A power management method for a centralized flywheel energy storage system

Abstract

The present application relates to the technical field of power management, in particular to a centralized flywheel energy storage system power management method, comprising the following steps: obtaining flywheel angular velocity and current data, analyzing energy direction and response time difference, determining power quality and grid-connected conditions, generating control instructions and power management records. In the present application, by introducing angular velocity polarity analysis, combined with current sign switching and time synchronization comparison mechanism, the response delay of the node is accurately identified, the consistency of voltage tolerance, harmonic current and start-stop logic is checked to comprehensively determine the power quality of the running node, the power supply stability is evaluated by a continuous voltage fluctuation statistical window, and the target node with grid-connected conditions is screened, the abnormal identification result and the ready state screening are combined to form a global power supply switching instruction, the charging and discharging switching identification accuracy, node response compliance, grid-connected control timeliness and power supply process stability are optimized, and the power management effect is fully improved.

Description

Technical Field

[0001] This invention relates to the field of power management technology, and in particular to a power management method for a centralized flywheel energy storage system. Background Technology

[0002] Power management technology involves the systematic management of the generation, storage, distribution, and dispatch of electrical energy in power systems. This includes ensuring the stability of power supply, controlling the operation of energy storage devices, balancing grid loads, and optimizing energy conversion processes. Common application scenarios cover grid-side energy dispatch, user-side power optimization, and coordinated operation of distributed energy systems. Effective management of the overall power supply and distribution system's operating status is achieved through dynamic control of the operating parameters of various energy storage devices such as batteries, supercapacitors, and flywheel energy storage devices. Traditional centralized flywheel energy storage systems utilize power management methods that address technical aspects such as energy charging and discharging control, power regulation, and energy flow distribution during the operation of flywheel energy storage devices in centralized power systems. This typically employs flywheel energy state estimation based on speed detection, combined with a constant voltage and frequency output strategy or a DC-side constant power regulation mechanism in the three-phase inverter circuit control to control the energy exchange process. Active and reactive power distribution is adjusted based on the flywheel energy storage device's motor operating parameters and the inverter's output waveform characteristics to meet grid connection requirements.

[0003] In the power management process of existing flywheel energy storage systems, energy state estimation mainly relies on speed detection, which lacks an in-depth identification mechanism for polarity change characteristics. This can easily lead to identification lag problems in the early stages of energy direction change. Traditional power regulation methods based on constant voltage, constant frequency, or constant power fail to fully consider the real-time response consistency of nodes and the matching of power quality. This may result in some nodes being included in the control logic even when their response is untimely or their power quality is substandard, which can cause abnormal fluctuations or system oscillations during energy allocation. This is especially true in multi-node grid-connected collaborative scenarios, which can easily lead to unstable operation risks and management strategy failures. Summary of the Invention

[0004] The purpose of this invention is to address the shortcomings of existing technologies by proposing a power management method for a centralized flywheel energy storage system.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: a power management method for a centralized flywheel energy storage system, comprising the following steps:

[0006] S1: Obtain real-time sampled values ​​of the angular velocity of each energy storage flywheel, filter time periods with opposite polarities, extract critical points for the transition between charging and discharging states, register the corresponding time points, and generate energy direction conversion time point data;

[0007] S2: Collect the RMS value sequence of node current sensors, determine the current sign change and record the occurrence time, combine the energy direction conversion time point data for time difference comparison, filter nodes that exceed the certification response time threshold, and generate a list of non-standard response nodes.

[0008] S3: Based on the non-standard response node list, read the node logic status, make a consistency judgment on the node voltage tolerance range, harmonic current limit and current start / stop status, and generate a power quality abnormal node report.

[0009] S4: Obtain the voltage transient sequence of the power supply cycle, extract the voltage fluctuation rate of the current power supply node, count the number of voltage fluctuation standard windows that meet the standard window in the continuous window during the node to be started stage, determine whether the corresponding duration exceeds the minimum grid connection waiting time of the protocol, and generate a list of grid-connected ready nodes.

[0010] S5: Based on the power quality anomaly node report and the grid-connected ready node list, select target nodes that can be started and switched for power supply, construct global power node control instructions, and generate centralized power management records.

[0011] As a further aspect of the present invention, the energy direction conversion time point data includes the critical point of angular velocity polarity conversion, the corresponding median time, and switching control registration information; the non-standard response node list includes the node number exceeding the certification response time threshold, the time point of symbol change recording, and the statistical results of current symbol change; the power quality abnormal node report includes voltage tolerance over-limit status, harmonic current limit violation, and start-stop status consistency abnormality; the grid-ready node list includes the number of power supply voltage fluctuation stability windows, duration evaluation results, and communication protocol minimum waiting time comparison conclusions; and the centralized power management record includes the target node power supply start command, power supply switching logic trigger conditions, and safety interlock response status.

[0012] As a further aspect of the present invention, the specific steps for obtaining the energy direction conversion time point data are as follows:

[0013] S111: Obtain the real-time sampled value of the angular velocity of each energy storage flywheel, detect the change in the sign of the angular velocity direction between two adjacent time points, extract the location of the sign change and the corresponding time value in the continuous time series, and obtain the time point sequence of angular velocity direction change.

