Anti-electromagnetic interference method for flywheel energy storage system of data center

Abstract

The invention relates to the technical field of anti-electromagnetic interference, in particular to an anti-electromagnetic interference method for a flywheel energy storage system of a data center, which comprises the following steps: acquiring change rate data of the rotating speed and the angular speed of a flywheel, judging the attribution of a disturbance interval, recording change characteristics, extracting an interference identifier, comparing the interference identifier with impedance change, and generating a switching node. Frequency response and reflection coefficient characteristics of the filtering unit are detected, a coupling channel generation label is identified, and output fluctuation is evaluated to generate an anti-interference capability state label. According to the method, the surface resistivity of the shielding layer is dynamically switched according to the comparison result of the specific interference triggering identification bit and the signal link impedance change state in each disturbance stage through classification judgment and correlation marking of inertial disturbance, so that the effect of suppressing electromagnetic interference in real time is optimized; the influence degree of electromagnetic interference can be dynamically evaluated in the system operation process, and the adaptability of the energy storage system to the electromagnetic interference is judged and adjusted in real time.

Description

Technical Field

[0001] This invention relates to the field of electromagnetic interference suppression technology, and in particular to an electromagnetic interference suppression method for a flywheel energy storage system in a data center. Background Technology

[0002] The field of electromagnetic interference (EMI) mitigation technology involves the identification, analysis, and suppression of EMI sources, aiming to ensure the stable operation of electronic equipment or systems in complex electromagnetic environments. This field covers the generation mechanisms, coupling paths, and sensitive carriers of EMI, focusing on limiting the impact of interference signals on target systems through various methods such as shielding, filtering, grounding, isolation, and layout optimization to ensure system electromagnetic compatibility. Especially in environments with high density, high speed, and high power electronic equipment, such as data centers, power systems, and avionics equipment, EMI mitigation measures not only require high suppression ratios and high stability but also must adapt to engineering constraints related to system operational continuity, safety, and thermal management.

[0003] The electromagnetic interference (EMI) mitigation method for flywheel energy storage systems in data centers falls under the category of EMI compatibility (EMC) methods for high-reliability power supply scenarios. Specifically, it addresses the vulnerability of flywheel energy storage systems to strong electromagnetic interference in data center environments. The method models the interference coupling paths within the energy storage system's power supply units, signal links, and control modules. Combined with hardware isolation, signal shielding, and filtering network design, it effectively suppresses typical EMI sources such as high-frequency pulses and harmonic interference. Its application can improve the stability of flywheel energy storage systems during critical moments such as power supply switching and sudden load responses in data centers, thereby enhancing the overall electromagnetic robustness and operational reliability of the data center power supply system.

[0004] Traditional anti-interference methods rely on static hardware isolation, fixed signal shielding measures, and filter network design. They do not consider the actual disturbance characteristics changes during the operation of flywheel energy storage systems, making the anti-interference strategy relatively simple and passive. They cannot respond and adjust in real time to the dynamic changes of electromagnetic interference sources and the internal state of the energy storage system. When faced with the superposition of electromagnetic interference of different frequencies and rapid changes in interference intensity, the system shielding measures are prone to failure or the filter frequency band setting deviates from the actual needs, resulting in a decline in electromagnetic compatibility performance and affecting the timeliness and reliability of power supply switching and load surge response. For example, when the energy storage system encounters rapidly changing high-frequency interference sources in a specific operating range, the fixed configuration of shielding and filtering strategies is difficult to suppress the interference in time, causing the system output voltage fluctuation to exceed the allowable range, triggering the risk of temporary power outages or equipment failures in the data center, and affecting the overall operational stability and service quality of the data center. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and propose an anti-electromagnetic interference method for flywheel energy storage systems in data centers.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a method for resisting electromagnetic interference in a flywheel energy storage system for a data center, comprising the following steps: S1: Obtain the output data of the flywheel energy storage unit during the operation of the power access node, determine the interval assignment of the angular velocity change rate and the four fixed angular velocity fluctuation interval thresholds, record the running period and change characteristics corresponding to the speed data segment in the disturbance interval, and generate an inertial disturbance assignment information set. S2: Based on the inertial disturbance attribution information set, combined with the power input port input impedance change status of the corresponding signal link, compare it item by item with the shielding layer surface resistivity switching threshold to determine whether the switching conditions are met, and generate a shielding state switching node set. S3: Call the set of shielding state switching nodes, detect the complex impedance frequency response curves of adjacent upstream and downstream filter units, and detect whether the maximum slope of the first derivative of the reflection coefficient exceeds the set coupling judgment threshold, obtain the channel number and coupling type label that meet the sudden change characteristics, and generate coupling node identification label information. S4: Based on the identification tag information of the coupling node, extract the output voltage time series at the corresponding channel of the flywheel energy storage output end, and calibrate the degree of interference-induced effect to generate an energy storage output fluctuation level identifier.

[0007] The present invention improves upon this invention by including the following: the inertial disturbance attribution information set includes a flywheel operation disturbance level label, a disturbance period boundary timestamp, a disturbance segment identification number, a speed change curve index number, and a corresponding disturbance state index; the shielding state switching node set includes a trigger node number, a shielding layer switching state label, a trigger time index, a channel impedance state classification value, and a shielding layer impedance switching command; the coupling node identification label information specifically includes a channel coupling identifier number, a slope change point quantity index, a reflection coefficient fluctuation level, a filter unit coupling type classification, and a frequency crossing region slope fitting number; and the energy storage output fluctuation level identifier specifically includes a steady-state deviation level, a fluctuation characteristic number, a distortion level label, a voltage anomaly persistence classification, and a corresponding coupling channel index identifier.

