in real time, and give priority to switching the battery clusters whose circulating current amplitude ranking is within a preset number; After the switching instruction is executed, the power output limits of the adjacent battery clusters are synchronously adjusted to a third preset ratio range.

Images