[0014] S112: Based on the time point sequence of angular velocity direction change, filter the time periods between adjacent change time points where the signs of the angular velocity changes are opposite, calculate the median adjustment time point of the angular velocity direction reversal interval, and form a median adjustment time point set;

[0015] S113: Read the time points in the median adjustment time point set that meet the rated speed setting range, register them to the switching control node, record the time data corresponding to the corresponding node, and generate energy direction conversion time point data.

[0016] As a further aspect of the present invention, the specific steps for obtaining the non-standard response node list are as follows:

[0017] S211: Collect the RMS value sequence of the Hall current sensor at the acquisition node, acquire the effective current value data of each node in the continuous sampling period at fixed time intervals, extract the current sign direction corresponding to each sampling point, and determine whether the current sign changes directionally between any two adjacent sampling times. If the current direction signs are different between the two times, it is considered a direction jump. Record the time point of the jump and generate a current direction change time sequence.

[0018] S212: Based on the current direction change time series, pair the two time series according to the node number, calculate the time difference, synchronously extract the phasor data record corresponding to the node, analyze the phase angle change of each sampling point in the phasor sequence, compare the deviation with the theoretical phase angle corresponding to the energy direction conversion time point data, record the node number of all time difference values ​​exceeding the certification response time threshold, and obtain the over-limit response node number sequence.

[0019] S213: Based on the sequence of out-of-limit response node numbers, sequentially match all node lists within the current monitoring period, extract the node items corresponding to the numbers and attach an out-of-limit response identifier, and mark the response behavior of the corresponding nodes as not meeting the synchronous phasor response requirements, and establish a list of non-standard response nodes.

[0020] As a further aspect of the present invention, the specific steps for obtaining the power quality anomaly node report are as follows:

[0021] S311: Read the current control logic state parameters of each node according to the non-standard response node list, read the corresponding node current parameter data, compare with the current standard value for normalization, and traverse all nodes to form a current parameter dataset.

[0022] S312: Based on the non-standard response node list, the voltage value of each node is compared with the rated reference voltage value, the voltage deviation is calculated, and the node comprehensive consistency error value is calculated by combining the current parameter dataset to obtain the power quality anomaly assessment sequence.

[0023] S313: Based on the power quality anomaly assessment sequence, cross-compare the node status and obtain the current start / stop signal mark of each node. If the start / stop status mark is "in operation", then simultaneously determine whether the voltage tolerance status and harmonic deviation status are both compliant. If any one of them is not satisfied, record the corresponding node as abnormal and establish a power quality anomaly node report.

[0024] As a further aspect of the present invention, the specific steps for obtaining the list of grid-connected ready nodes are as follows:

[0025] S411: Obtain the voltage transient sequence within each power supply cycle, extract the voltage effective value sequence of the current power supply node, compare it with the set voltage fluctuation benchmark value range, count the sampling segments that meet the standard in the continuous power supply time window, and obtain the compliance window statistics.

[0026] S412: Based on the compliance window statistics, compare with the minimum grid connection waiting time requirement, judge the fluctuation status in the continuous time window corresponding to each power supply node, and determine whether it always falls within the set offset upper limit range. If the condition is met, mark the current power supply cycle of the corresponding node as a stable state cycle and generate a voltage fluctuation compliance identifier set.

[0027] S413: Based on the voltage fluctuation compliance identifier set, filter the node numbers that meet the compliance identifier and are in the ready-to-start stage, and according to the corresponding control status parameters, detect whether there is an identifier that is configured as a standby grid-connected node. If there is, record it as a candidate number set to obtain a list of grid-connected ready nodes.

[0028] As a further aspect of the present invention, the specific steps for obtaining the centralized power management record are as follows:

[0029] S511: Based on the power quality abnormal node report and the grid-connected ready node list, filter the target nodes with switching control authority, exclude all nodes marked in the abnormal report, read the start / stop status of each node in the set in the latest power supply cycle, and generate a power supply node switching candidate set.

[0030] S512: Based on the power supply node switching candidate set, obtain the safety interlock information in the configuration corresponding to each candidate node, check the linkage protection status, fault blocking status, grid connection delay status, signal locking status and trigger flag status of each node. If all status items are released, it is determined to be a power supply node, and a controllable interlock power supply node set is obtained.

[0031] S513: Based on the controllable interlocked power supply node set, read the output configuration and preset grid connection code of each node in the control logic table, construct a unified control instruction sequence for driving node on / off control, write the node number, execution flag, output port number, interlock number and timestamp parameters in sequence, and write them into the instruction cache table and central control dispatch log in ascending order of node number to establish a centralized power management record.

[0032] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0033] In this invention, angular velocity polarity analysis is introduced into the monitoring of changes in the state of power to clarify the boundary of energy flow conversion. Combined with the current sign switching and time synchronization comparison mechanism, accurate identification of node response delay is achieved. Through the consistency verification of voltage tolerance, harmonic current and start-stop logic, a comprehensive judgment of the power quality of operating nodes is achieved. The power supply stability is evaluated by the continuous voltage fluctuation statistical window and target nodes with grid connection conditions are selected accordingly. The abnormal identification results and ready state screening are integrated to form a global power supply switching command. Systematic optimization is achieved in improving the accuracy of charge and discharge switching identification, strengthening the compliance of node response, enhancing the timeliness of grid connection regulation and the stability of the power supply process, and fully improving the power management effect. Attached Figure Description

[0034] Figure 1 This is a flowchart of the main steps of the present invention;

[0035] Figure 2 This is a flowchart of the energy direction conversion time point data acquisition process of the present invention;

[0036] Figure 3 This is a flowchart of the process for obtaining the non-standard response node list in this invention;

[0037] Figure 4 This is a flowchart of the power quality anomaly node report acquisition process of the present invention;

[0038] Figure 5 This is a flowchart of the process for obtaining the list of grid-ready nodes in this invention;

[0039] Figure 6 This is a flowchart of the centralized power management record acquisition process of the present invention. Detailed Implementation

[0040] 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.