[0008] The present invention is improved in that the step of obtaining the inertial disturbance attribution information set is specifically as follows: S111: Acquire the output data of the flywheel energy storage unit during operation at the power access node, including instantaneous speed value, angular velocity change rate and maximum angular velocity fluctuation amplitude within three cycles, synchronously record the speed difference and sampling time difference corresponding to each moment in each cycle, calculate the angular velocity change rate sequence range corresponding to each cycle, and generate the cycle angular velocity change interval value. S112: Based on the periodic angular velocity change interval value, extract three continuous periods of equal time width from the sampling sequence, locate and extract the maximum angular velocity change rate within each period, and calculate the maximum speed difference and average change trend within the corresponding period to generate three fluctuation characteristic statistical values. S113: Based on the statistical values ​​of the three wave characteristics, according to the maximum value of the rate of change of angular velocity, the interval assignment is determined by comparing with the four angular velocity wave interval thresholds. The wave segment number and the corresponding sampling time period position are recorded as being in the disturbance interval. A corresponding operating status label is established, and an inertial disturbance assignment information set is generated.

[0009] The present invention is improved in that the step of obtaining the set of shielding state switching nodes is specifically as follows: S211: Based on the inertial disturbance attribution information set, according to the time period in the medium disturbance and high disturbance range, the disturbance identifier bit, node sampling time and corresponding rotational speed change label of each data node are collected in the time period, and the disturbance identifier state sequence is segmented and classified in chronological order to obtain the node number and trigger identifier distribution in each disturbance continuous segment, and generate the disturbance identifier sequence feature value. S212: Based on the feature value of the disturbance identification sequence, perform impedance detection on the power access port in the signal link, obtain the instantaneous input impedance value change sequence corresponding to the port, compare it with the surface resistivity switching threshold set by the shielding structure, determine whether the corresponding impedance value exceeds the switching judgment boundary, and record the number and impedance change direction of the switching condition to generate the node impedance change trend value. S213: Call the node impedance change trend value, mark each number that meets the switching conditions and the corresponding time as a valid switching point, count the synchronization of disturbance indicators and the consistency of impedance directionality at the switching point, and filter switching event nodes with complete response sequences. Calculate and obtain the switching event response value, and include the numbers whose response values ​​are greater than the set event validity threshold into the filter set. Obtain the trigger time, number index and channel status mark of each valid node, and establish a set of shielded state switching nodes.

[0010] The present invention is improved in that the step of obtaining the identification tag information of the coupling node is specifically as follows: S311: Call the set of shielding state switching nodes, and according to the trigger node number, sequentially search the corresponding module number in the frequency selective filter signal chain structure, locate the adjacent upstream and downstream filter unit module indexes, map the three structure pairs by number, obtain their corresponding relative sequence positions in the signal channel, and generate filter module combination index value. S312: Based on the combined index value of the filter module, call the micrometer node parameters at the frequency crossing point in each combined structure, collect the equivalent inductance value, equivalent capacitance value and reflection coefficient frequency domain response data, insert the equivalent parameter values ​​into the complex impedance frequency response curve, extract the complex impedance slope change point value under the frequency band according to the derivative change behavior of the response curve, calculate the deviation between the maximum slope value of the first derivative of the reflection coefficient and the set coupling change threshold, calculate and obtain the coupling behavior strength value in the combined structure, and generate a coupling response strength set; S313: Based on the set of coupling response strengths, extract the combined structure numbers whose strength values ​​are greater than the coupling mutation threshold, match the original node and channel index, generate coupling feature identifier bits at the corresponding positions, and generate a hierarchy based on the coupling position and parameter jump mode label. Extract the coupling type label, channel number and structure positioning index on each signal channel to establish coupling node identification label information.

[0011] The present invention is improved in that the step of obtaining the energy storage output fluctuation level identifier is specifically as follows: S411: Based on the identification tag information of the coupling node, according to the marked coupling channel number, the corresponding number is located sequentially in the output channel of the flywheel energy storage system, the output voltage time series of the channel is collected in the continuous sampling period before and after the disturbance event, and the sampling frequency and total number of periods of the sequence are recorded to generate a voltage time series sampling dataset. S412: Based on the voltage time-series sampling dataset, analyze the trend of voltage effective value change in each cycle, extract the voltage effective value, total harmonic distortion and number of fluctuation duration cycles in each cycle, use the initial steady-state voltage reference value to perform differential calculation and normalization transformation on the index, calculate and obtain the fluctuation intensity factor under the disturbance range, and generate a fluctuation intensity factor set. S413: Call the set of fluctuation intensity factors, combine it with the preset voltage fluctuation level division interval, mark the fluctuation level label corresponding to each channel, and organize the channel number, fluctuation interval and level sequence number to establish the energy storage output fluctuation level identifier.

[0012] The present invention has an improvement, wherein the method further includes the following steps: S5: Call the energy storage output fluctuation level identifier, assign the anti-interference capability status label of the current flywheel energy storage system under the measured interference scenario according to the set voltage stability level classification standard, and evaluate whether the current system status has the conditions to continue to operate by combining the current operation task level matching relationship, and generate the anti-interference capability status label. The anti-interference capability status label includes an anti-interference capability classification flag, a task adaptation status label, a stability level matching value, a system allowed operation status instruction, and an anomaly suppression mode label.

[0013] The present invention is improved in that the step of obtaining the anti-interference capability status tag is specifically as follows: S511: Call the energy storage output fluctuation level identifier, extract the voltage deviation upper limit, harmonic distortion upper limit and duration critical value of each level interval according to the preset voltage stability level classification standard, and match each group of fluctuation levels with the classification standard by channel number as index to determine the level segment to which it belongs and generate voltage stability level value. S512: Based on the voltage stability level value, compare it with the task level requirements undertaken by the flywheel energy storage unit in the current operating cycle, extract the power supply continuity tolerance boundary, voltage stability minimum level standard and control response time limit threshold set under the task level, perform difference calculation and sort the voltage level and task standard, filter the channel number that meets the operating requirements, and generate task operation adaptability index. S513: Call the task operation adaptation index, extract the channel identifiers with positive level matching status, record the channel number, level range and allowed operation status identifier value, and combine and classify the adaptation status and stability status according to the evaluation rules to generate anti-interference capability status labels.