[0041] 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.

[0042] Please see Figure 1 A power management method for a centralized flywheel energy storage system includes the following steps:

[0043] S1: Obtain the real-time sampling value of the angular velocity of each energy storage flywheel in the centralized flywheel energy storage system, detect the change in the sign of the angular velocity direction between two adjacent time points, filter time periods with opposite polarities, and extract the median value of the corresponding time period as the critical point for the transition between charging and discharging states (set according to 60%-70% of the rated speed of the flywheel energy storage system to meet the grid connection switching requirements of IEEE 1547 "Distributed Power Sources and Power System Interconnection Standard"), register the corresponding time point in the switching control node, and generate energy direction conversion time point data;

[0044] S2: Collect the RMS value sequence of Hall current sensors at corresponding nodes in the centralized flywheel energy storage system, determine the change of current sign in the continuous time sequence and record the occurrence time point, combine the energy direction conversion time point data to perform IEEE C37.118 synchronization phasor (power system synchronization phasor measurement standard) time difference comparison, screen nodes that exceed the UL1741SA certification response time threshold (safety certification standard for energy storage converters) and record the number, and generate a list of non-standard response nodes;

[0045] S3: Based on the list of non-standard response nodes, read the corresponding node logic status in the control logic of the centralized flywheel energy storage system, make a consistency judgment on the node voltage tolerance range (the allowable range of distribution voltage deviation specified by the voltage standard), harmonic current limit (the maximum value of harmonic current injected into the power grid by the equipment as specified in the electromagnetic compatibility standard), and current start-stop status, and generate a power quality abnormal node report.

[0046] S4: Obtain the voltage transient sequence of each power supply cycle in the centralized flywheel energy storage system, extract the voltage fluctuation rate of the current power supply node, and use the IEC62586-2 power quality monitoring algorithm (the measurement method conforming to the IEC62586 standard for power quality monitoring equipment) to count the number of windows that meet the voltage fluctuation standard (the effective voltage value changes by no more than ±10% within a 10-minute interval) within the continuous window during the waiting node stage. Determine whether the corresponding duration exceeds the minimum grid connection waiting time of the IEEE2030.5 protocol (smart grid communication protocol) and generate a list of grid-ready nodes.

[0047] S5: Based on the power quality anomaly node report and the grid-ready node list, select target nodes that can be started and switched for power supply, and combine them with the IEC61508SIL2 safety interlock (safety integrity level 2 requirement in functional safety standards) logic to construct global power node control instructions and generate centralized power management records.

[0048] The energy direction conversion timing data includes the critical point of angular velocity polarity conversion, the corresponding median time, and switching control registration information. The non-standard response node list includes the node number exceeding the certification response time threshold, the time point of symbol change record, and the statistical results of current symbol change. The power quality abnormal node report includes voltage tolerance over-limit status, harmonic current limit violation, and abnormal start-stop status consistency. The grid-ready node list includes the number of power supply voltage fluctuation stability windows, duration evaluation results, and communication protocol minimum waiting time comparison conclusions. The centralized power management record includes the target node power supply start command, power supply switching logic trigger conditions, and safety interlock response status.

[0049] Please see Figure 2 The specific steps of S1 are as follows:

[0050] S111: Obtain the real-time sampled values ​​of the angular velocity of each energy storage flywheel in the centralized flywheel energy storage system, detect the change in the sign of the angular velocity direction between two adjacent time points, extract the location of the sign change and the corresponding time value in the continuous time series, and obtain the time point sequence of angular velocity direction change.

[0051] Real-time sampling values ​​of the angular velocity of each flywheel in a centralized flywheel energy storage system are obtained. Data is continuously collected from the speed sensors in the flywheel units at each time interval to establish a one-to-one correspondence between time points and angular velocity values. For example, with a sampling period of 1 second and an initial time point of 10 seconds, after 5 consecutive sampling periods, the following data sequence can be obtained: time points are 10 seconds, 11 seconds, 12 seconds, 13 seconds, and 14 seconds, corresponding to angular velocities of 5200 rad / s, 5400 rad / s, 5600 rad / s, 5800 rad / s, and 6000 rad / s, respectively. During the detection of changes in the sign of the angular velocity direction, the difference between the angular velocity values ​​at each adjacent time point is calculated. If the difference between the current angular velocity and the previous angular velocity is greater than 0, it is considered acceleration; otherwise, it is considered deceleration. When the direction of the difference changes abruptly from positive to negative or from negative to positive, this time point is marked as the angular velocity direction change point. The corresponding calculation method is: let the angular velocity at time j be... At time j-1, ,when When a change in direction is determined, this difference multiplication judgment method is executed sequentially in each group of consecutive sampling time points to filter out the time points of change in direction. Finally, all time points that meet the above change conditions and their corresponding directions are recorded as direction change points, and a sequence set containing time points is constructed as the core data used in subsequent processing to generate the angular velocity direction change time point sequence.

[0052] Table 1 Real-time sampling data of angular velocity

[0053]

[0054] As shown in Table 1, the direction of the angular velocity difference between adjacent time points can be determined using the above data, and the moment of reversal can be obtained.

[0055] S112: Based on the time sequence of angular velocity direction changes, filter time periods with opposite signs of angular velocity changes between adjacent time points using the following formula:

[0056] ;

[0057] Calculate the midpoint of the angular velocity direction reversal interval adjustment time point This forms a set of median-adjusted time points, where... and Representing the first The start and end times of the time interval when the angular velocity direction reverses. , These represent the initial and final angular velocity values ​​for that time period, respectively. This represents the angular velocity at all sampling times within that time period. This represents the maximum angular velocity value within this segment;

[0058] Based on the time point sequence of angular velocity direction change, the start and end time points of any segment of the angular velocity reversal interval are selected. The current interval being analyzed is defined as the k-th segment, with a start time of 10s and an end time of 14s, corresponding to angular velocities of 5200 rad / s and 6000 rad / s respectively. The angular velocity sequence collected within this interval is {5200, 5400, 5600, 5800, 6000}. First, the median time of the interval is calculated by adding the start and end time points and then dividing by 2. The time is the unweighted median, and then the angular velocity correction factor is calculated. Then the final median adjustment time point was calculated. Next, determine whether the angular velocity value corresponding to that time point is within the 60% to 70% range of the system's rated speed. Assuming the rated speed of the flywheel energy storage system is 9000 rad / s, then the speed range corresponding to 60% to 70% is 5400 rad / s to 6300 rad / s. At this time, query the sampling point near 13.5996 seconds. Its angular velocity can be obtained by linear interpolation, which is between 5800 and 6000 rad / s. Use the interpolation formula... Substitute , , , , Calculated Since the value is within the range of 5400 to 6300, this time point is selected, retained, and registered as the state switching critical point. This process will be executed sequentially for multiple angular velocity direction change ranges, batch-acquire all median adjustment time points and complete effective filtering, and finally generate a set of median adjustment time points.

[0059] The median adjustment time point refers to a time point in a flywheel energy storage system that, within a time period of change in a certain angular velocity direction, is calculated by combining the median position of the start and end times of that period with a proportional correction factor formed by the amplitude of the angular velocity change. This time point is dynamically offset and adjusted. It not only reflects the time axis position of the angular velocity change process but also incorporates the influence of the intensity of the angular velocity change on the switching timing. Its essential function is to provide a more representative and responsive position than the simple median, to identify the most physically representative moment in the transition between charging and discharging states of the flywheel system, thus serving as a time reference for energy direction conversion criteria and ensuring that the system switching logic is more in line with the actual dynamics of the operating state.

[0060] The operational logic of this formula mainly revolves around the correction mechanism of the time median point based on the trend of angular velocity change. First, the formula adopts... The median time of the angular velocity reversal interval is used to ensure that the base time positioning is centered on the midpoint of the interval, and then a scaling factor is introduced. This factor, by calculating the absolute value of the difference between the initial and final angular velocities and normalizing it to the maximum angular velocity value within that interval, characterizes the relative proportion of the angular velocity change amplitude to the total dynamic range of that segment, and is expressed in a weighted form. Linear amplification of the original time median appropriately delays the switching time points corresponding to periods of drastic angular velocity change, capturing trend continuity and avoiding fluctuations caused by early switching. This multiplication process establishes an amplified coupling relationship between the angular velocity change trend and time progression, thus leading to the final expression... By converting angular velocity fluctuations into correction coefficients on a time scale, the adjustment of the switching point's reasonable offset in the time dimension is achieved. Absolute value operations ensure directional consistency, maximum value operations complete dynamic normalization, and multiplication structures provide linear enhancement of the correction magnitude. The overall logic maintains consistency from physical dimensions to actual feedback.

[0061] S113: Read the time points in the median adjustment time point set that meet the rated speed setting range (60%-70% of rated speed), register them to the switching control node, record the time data corresponding to the corresponding node, and generate energy direction conversion time point data;

[0062] Based on the median adjustment time points that fall within the range of 60% to 70% of the rated speed, the time point values ​​are sequentially registered in the switching control node of the control management module in the flywheel energy storage system. This time point is recorded as the trigger signal node for switching the energy transmission direction. The switching command time of the control program in the system is set through this recorded value. At the same time, corresponding data items are generated for the subsequent energy income and expenditure logic of the system to judge and process. All time point data recorded in the control node are centrally integrated and output to establish energy direction conversion time point data.

[0063] Please see Figure 3 The specific steps of S2 are as follows:

[0064] S211: Collect the RMS value sequence of Hall current sensors at each node in the centralized flywheel energy storage system, acquire the effective current value data of each node in the continuous sampling period at fixed time intervals, extract the current sign direction corresponding to each sampling point, and determine whether the current sign changes directionally between any two adjacent sampling times. If the current direction signs are different between the two times, it is considered a direction jump. Record the time point of the jump and generate a current direction change time series.