[0014] Compared with the prior art, the advantages and positive effects of the present invention are as follows: In this invention, by analyzing the instantaneous rotational speed, angular velocity change rate, and intra-cycle angular velocity fluctuation amplitude in the output data of the flywheel energy storage system, the inertial disturbance is classified and associated. Based on the disturbance characteristics, the operation stages of medium and high disturbances are clearly distinguished. Based on the comparison results of specific interference triggering flags and signal link impedance changes in each disturbance stage, the surface resistivity of the shielding layer is dynamically switched to optimize the real-time suppression of electromagnetic interference. Through quantitative analysis and difference marking of the complex impedance characteristics and reflection coefficient change trends of the filter unit, the frequency domain coupling channel of electromagnetic interference is accurately identified. Combined with the comprehensive differential calculation of the voltage fluctuation amplitude, harmonic distortion, and disturbance duration at the energy storage output terminal, the assessment of the electromagnetic interference impact is more quantitative and accurate. The impact of electromagnetic interference can be dynamically assessed during system operation, and anti-interference capability status flags can be automatically generated accordingly. This enables real-time judgment and adjustment of the energy storage system's adaptability to electromagnetic interference, improving the response agility and electromagnetic interference resistance of the energy storage system in high-frequency pulse and harmonic interference scenarios, and ensuring the power supply reliability and operational stability of the data center in power switching and load change scenarios. Attached Figure Description

[0015] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a flowchart illustrating the process of obtaining the inertial disturbance attribution information set according to the present invention; Figure 3This is a flowchart illustrating the process of obtaining the set of nodes for shielding state switching according to the present invention; Figure 4 This is a flowchart illustrating the process of obtaining the identification tag information of the coupling node in this invention; Figure 5 This is a flowchart illustrating the process of obtaining the energy storage output fluctuation level identifier according to the present invention; Figure 6 This is a flowchart illustrating the process of obtaining the anti-interference capability status label according to the present invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0017] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0018] Please see Figure 1 This invention provides a technical solution: a method for resisting electromagnetic interference in a flywheel energy storage system for a data center, comprising the following steps: S1: Acquire the output data of the flywheel energy storage unit during the operation of the power access node, including instantaneous speed value, angular velocity change rate and maximum angular velocity fluctuation amplitude within three periods, and determine the interval assignment of the angular velocity change rate and four fixed angular velocity fluctuation interval thresholds, record the running period and change characteristics corresponding to the speed data segment within the disturbance interval, and generate an inertial disturbance assignment information set; S2: Based on the inertial disturbance attribution information set, according to the time period in the medium and high disturbance range, the interference triggering flag bit in the time period is extracted. Combined with the power input port input impedance change status of the corresponding signal link, it is compared with the shielding layer surface resistivity switching threshold item by item to determine whether the switching condition is met. The node number and trigger time that meet the condition are recorded to generate the shielding state switching node set. S3: Call the shielding state switching node set, locate the module in the frequency selection filter signal chain according to the trigger node number, detect the complex impedance frequency response curve of the adjacent upstream and downstream filter units, collect the equivalent inductance value, equivalent capacitance value and reflection coefficient frequency domain curve of the micro-measurement signal node at the frequency crossing point between modules, calculate the value of the complex impedance slope change point based on the curve fitting result, and detect whether the maximum slope of the first derivative of the reflection coefficient exceeds the set coupling judgment threshold, obtain the channel number and coupling type label that meet the sudden change characteristics, and generate coupling node identification label information; S4: Based on the identification tag information of the coupling node, according to the marked coupling channel number, extract the output voltage time series at the corresponding channel of the flywheel energy storage output end, analyze the three information of effective value fluctuation amplitude, voltage harmonic distortion degree and disturbance duration of the sequence, perform differential calculation with the initially set output voltage steady state reference value, and calibrate the significance of the interference-introduced effect to generate an energy storage output fluctuation level identifier. S5: Call the energy storage output fluctuation level identifier, assign the anti-interference capability status label of the current flywheel energy storage system under the measured interference scenario according to the set voltage stability level classification standard, and evaluate whether the current system status has the conditions to continue to operate by combining the current operation task level matching relationship, and generate the anti-interference capability status label. The inertial disturbance attribution information set includes flywheel operation disturbance level label, disturbance period boundary timestamp, disturbance section identification number, speed change curve index number, and corresponding disturbance state index. The shielding state switching node set includes trigger node number, shielding layer switching state label, trigger time index, channel impedance state classification value, and shielding layer impedance switching command. The coupling node identification label information specifically includes channel coupling identifier number, slope change point number index, reflection coefficient fluctuation level, filter unit coupling type classification, and frequency crossing region slope fitting number. The energy storage output fluctuation level identifier specifically refers to steady-state deviation level, fluctuation characteristic number, distortion level label, voltage anomaly persistence classification, and corresponding coupling channel index identifier. The anti-interference capability status label includes anti-interference capability level mark, task adaptation status label, stability level matching value, system allowed operation status command, and anomaly suppression mode label.

[0019] Please see Figure 2 The specific steps for obtaining the inertial disturbance attribution information set are as follows: S111: Acquire the output data of the flywheel energy storage unit during operation at the power access node, including instantaneous speed value, angular velocity change rate and maximum angular velocity fluctuation amplitude within three cycles, synchronously record the speed difference and sampling time difference corresponding to each moment in each cycle, calculate the angular velocity change rate sequence range corresponding to each cycle, and generate the cycle angular velocity change interval value. To obtain the instantaneous rotational speed of the flywheel energy storage unit during operation at the power access node and the rate of change of angular velocity in each cycle of continuous sampling, a high-frequency speed sensor is first configured at the power access node. The sampling frequency records the flywheel rotation speed value, and the sampling time period is set to [value]. Then each period contains Each instantaneous rotational speed value is calculated using the rotational speed difference between consecutive sampling points. Divide by the corresponding time interval The rate of change of angular velocity is obtained, in order to for Periodic grouping, forming Each cycle unit extracts all speed values ​​for each cycle and calculates the maximum speed difference. Record the time difference between the first and last sampling points in each cycle. , and then The periodic average angular velocity change rate was obtained and formed into arrays. And record the upper and lower bounds of the rate of change for each cycle to form a range of change. For example, in a certain operating cycle, the rotational speed changes from... Increase to , , The rate of change of angular velocity during this period is... Further record the upper and lower limits of the rate of change for each cycle, forming The structure is an array of intervals, which ultimately collects all the intervals of angular velocity change rate for all periods, generating the interval values ​​of angular velocity change for each period.