[0065] The RMS value sequence of Hall current sensors at each node in the centralized flywheel energy storage system is collected. Each node is set to a 1-second sampling period for synchronous recording, forming a current data sequence arranged by time within a monitoring cycle. The positive or negative value of each data point needs to be extracted, and its directionality is determined based on the current value. Sampling points with current values ​​greater than 0 are marked as positive, those less than 0 are marked as negative, and those equal to 0 are marked as zero-current nodes. The current direction labels of two adjacent sampling time points are compared sequentially. If the direction of the previous time point is positive and the direction of the next time point is negative, or the previous time point is negative and the next time point is positive, then this time point is recorded as a direction change time point. Subsequently, all recorded direction change time points are organized into a change time sequence as a time sequence feature of current response jump behavior. In actual operation, if the current sequence obtained is as shown in Table 6, then the direction change points at the 2nd and 5th seconds can be identified, because the 1st to 2nd second is from forward to reverse, and the 4th to 5th second is from negative to forward. Such time points constitute current direction change events.

[0066] Table 2 Hall Current RMS Sampling Table

[0067]

[0068] As shown in Table 2, the RMS current change process within a continuous 6-second cycle is recorded. By judging the positive and negative values ​​of adjacent data, the 2nd and 5th seconds can be identified as the current direction change points, thereby establishing a time series of current direction change.

[0069] S212: Based on the current direction change time series, pair the two time series according to the node number, calculate the time difference, synchronously extract the phasor data record corresponding to the node, analyze the phase angle change of each sampling point in the phasor sequence, compare the deviation with the theoretical phase angle corresponding to the energy direction conversion time point data, record the node number of all time difference values ​​exceeding the certification response time threshold, and obtain the over-limit response node number sequence.

[0070] Based on the time series of current direction changes and the time point data of energy direction transitions, for each recorded node number, all time point records of the corresponding node in the two data series are extracted. The time pairs in any two sequences are then calculated for difference and arranged item by item by node. The time difference of all nodes is used as the criterion for determining the delay response. If a node has multiple transition times, the corresponding record pair can be selected by calculating the minimum time difference. Then, the phase angle sequence values ​​are extracted from the data stream of the node phasor measurement system within the same time period, and matched with the reference phase angle related to the energy direction switching time. If... If any response time difference exceeds the authentication response time threshold, the node number is recorded as an out-of-limit node. For example, if a node's energy direction switching is recorded at 2.000 seconds and its current jump is recorded at 2.072 seconds, the response time difference is 72 milliseconds. If the system's response time threshold is set to 60 milliseconds, then this node should be marked as an abnormal response node. Similarly, if a node's phasor sampling angle offset reaches 6.2 degrees, exceeding the 5-degree error limit, it is also classified as abnormal. This process iterates through all nodes, filters all node numbers that meet any out-of-limit condition, and generates an out-of-limit response node number sequence.

[0071] S213: Based on the sequence of out-of-limit response node numbers, match all nodes in the current monitoring period in sequence, extract the node items corresponding to the numbers and attach an out-of-limit response identifier. The response behavior of the corresponding node is marked as not meeting the synchronous phasor response requirements, and a list of non-standard response nodes is established.

[0072] Based on the sequence of out-of-limit response node numbers, all monitored nodes in the system are indexed sequentially. Nodes with identified numbers are filtered and marked as non-compliant in synchronous response. This identifier is added as an additional column in the monitoring dataset. A status label value is configured for each node, such as status 1 representing an abnormal response and status 0 representing a normal response. Then, all nodes with a status value of 1 are sorted by number and summarized to form an abnormal node list. The data structure retains fields such as node number, response status in the current period, corresponding time period number, and data record ID. This list is mounted in the system's main control module for data updates or traceability marking in subsequent operating cycles. This process completes the standard judgment and summary of system node response behavior and establishes a non-standard response node list.

[0073] Please see Figure 4 The specific steps of S3 are as follows:

[0074] S311: Read the current control logic state parameters of each node according to the non-standard response node list, read the corresponding node current parameter data, compare with the current standard value for normalization, and traverse all nodes to form a current parameter dataset.

[0075] Based on the non-standard response node list, the start / stop status, measured current value, and current limit of each target node in the current control cycle are extracted one by one. Using the node number as an index, an initial mapping relationship between node status and electrical parameters is constructed. Example data from nodes 1 to 3 are extracted, recording their measured currents as 6.3A, 5.8A, and 6.1A, respectively, with a corresponding limit of 6.0A. The offsets are calculated as 0.3A, 0.2A, and 0.1A respectively using absolute differences. Furthermore, each offset is divided by its corresponding limit to obtain normalized offset rates of 0.05, 0.0333, and 0.0167, respectively. The mean normalized current offset rate of the node cluster is obtained using the following formula:

[0076] ;

[0077] The results are expressed as normalized current offset mean, providing a basis for the next step of composite judgment. The initial electrical parameter acquisition relationship of the nodes is shown in the table below, where "node number" is 1 to 3, and the measured current (Hall current value) is 3. (A) and the standard value of current (A) column is used for subsequent normalization processing.

[0078] Table 3 Initial Electrical Parameter Acquisition Data

[0079]

[0080] S312: Based on the non-standard response node list, compare the voltage value of each node with the rated reference voltage value, calculate the voltage deviation, and combine it with the current parameter dataset, using the formula:

[0081] ;

[0082] Compute node integrated consistency error value The power quality anomaly assessment sequence was obtained, where, Indicates the first The Hall current value at each sampling point, in amperes (A). For the first The standard current value at each sampling point, in amperes (A). For the first The actual voltage value of each node, in volts (V). For the first The voltage reference value for each node, in volts (V). This indicates the total number of current sampling points in the current cycle. This indicates the number of voltage nodes in the current cycle;

[0083] Further statistical analysis of the measured node voltage values ​​and the reference voltage value revealed that the measured voltages of nodes 1 to 3 (229.5V, 226.0V, and 231.2V) were subtracted from the reference value of 220.0V, and the absolute values ​​were taken to obtain offsets of 9.5V, 6.0V, and 11.2V, respectively. These offsets were then divided by the reference value of 220V to obtain normalized voltage offset rates of 0.0432, 0.0273, and 0.0510, respectively. The average voltage offset of the node cluster was calculated using the mean expression.