[0020] S112: Based on the periodic angular velocity change interval, three continuous periods of equal time width are extracted from the sampling sequence. The maximum value of the angular velocity change rate in each period is located and extracted. The maximum speed difference and average change trend in the corresponding period are calculated to generate three fluctuation characteristic statistical values. Based on the range of periodic angular velocity changes, three continuous periods of equal time width are extracted from the sampling sequence. The cycle is divided into segments A (cycle 13), B (cycle 46), and C (cycles 7-9). The maximum value of the rate of change of angular velocity in each segment is located and extracted. For example, the maximum rate of change in segment A is... Section B is Section C is The maximum rate of change values ​​in each segment are combined into a vector. And calculate the maximum speed difference for each segment, for example, segment B from Rise to , The duration is The average rate of change Finally, the maximum rate of change, the speed difference, and the average rate of change of each segment are combined as a characteristic parameter group to form a characteristic statistics table, as shown in Table 1.

[0021] Table 1. Statistical Table of Three-Segment Periodic Fluctuation Characteristics As shown in Table 1, after extracting the complete numerical values ​​of each periodic feature, it can be used for the next step of interval attribution judgment to generate three segments of fluctuation feature statistics.

[0022] S113: Based on the statistical values ​​of the three-segment fluctuation characteristics, according to the maximum value of the rate of change of angular velocity, the interval assignment is determined by comparing with the threshold values ​​of the four angular velocity fluctuation intervals. The fluctuation segment number and the corresponding sampling time period position are recorded as being in the disturbance segment. A corresponding operating status label is established, and an inertial disturbance assignment information set is generated. Based on the maximum angular velocity change rate among the three fluctuation characteristic statistics, the interval assignment is determined by comparing it with the angular velocity fluctuation interval threshold. The four angular velocity fluctuation interval thresholds are set based on: combining the allowable load response bandwidth and speed fluctuation tolerance of the flywheel system design, the stable segment threshold is determined as follows. The light disturbance segment is The middle disturbance segment is The high-disturbance segment is The reference standard for this setting is the system's rated speed response rate. The maximum allowed is not more than its The rationale for segmenting the system downwards by a factor of two is that the maximum dynamic response fluctuation during system startup, shutdown, and load surge phases generally does not exceed [a certain value]. This sets the classification boundaries; according to this standard, segment A... Included in the middle disturbance segment, segment B Classified as high-disturbance segment, segment C Classified as a lightly disturbed segment, the corresponding sampling time period is recorded as follows: , , And by mapping segment numbers to time period numbers one-to-one, fluctuation segment identifier pairs are generated. The medium and high disturbance segments are respectively assigned the disturbance segment status codes. and This ultimately forms an array of disturbance segment numbers. Time period index It is bound to it and recorded as a disturbance state mapping table with the structure { : , The result of the disturbance is written into the current time period's operating status label mapping table, and finally, an inertial disturbance attribution information set is established.

[0023] Please see Figure 3The specific steps for obtaining the set of nodes for masking state switching are as follows: S211: Based on the inertial disturbance attribution information set, according to the time period in the medium and high disturbance range, the disturbance identifier bit, node sampling time and corresponding rotational speed change label of each data node are collected in the time period, and the disturbance identifier state sequence is segmented and classified in chronological order to obtain the node number and trigger identifier distribution in each continuous disturbance segment, and generate the disturbance identifier sequence feature value. Obtain disturbance triggering state data within the medium and high disturbance range. Using the time intervals identified as medium and high disturbance levels in the previous step as the input range, and the output field of the flywheel control system's built-in disturbance recording module as the input source, extract the disturbance flag bit for each data node within this time period, and use the state variables... This indicates whether the node is marked as having a valid disturbance state, and also records the timestamp for each node. And by the node speed difference label Mutation identification is performed, where a mutation is defined as the difference in rotational speed between two adjacent nodes exceeding a preset mutation identification threshold. This is used to identify and label the corresponding nodes, if within a time period and If a sudden change in rotational speed occurs, then the node... These nodes are marked as mutation nodes. They are then sorted in ascending order of timestamps. Segments with consecutive perturbation state bits set to 1 are grouped into segments, each segment labeled as a perturbation sequence block, and its starting node number is recorded. End number and all of that section Sequence Distribution. Taking a three-segment perturbation example, if the perturbation identifier 1 appears consecutively at node numbers 1014, 2730, and 51~55, then three sets of perturbation identifier subsequences are formed, the structure of which can be recorded as follows: The generated three-segment node perturbation sequence is the perturbation identifier sequence feature value.

[0024] S212: Based on the characteristic value of the disturbance identification sequence, perform impedance detection on the power input port in the signal link, obtain the instantaneous input impedance value change sequence corresponding to the port, compare it with the surface resistivity switching threshold set by the shielding structure, determine whether the corresponding impedance value exceeds the switching judgment boundary, and record the number and impedance change direction of the switching condition, and generate the node impedance change trend value. Based on the node number of each disturbance segment located by the characteristic value of the disturbance identifier sequence, locate its corresponding power input port on the signal link physical diagram, and measure its input impedance value using an impedance monitoring probe. With sampling period Establish a node impedance change sequence based on this. Then, normalization is used to transform it into ,in , This represents the current period's total channel impedance variation range, where the variation is the ratio of the impedance difference between preceding and following nodes to the total range. The surface resistivity switching threshold is set based on the following: the shielding material is a nickel carbide composite material, and its surface resistivity in its low-resistivity state is... The high impedance switching point is set as its times This serves as the switching boundary, detecting whether the resistivity of a channel associated with a node in the impedance sequence exceeds this critical value. For example, the impedance measurement value corresponding to node 37 is... Its previous node value is ,but If the node is marked as having a valid disturbance at this time If a node is identified as a suspicious node, it will be further evaluated, and the nodes that meet the criteria will be numbered and their impedance increase / decrease directions will be marked. or The node impedance change trend value is generated by recording the sequence list.

[0025] S213: Call the node impedance change trend value, mark each node that meets the switching conditions with its corresponding time as a valid switching point, statistically analyze the synchronization of disturbance indicators and the consistency of impedance directionality at the switching points, and filter switching event nodes with complete response sequences using the formula: ; The operation obtains the switching event response value, and the numbers of the response values ​​that are greater than the set event validity threshold are included in the filter set. The trigger time, number index and channel status flag of each valid node are obtained, and a set of blocking status switching nodes is established. in, The response value represents the set of nodes switching between shielded and shielded states. Representing the The normalized value of the input impedance change at the nth node represents the value of the nth node. The standardized value of the impedance increase or decrease at each node relative to the impedance variation range in the entire channel. For the first The disturbance flag for each node indicates its valid status, with a value of 0 or 1, representing whether the node is within a valid disturbance segment. For the first The normalized value of the sampling time of each node represents the proportion of that time point within the current disturbance segment. The normalized value of the initial sampling time of the disturbance segment is set to 0 or the starting reference point. This is the index of the perturbed node. This represents the total number of nodes detected in the disturbed segment. Call the node impedance change trend value, and select the one that satisfies , The nodes are sorted by number and matched with their corresponding timestamps. Combine and mark as valid switching points, then assign these switching points to disturbance flag states. The impedance direction is checked for consistency. If it is within the perturbation subsequence and the impedance change direction is positive with the perturbation indicator, it is included in the screening pool. The formula is input using this node set: ; in, This is the normalized input impedance change. This is the valid disturbance status flag. This represents the normalized time span of the sampling time at this node relative to the time of the starting node in this segment. This represents the total number of valid nodes participating in the determination. Taking three groups of nodes as an example, if node 20 is present: , , Node 25: , , Node 29: , , ; Substitute into the calculation: ; Event validity threshold set to This value comes from system simulation experiments. The 80th percentile value during the second disturbance handover is used as the cutoff point to form the judgment benchmark under the average high response condition. Substituting this value into the formula yields the following result: Exceeding the threshold This indicates that the switching event has occurred. The corresponding node number, timestamp, and switching direction marker are recorded and written into the status table, ultimately establishing a set of nodes for the masked state switching. The advantage of this formula is that, through the coupled calculation of three factors—disturbance amplitude, disturbance effectiveness, and sampling delay—a quantitative response value index is formed for instantaneous switching behavior without the need to introduce master control timing or external time scales. This makes the response judgment rely solely on the data-driven process of the node itself, making it more real-time and independent.

[0026] Please see Figure 4 The specific steps for obtaining the tag information of the coupled node identification are as follows: S311: Call the set of shielding state switching nodes, and according to the trigger node number, sequentially search the corresponding module number in the frequency selection filter signal chain structure, locate the adjacent upstream and downstream filter unit module indexes, map the three into a structure pair by number, obtain their corresponding relative sequence positions in the signal channel, and generate the filter module combination index value. The system retrieves the trigger node number recorded in the shielding state switching node set and establishes a number index in the frequency-selective filter structure diagram configured in the filtering system. This number is then used to sequentially locate the filter module connected to each node. The system's filter structure employs a multi-level cascaded configuration, with each node corresponding to a primary module. The system then searches the node number mapping table to find the number of the main filter module to which the node belongs. And extract the upstream module numbers in this order. Downstream module number The three are combined in structural order to form This is used to identify its positional relationship in the signal path and assign a unique mapping number to the structure. For example, if node number 12 is, then the corresponding main module is... If the structural chain is linear, then the combination can be obtained as follows: The system has a 48-channel structure, with modules numbered consecutively; therefore, the total number of combined structures is [number missing]. The actual sequence position of the combined structure in the signal channel is recorded using a numbered index table, forming a structure identification index vector, which ultimately generates the combined index value of the filtering module.

[0027] S312: Based on the filter module combination index value, call the micrometer node parameters at the frequency crossing point in each combination structure, collect the equivalent inductance value, equivalent capacitance value, and reflection coefficient frequency domain response data, insert the equivalent parameter values ​​into the complex impedance frequency response curve, extract the complex impedance slope change points in the frequency band based on the derivative change behavior of the response curve, and calculate the deviation between the maximum slope value of the first derivative of the reflection coefficient and the set coupling abrupt change threshold, using the formula: ; The calculation obtains the coupling strength values ​​in the combined structure and generates a set of coupling response strengths. in, For the first The coupling strength value of the group filter combination structure. For the first Normalized reflection coefficient of the group structure at the frequency crossing point. For the first The rate of change of the group reflection coefficient with respect to the normalized frequency; the first derivative reflects the rate of change of signal reflection with frequency. For the first Complex impedance value at each frequency point For the first The rate of change of the complex impedance with respect to the normalized frequency at each frequency sampling point, and the first derivative represents the local slope of the complex impedance curve at that point. The normalized frequency value represents the relative position of the sampling frequency point within the total area of ​​the crossing region. For the first Normalized width of the frequency crossover region, This is the index number of the frequency sampling point. This represents the total number of sampling points in the frequency crossing region. The total number of combined structural samples participating in the fitting. This is the index of the filter combination structure; Based on the filter module combination index value, the frequency crossing point of the central module in each combination structure is called, and the parameters of the micro-measurement node embedded at that frequency point are read, including the equivalent inductance. Equivalent capacitance With reflection coefficient response curve Obtain frequency range Discrete frequency points within The corresponding complex impedance The sequence is analyzed, and its numerical derivative is calculated. The point of maximum derivative change is extracted, reflecting the location of the abrupt change. Then, the difference between the first derivative value and the mean derivative value between frequency points is calculated. The absolute values ​​of the derivatives at all points are summed and multiplied by the rate of change of the reflection coefficient, and then substituted into the following formula: ; in, For the first The reflection coefficient of the group structure is normalized to its first derivative. For the first The sum of the rates of change of complex impedance at all frequency points within the crossing region. This represents the normalized value of the frequency crossing interval width. Here, represents the total number of combinatorial structures. This represents the total number of structures in this batch. An example is given below to illustrate this; if the number... The derivative values ​​in the group are as follows: , , ; Substituting into the formula, we get: ; The coupling mutation threshold is set to This value is based on the average level of the typical rate of change of reflection of the system in a non-interactive state. It is determined by a 60% increase, that is: To ensure the distinguishability of abrupt changes in transient reflection, when the calculated value is as follows... When the structure is determined to be a strongly coupled abruptly altered structure, it is included in the index. The coupling set is entered into the structure, and finally the coupling response intensity set is established.