[0084] ;

[0085] Based on the current data obtained above, the normalized squared current term for each node is obtained, i.e.:

[0086] ;

[0087] ;

[0088] ;

[0089] Taking the square root of the mean of the three values, we get:

[0090] ;

[0091] Finally, the three normalized indicators were combined using a weighted summation formula:

[0092] ;

[0093] The result indicates that the node overall consistency error is 1.0861. If the judgment threshold is set to 1.000, it means that the current state should be included in the power quality anomaly assessment sequence.

[0094] The node overall consistency error value is a composite index used to measure the overall consistency between the current and voltage operating states of each node in a centralized flywheel energy storage system and the standard operating state. It integrates three factors—normalized current offset rate, normalized current fluctuation intensity, and normalized voltage offset rate—to reflect the deviation intensity of each node in the power quality dimension. Specifically, the current offset term indicates whether the node's load current exceeds the limit, the current fluctuation term indicates whether there are drastic fluctuations in the load current, and the voltage offset term reveals whether the supply voltage exceeds the standard tolerance range. When this error value increases, it indicates that the corresponding node has increasingly serious inconsistencies or abnormalities in current or voltage; when the value is close to zero, it indicates that the node's operating state is more stable and closer to the standard state. Therefore, this index can serve as a key quantitative basis for power quality consistency analysis, supporting the identification, screening, and reporting of abnormal nodes.

[0095] The formula's operational logic lies in unifying and weighting different types of offset characteristic indicators to form a composite index for comprehensively judging the consistency of node electrical parameters. First, the normalized current offset rate reflects the degree of deviation between the node current and its limit; a larger value indicates a more unstable node load. Second, the square root of the normalized square mean of the current measures the overall intensity of current fluctuations. By squaring and then taking the square root, it can more sensitively capture abnormal points with large offset amplitudes, amplifying extreme changes. The voltage offset rate reflects whether the voltage is within the standard tolerance range, and its normalized value has the same dimensions as the previous two. All three are dimensionless values ​​and can be directly superimposed to form a unified error evaluation index. The formula structure uses linear addition to integrate the individual offset results. Through this weighted fusion mechanism, current offset, fluctuation intensity, and voltage anomalies work together in a unified index system, thus constructing a composite error value that can be used for consistency judgment. This additive structure facilitates subsequent threshold setting for judgment and is easy to implement in engineering.

[0096] S313: Based on the power quality anomaly assessment sequence, cross-compare the node status and obtain the current start / stop signal mark of each node. If the start / stop status mark is "running", then simultaneously determine whether the voltage tolerance status and harmonic deviation status are both compliant. If any one of them is not satisfied, record the corresponding node as an anomaly and establish a power quality anomaly node report.

[0097] After completing the quantitative assessment of current offset, amplitude fluctuation, and voltage offset, and combining the node start-up and shutdown status data, a consistency judgment is made by comparing the node electrical parameter offset status. If a node is currently in the "start-up" state but the overall consistency error exceeds the set value of 1.000, the node is marked as an inconsistent response node, and a power quality abnormal node report is generated. For example, node 1 is in the "start-up" state, and all its offset indicators have reached the highest level, so it is included in the abnormal report. Although nodes 2 and 3 have offsets, they have not exceeded the limit in the overall error measurement and are not included in the abnormal node sequence. The final summary of node status is as follows:

[0098] Table 4 Node Status Table

[0099]

[0100] This table clearly presents a comparison between the node response offset and the operating status, providing an accurate basis for fault trend analysis.

[0101] Please see Figure 5 The specific steps of S4 are as follows:

[0102] S411: Obtain the voltage transient sequence in each power supply cycle of the centralized flywheel energy storage system, extract the voltage effective value sequence of the current power supply node, compare it with the set voltage fluctuation benchmark value range, count the sampling segments that meet the standard in the continuous power supply time window, and obtain the compliance window statistics.

[0103] Based on the voltage transient sequence of the current power supply cycle in the centralized flywheel energy storage system, a sample sequence of effective voltage values ​​is collected every 10 minutes to obtain the time-series data column corresponding to the node. Any two adjacent voltage values ​​are extracted from the sample sequence, their difference is calculated, and the absolute value is compared. 220V is set as the reference voltage for the node. According to the ±10% offset range corresponding to the IEC62586 voltage fluctuation standard, the allowable upper limit is calculated to be 242V and the lower limit to be 198V. Taking node number N1 as an example, its corresponding voltage values ​​are 230V and 235V. The differences between 240V and 243V and 220V are calculated to be 10V, 15V, 20V, and 23V, respectively, with the difference percentages being 4.55%, 6.82%, 9.09%, and 10.45%. The first three are all less than 10%, while the last one is greater than 10%. Therefore, only the first three are considered compliant range data and are regarded as effective voltage windows. By counting the number of 10-minute windows formed by the compliant voltage value sequence of each node in each cycle, this method is used to iterate through all the node numbers to be started in the system and form statistical records, finally obtaining the compliance window statistics.