[0028] S313: Based on the set of coupling response strengths, extract the combined structure numbers whose strength values ​​are greater than the coupling mutation threshold, match the original node and channel index, generate coupling feature identifier bits at the corresponding positions, and generate hierarchical levels based on the coupling position and parameter jump mode labels. Extract the coupling type label, channel number and structure positioning index on each signal channel to establish coupling node identification label information. Based on the structure numbers whose coupled response intensity values ​​are greater than the coupled abrupt change threshold, extract the corresponding channel numbers from the original signal channel mapping table. By combining the central module number and the structural index of its adjacent modules, the channel topology information located on the signal chain is determined. This structural point is marked as a coupling point in the channel list, and its structural mutation category is determined according to its... Value classification, set to: light coupling Coupling Strong coupling This is used to assign coupling type labels. Numbered by composite structure For example, its It is classified as a medium coupling type. and in the channel number The type is marked, and the coupling feature identifier is set to 1. The structure pair number is also recorded. Channel number ,type Finally, the coupling node identification tag information is established in the coupling node index table.

[0029] Please see Figure 5 The specific steps for obtaining the energy storage output fluctuation level identifier are as follows: S411: Based on the identification tag information of the coupling node, according to the marked coupling channel number, the corresponding number is located sequentially in the output channel of the flywheel energy storage system, the output voltage time series of the channel is collected in the continuous sampling period before and after the disturbance event, and the sampling frequency and total number of periods of the sequence are recorded to generate a voltage time series sampling dataset. Based on the coupling channel number recorded in the coupling node identification tag information, the corresponding channel number is sequentially retrieved and locked in the output channel mapping table of the flywheel energy storage system. The output module ID to which the channel node belongs is extracted, and its electrical path is connected to the voltage monitoring port. The total duration before and after coupling is determined through a set event trigger window. The voltage sequence is sampled, and the sampling frequency set for each channel is [value missing]. Corresponding to the generation per second Group voltage data, collected continuously for 3 seconds Group data, using sampling period intervals It is a cycle, and each cycle contains The sample points are divided into: For each cycle, attributes such as sampling number, timestamp, and measured voltage value are established. The cycle length, time span, and boundary positions between adjacent cycles required for spectrum analysis of the entire segment are recorded, using triplet data. A structured sequence index is established based on the sampling parameters to locate the voltage data structure of a single cycle during subsequent processing. Each channel's data is numbered. The indexes are combined into a dictionary structure, ultimately forming a voltage timing sampling dataset.

[0030] S412: Based on the voltage time-series sampling dataset, analyze the trend of voltage RMS value change in each cycle, extract the voltage RMS value, total harmonic distortion, and number of fluctuation duration cycles for each cycle, and use the initial steady-state voltage reference value to perform differential calculation and normalization transformation on the indicators, using the formula: ; The volatility intensity factor under the disturbance range is obtained through calculation, and a volatility intensity factor set is generated. in, Indicates the first The volatility intensity factor of the channel under the influence of disturbances Indicates the first The normalized value of the effective value of the periodic voltage. This represents the normalized value of the steady-state voltage reference value. Indicates the first Normalized value of periodic voltage harmonic distortion For the first Channel harmonic influence weighting factor This represents the normalized value indicating the duration of the disturbance across periods. For periodic indexes, The total number of cycles, Number the channel; Based on the voltage timing sampling dataset, each channel is numbered. Each cycle below Calculate the effective value of voltage Voltage distortion And count the number of consecutive fluctuation cycles. Normalize all periodic voltage RMS values ​​to ,in The upper limit steady-state voltage reference value set for the system. As a normalized reference, the effectiveness of voltage fluctuations is determined by... Definition: A deviation of more than 3% of the effective value is considered a disturbance; Total Harmonic Distortion (THD) After normalization ,set up It is the maximum harmonic ratio that should not be exceeded in the system; the normalized value of the disturbance duration period is... This corresponds to the proportion of effective fluctuations occurring within the current channel over the total period. Channel harmonic influence weighting factor. The main load power supply channel is assigned a value based on the group setting of the channel. Auxiliary control power supply channel assignment This setting is based on different strategies of the channel load characteristic evaluation system for harmonic fluctuation tolerance, and serves as a constant for actual configuration. It is now defined by channel number. For example, some of its periodic sampling data are shown in Table 2.

[0031] Table 2 Periodic Sampling Data Table In continuous period The CCP discovered 11 cycles deviating from the benchmark, which is recorded as ,but Substitute into the formula: ; set up Calculate the 5th period term as The sum of each item is calculated sequentially and set as follows: ,but: ; Finally, the volatility intensity factor value of channel 3 was obtained. .

[0032] S413: Call the fluctuation intensity factor set, combine it with the preset voltage fluctuation level division range, mark the fluctuation level label corresponding to each channel, and organize the channel number, fluctuation range and level sequence number to establish the energy storage output fluctuation level identifier. Call the volatility intensity factor value and number all channels. Calculated The values ​​are classified into levels, and the preset voltage fluctuation level ranges are as follows: Stability level is Slight fluctuations Moderate fluctuation is Severe fluctuations According to channel 3 The level is determined to be moderate fluctuation, level number 2, corresponding to the label "M". Channel numbers and corresponding... Numerical values, level numbers, and labels are organized into fluctuation characteristic record entries and written into an identification label table. A table structure is established with the channel number as the primary index to generate an energy storage output fluctuation level identifier. This structure can be directly used in downstream task state matching and steady-state margin allocation strategies.

[0033] Please see Figure 6 The specific steps for obtaining the anti-interference capability status label are as follows: S511: Call the energy storage output fluctuation level identifier, extract the voltage deviation upper limit, harmonic distortion upper limit and duration critical value of each level range according to the preset voltage stability level classification standard, and match each group of fluctuation levels with the classification standard by channel number, determine the level range to which it belongs, and generate voltage stability level value. The system retrieves the channel numbers and corresponding voltage fluctuation level numbers recorded in the energy storage output fluctuation level identifier. Based on the system's preset voltage stability level grading standard, it compares the actual fluctuation parameters of each channel with the level threshold. The grading standard is divided into four levels: L0, L1, L2, and L3, corresponding to "stable," "mild," "moderate," and "severe" statuses, respectively. Level determination references three fluctuation indicators, namely, the upper limit of voltage deviation. Total Harmonic Distortion Upper Limit and the upper limit of the duration of the disturbance The specific parameter settings are shown in Table 2.

[0034] Table 3. Voltage Stability Level Classification Standards As shown in Table 3, the grading standards are formulated based on the national standard GB / T15543-2008 and conventional indicators for power grid stability assessment. This is achieved by measuring the actual fluctuation indicators of each channel (such as...). , , Compare it with the table to determine if it falls within the interval. The corresponding level number is 1, and finally, a voltage stability level value is generated for each channel.

[0035] S512: Based on the voltage stability level value, compare it with the task level requirements undertaken by the flywheel energy storage unit in the current operating cycle, extract the power supply continuity tolerance boundary, the minimum voltage stability level standard and the control response time limit threshold set under the task level, perform difference calculation and sorting between the voltage level and the task standard, filter the channel number that meets the operating requirements, and generate the task operation adaptability index. Based on the voltage stability level value, a system task adaptation assessment is performed for the level number corresponding to each channel. The tasks undertaken by the flywheel energy storage system during its operating cycle are divided into three categories according to the scheduling plan: Task T1 is the main load power supply guarantee task, requiring a voltage level no lower than L1 and a tolerance for power outages of no more than 2 cycles; Task T2 is a backup support task, allowing the level to drop to L2, with a power outage period of no more than 4 cycles; Task T3 is a non-critical voltage regulation task, accepting a level of L3, but not lower than this threshold. Three standards are set for each task, namely the power supply continuity tolerance boundary. Minimum stability level requirements Response time threshold Taking channel number 5 as an example, if its voltage level is L2 and its current task is T1, the corresponding task requirements are: The difference A positive sorting order difference indicates that the current task has not been satisfied. If it's task T2, ... This indicates a positive match. All channels are matched with their corresponding tasks to form a judgment sequence, outputting the channel number, level difference, and sorting status. Channel numbers that meet the task requirements are then filtered out, generating a task performance compatibility index.

[0036] S513: Call the task running adaptability index, extract the channel identifiers of the adaptation result as positive level matching status, record the channel number, level range and allowed running status identifier value, and combine and classify the adaptation status and stability status according to the evaluation rules to generate anti-interference capability status labels. Call the task execution adaptation metric and extract the adaptation result as the positive level matching status. The system collects channel information, organizes each channel's number, corresponding stability level range, task number, and status code, and records and encodes these to construct a composite identification system. The system categorizes statuses into four types: 00 for stable-matching, 01 for minor fluctuation-matching, 10 for moderate fluctuation-mismatching, and 11 for severe fluctuation-mismatching. These statuses are combined with the level number and task adaptation status for encoding. For example, if channel 5 has a level of L1 and task T1 meets the level requirement, the status code is 00. The encoded structure is mapped to a status classification label dictionary, written into the status identification field, and anti-interference capability status labels are generated for subsequent task-level control and scheduling.

[0037] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for resisting electromagnetic interference in a flywheel energy storage system for a data center, characterized in that, Includes the following steps: S1: Obtain the output data of the flywheel energy storage unit during the operation of the power access node, determine the interval assignment of the angular velocity change rate and the four fixed angular velocity fluctuation interval thresholds, record the running period and change characteristics corresponding to the speed data segment in the disturbance interval, and generate an inertial disturbance assignment information set. S2: Based on the inertial disturbance attribution information set, combined with the power input port input impedance change status of the corresponding signal link, compare it item by item with the shielding layer surface resistivity switching threshold to determine whether the switching conditions are met, and generate a shielding state switching node set. S3: Call the set of shielding state switching nodes, detect the complex impedance frequency response curves of adjacent upstream and downstream filter units, and detect whether the maximum slope of the first derivative of the reflection coefficient exceeds the set coupling judgment threshold, obtain the channel number and coupling type label that meet the sudden change characteristics, and generate coupling node identification label information. S4: Based on the identification tag information of the coupling node, extract the output voltage time series at the corresponding channel of the flywheel energy storage output end, and calibrate the degree of interference-induced effect to generate an energy storage output fluctuation level identifier.

2. The electromagnetic interference suppression method for a flywheel energy storage system in a data center according to claim 1, characterized in that, The inertial disturbance attribution information set includes flywheel operation disturbance level label, disturbance period boundary timestamp, disturbance section identification number, speed change curve index number, and corresponding disturbance state index. The shielding state switching node set includes trigger node number, shielding layer switching state label, trigger time index, channel impedance state classification value, and shielding layer impedance switching command. The coupling node identification label information specifically includes channel coupling identifier number, slope change point number index, reflection coefficient fluctuation level, filter unit coupling type classification, and frequency crossing region slope fitting number. The energy storage output fluctuation level identifier specifically refers to steady-state deviation level, fluctuation feature number, distortion level label, voltage anomaly persistence classification, and corresponding coupling channel index identifier.

3. The electromagnetic interference suppression method for a flywheel energy storage system in a data center according to claim 2, characterized in that, The specific steps for obtaining the inertial disturbance attribution information set are as follows: S111: Acquire the output data of the flywheel energy storage unit during operation at the power access node, including instantaneous speed value, angular velocity change rate and maximum angular velocity fluctuation amplitude within three cycles, synchronously record the speed difference and sampling time difference corresponding to each moment in each cycle, calculate the angular velocity change rate sequence range corresponding to each cycle, and generate the cycle angular velocity change interval value. S112: Based on the periodic angular velocity change interval value, extract three continuous periods of equal time width from the sampling sequence, locate and extract the maximum angular velocity change rate within each period, and calculate the maximum speed difference and average change trend within the corresponding period to generate three fluctuation characteristic statistical values. S113: Based on the statistical values ​​of the three wave characteristics, according to the maximum value of the rate of change of angular velocity, the interval assignment is determined by comparing with the four angular velocity wave interval thresholds. The wave segment number and the corresponding sampling time period position are recorded as being in the disturbance interval. A corresponding operating status label is established, and an inertial disturbance assignment information set is generated.