[0104] S412: Based on the compliance window statistics, compare with the minimum grid connection waiting time requirement, judge the fluctuation status in the continuous time window corresponding to each power supply node, and determine whether it always falls within the set offset upper limit range. If the condition is met, mark the current power supply cycle of the corresponding node as a stable state cycle and generate a voltage fluctuation compliance identifier set.

[0105] Based on the obtained compliance window statistics, and using the 30-minute grid connection preparation time specified in the IEEE 2030.5 protocol as the criterion, it is equivalent to three consecutive compliance windows. For each power supply node number, the number of compliance tags within a 10-minute unit time window is counted. The historical fluctuation window records of nodes N1 to N3 are read and judged. If the effective voltage values ​​of node N1 within the sampling period are all 231V, 232V, and 234V, and the continuous fluctuation differences are 1V and 2V respectively, then the node is judged to meet the continuous window stability condition. If the fluctuation values ​​of node N2 are 236V, 242V, and 239V, and the differences are 6V and -3V respectively, then the node is judged to meet the continuous window stability condition in the first to second periods. If the fluctuation range of a window exceeds the standard, it cannot form a continuous window compliance chain. If the fluctuations of node N3 are 228V, 229V, and 231V respectively, with differences of 1V and 2V, the condition is also met. Then, nodes N1 and N3 are recorded as compliant nodes. At the same time, the fluctuation rate and voltage offset rate between their adjacent windows are counted to see if they are less than the set threshold values. The fluctuation rate threshold is set to 5V / 10min, and the offset rate threshold is set to 10%. The setting is based on the recommended limit range of IEC62586 and refers to the test report of national standard GB / T12325. The maximum fluctuation rate and offset rate of the nodes obtained are both less than the set thresholds. Finally, the voltage fluctuation compliance mark set is obtained.

[0106] S413: Based on the voltage fluctuation compliance identifier set, filter the node numbers that meet the compliance identifier and are in the standby stage. According to the corresponding control status parameters, check whether there is an identifier that is configured as a standby grid-connected node. If there is, record it as a candidate number set to obtain a list of grid-connected ready nodes.

[0107] Based on the obtained voltage fluctuation compliance identifier set, the system database is queried for the set of nodes marked "pending startup". The node numbers with grid connection candidate identifier fields are extracted. Subsequent screening is performed on nodes numbered N1 and N3. The difference between their effective voltage value in the most recent sampling period and the voltage reference center value set by the system is compared. The reference value is set to 220V. The voltage value of node N1 is 231V, with a difference of 11V, accounting for 5%; the voltage value of node N3 is 229V, with a difference of 9V, accounting for 4.1%. Only N3 meets the screening requirement of less than 5%. Finally, nodes with a voltage deviation rate of less than 5% are selected from the candidate node set, and their control status markers and related numbers are recorded to obtain the grid-connected ready node list.

[0108] Table 5 Node Voltage Deviation Screening Data Table

[0109]

[0110] As shown in Table 5, node N3 is the only compliant node that meets all grid connection screening conditions and is ultimately included in the list of grid-ready nodes.

[0111] Please see Figure 6 The specific steps of S5 are as follows:

[0112] S511: Based on the power quality anomaly node report and the grid-connected ready node list, filter the target nodes with switching control authority, exclude all nodes marked in the anomaly report, read the start / stop status of each node in the set in the latest power supply cycle, and generate a power supply node switching candidate set.

[0113] Based on the power quality anomaly node report and the list of grid-connected ready nodes, the following steps are taken: First, extract the sets of anomaly node numbers and ready node numbers. For example, the set of anomaly nodes could be {N2, N3, N8}, and the set of grid-connected ready nodes could be {N1, N3, N5, N7, N9}. Then, through a set interpolation operation, the anomaly numbers are removed from the ready set, resulting in an initial set of available nodes {N1, N5, N7, N9}. Next, for each node in this set, the current cycle start / stop flag parameters recorded in the status management module are called, the results are read, and it is determined whether the node is in a "stopped" state. For example, the read results might be: N1 = Running, N5 = Stopped, N7 ​​= Stopped, N9 = Stopped. =If running, the running node numbers are further eliminated, and N5 and N7 in the stopped state are retained. Then, the node status confirmation process is entered. In this process, the power failure confirmation flag and restart permission bit status fields of nodes N5 and N7 are read again. If and only if both fields are in the "allowed" state, the node is marked as a candidate target that can be started. Assuming that the power failure confirmation flag of N5 is "allowed" and the restart permission bit is "prohibited", and both fields of N7 are "allowed", then only N7 will pass the verification and be marked as a candidate target node. Finally, the node numbers that can be used for subsequent switching control operations are formed into a set. This set is the power supply node switching candidate set, as shown in Table 6.

[0114] Table 6 Preliminary Screening Record of Power Supply Nodes

[0115]

[0116] As shown in Table 6, only node N7 ultimately meets the candidate criteria.

[0117] S512: Based on the power supply node switching candidate set, obtain the safety interlock information in the corresponding configuration of each candidate node, check the linkage protection status, fault blocking status, grid connection delay status, signal lock status and trigger flag status of each node. If all status items are released, it is determined to be a power supply node, and a controllable interlock power supply node set is obtained.