4. The electromagnetic interference suppression method for a flywheel energy storage system in a data center according to claim 3, characterized in that, The specific steps for obtaining the set of nodes for switching the shielding state are as follows: S211: Based on the inertial disturbance attribution information set, according to the time period in the medium disturbance and high disturbance range, the disturbance identifier bit, node sampling time and corresponding rotational speed change label of each data node are collected in the time period, and the disturbance identifier state sequence is segmented and classified in chronological order to obtain the node number and trigger identifier distribution in each disturbance continuous segment, and generate the disturbance identifier sequence feature value. S212: Based on the feature value of the disturbance identification sequence, perform impedance detection on the power access port in the signal link, obtain the instantaneous input impedance value change sequence corresponding to the port, compare it with the surface resistivity switching threshold set by the shielding structure, determine whether the corresponding impedance value exceeds the switching judgment boundary, and record the number and impedance change direction of the switching condition to generate the node impedance change trend value. S213: Call the node impedance change trend value, mark each number that meets the switching conditions and the corresponding time as a valid switching point, count the synchronization of disturbance indicators and the consistency of impedance directionality at the switching point, and filter switching event nodes with complete response sequences. Calculate and obtain the switching event response value, and include the numbers whose response values ​​are greater than the set event validity threshold into the filter set. Obtain the trigger time, number index and channel status mark of each valid node, and establish a set of shielded state switching nodes.

5. The electromagnetic interference suppression method for a flywheel energy storage system in a data center according to claim 4, characterized in that, The specific steps for obtaining the identification tag information of the coupling node are as follows: S311: Call the set of shielding state switching nodes, and according to the trigger node number, sequentially search the corresponding module number in the frequency selective filter signal chain structure, locate the adjacent upstream and downstream filter unit module indexes, map the three structure pairs by number, obtain their corresponding relative sequence positions in the signal channel, and generate filter module combination index value. S312: Based on the combined index value of the filter module, call the micrometer node parameters at the frequency crossing point in each combined structure, collect the equivalent inductance value, equivalent capacitance value and reflection coefficient frequency domain response data, insert the equivalent parameter values ​​into the complex impedance frequency response curve, extract the complex impedance slope change point value under the frequency band according to the derivative change behavior of the response curve, calculate the deviation between the maximum slope value of the first derivative of the reflection coefficient and the set coupling change threshold, calculate and obtain the coupling behavior strength value in the combined structure, and generate a coupling response strength set; S313: Based on the set of coupling response strengths, extract the combined structure numbers whose strength values ​​are greater than the coupling mutation threshold, match the original node and channel index, generate coupling feature identifier bits at the corresponding positions, and generate a hierarchy based on the coupling position and parameter jump mode label. Extract the coupling type label, channel number and structure positioning index on each signal channel to establish coupling node identification label information.

6. The electromagnetic interference suppression method for a flywheel energy storage system in a data center according to claim 5, characterized in that, The specific steps for obtaining the energy storage output fluctuation level identifier are as follows: S411: Based on the identification tag information of the coupling node, according to the marked coupling channel number, the corresponding number is located sequentially in the output channel of the flywheel energy storage system, the output voltage time series of the channel is collected in the continuous sampling period before and after the disturbance event, and the sampling frequency and total number of periods of the sequence are recorded to generate a voltage time series sampling dataset. S412: Based on the voltage time-series sampling dataset, analyze the trend of voltage effective value change in each cycle, extract the voltage effective value, total harmonic distortion and number of fluctuation duration cycles in each cycle, use the initial steady-state voltage reference value to perform differential calculation and normalization transformation on the index, calculate and obtain the fluctuation intensity factor under the disturbance range, and generate a fluctuation intensity factor set. S413: Call the set of fluctuation intensity factors, combine it with the preset voltage fluctuation level division interval, mark the fluctuation level label corresponding to each channel, and organize the channel number, fluctuation interval and level sequence number to establish the energy storage output fluctuation level identifier.

7. The electromagnetic interference suppression method for a flywheel energy storage system in a data center according to claim 6, characterized in that, The method further includes the following steps: S5: Call the energy storage output fluctuation level identifier, assign the anti-interference capability status label of the current flywheel energy storage system under the measured interference scenario according to the set voltage stability level classification standard, and evaluate whether the current system status has the conditions to continue to operate by combining the current operation task level matching relationship, and generate the anti-interference capability status label. The anti-interference capability status label includes an anti-interference capability classification flag, a task adaptation status label, a stability level matching value, a system allowed operation status instruction, and an anomaly suppression mode label.

8. The electromagnetic interference suppression method for a flywheel energy storage system in a data center according to claim 7, characterized in that, The specific steps for obtaining the anti-interference capability status tag are as follows: S511: Call the energy storage output fluctuation level identifier, extract the voltage deviation upper limit, harmonic distortion upper limit and duration critical value of each level interval according to the preset voltage stability level classification standard, and match each group of fluctuation levels with the classification standard by channel number as index to determine the level segment to which it belongs and generate voltage stability level value. S512: Based on the voltage stability level value, compare it with the task level requirements undertaken by the flywheel energy storage unit in the current operating cycle, extract the power supply continuity tolerance boundary, voltage stability minimum level standard and control response time limit threshold set under the task level, perform difference calculation and sort the voltage level and task standard, filter the channel number that meets the operating requirements, and generate task operation adaptability index. S513: Call the task operation adaptation index, extract the channel identifiers with positive level matching status, record the channel number, level range and allowed operation status identifier value, and combine and classify the adaptation status and stability status according to the evaluation rules to generate anti-interference capability status labels.

Images