[0118] Based on the set of numbers extracted from the candidate power supply node switching set, such as {N7}, the safety interlock parameters set in the configuration file of each node are read sequentially. These parameters mainly include five control fields: linkage protection status, fault blocking status, grid connection delay status, signal lockout status, and trigger flag status. For each status field, the current status value is read and matched against the set release status. If any status value is not equal to "release", the node is determined to be currently interlocked and does not have the ability to execute control commands. If all fields are in the "release" state, the node passes the safety check and is recorded as eligible. Taking node N7 as an example, the interlock configuration item reading results of the controllable interlock power supply node set are as follows: linkage protection status = released, fault blocking status = released, grid connection delay status = released, signal lock status = released, trigger flag status = released. All field values ​​match "release", which meets the requirements and is recorded as the current switchable target node. If node N5 participates in the judgment, its grid connection delay status is "delay not reached" or its signal lock status is "locked", then it will not be included in the current set. Finally, only the node number with all interlock field statuses in compliance is written into the controllable interlock power supply node set, as shown in Table 7.

[0119] Table 7 Node Interlock Status Assessment Record Table

[0120]

[0121] Referring to Table 7, only N7 ultimately meets the interlock release requirements and is included in the controllable interlock power supply node set.

[0122] S513: Based on the controllable interlocked power supply node set, read the output configuration and preset grid connection code of each node in the control logic table, construct a unified control instruction sequence for driving node on / off control, write the node number, execution flag, output port number, interlock number and timestamp parameters in sequence, and write them into the instruction cache table and central control dispatch log in ascending order of node number to establish a centralized power management record.

[0123] After identifying the interlock release node, based on the node number in the controllable interlock power supply node set, the current system timestamp record field in the system main controller and the central control command status field in the bus scheduling management module are called sequentially. The result values ​​are read to form a structured input source, where the timestamp field is in datetime format and the main control status field indicates whether each control path is active in bit form. Then, the node control parameter table is called to extract the preset output port number, execution status flag bit, and corresponding control flag code corresponding to the target node number. The above data is then assembled into a structure according to the node number, current timestamp, and execution flag bit. The five fields—output channel number, control identifier code, and output channel number—form a control instruction data frame, which is written to the command cache table through the master command management interface. Then, the assembled control data frame is written to the log recording unit in ascending order of node number, forming a node execution control record ordered by timestamp. This control instruction frame is activated on the main controller interface to drive the synchronous triggering of the power supply switching command. Taking node N7 as an example, its control parameter field read values ​​are: output port number = OUT03, execution flag = 1, control identifier code = CMD_315, and system timestamp is 2024-04-20 17:28:43. The data frame content after assembling the five fields is [N7, 2024-04-20 17:28:43, 1, OUT03, CMD_315], constituting a complete instruction record, as shown in Table 8.

[0124] Table 8 Control Node Data Frame Record Table

[0125]

[0126] Table 8 lists the actual generated results of node control data frames, which are used for processing by the control instruction recording and scheduling module.

[0127] 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 power management method for a centralized flywheel energy storage system, characterized in that, Includes the following steps: S1: Obtain real-time sampled values ​​of the angular velocity of each energy storage flywheel, filter time periods with opposite polarities, extract critical points for the transition between charging and discharging states, register the corresponding time points, and generate energy direction conversion time point data; S2: Collect the RMS value sequence of the Hall current sensor at the node, determine the change in current sign and record the time point of occurrence, compare the time difference with the energy direction conversion time point data, filter nodes that exceed the certification response time threshold, and generate a list of non-standard response nodes. S3: Based on the non-standard response node list, read the node logic status, make a consistency judgment on the node voltage tolerance range, harmonic current limit and current start / stop status, and generate a power quality abnormal node report. S4: Obtain the voltage transient sequence of the power supply cycle, extract the voltage fluctuation rate of the current power supply node, count the number of voltage fluctuation standard windows that meet the standard window in the continuous window during the node to be started stage, determine whether the corresponding duration exceeds the minimum grid connection waiting time of the protocol, and generate a list of grid-connected ready nodes. S5: Based on the power quality anomaly node report and the grid-connected ready node list, select target nodes that can be started and switched for power supply, construct global power node control instructions, and generate centralized power management records.

2. The power management method for a centralized flywheel energy storage system according to claim 1, characterized in that, The energy direction conversion timing data includes the critical point of angular velocity polarity conversion, the corresponding median time, and switching control registration information. The non-standard response node list includes the node number exceeding the certification response time threshold, the time point of symbol change recording, and the statistical results of current symbol change. The power quality abnormal node report includes voltage tolerance over-limit status, harmonic current limit violation, and abnormal start-stop status consistency. The grid-ready node list includes the number of power supply voltage fluctuation stability windows, duration evaluation results, and communication protocol minimum waiting time comparison conclusions. The centralized power management record includes the target node power supply start command, power supply switching logic trigger conditions, and safety interlock response status.

3. The power management method for a centralized flywheel energy storage system according to claim 1, characterized in that, The specific steps for obtaining the energy direction conversion time point data are as follows: S111: Obtain the real-time sampled value of the angular velocity of each energy storage flywheel, detect the change in the sign of the angular velocity direction between two adjacent time points, extract the location of the sign change and the corresponding time value in the continuous time series, and obtain the time point sequence of angular velocity direction change. S112: Based on the time point sequence of angular velocity direction change, filter the time periods between adjacent change time points where the signs of the angular velocity changes are opposite, calculate the median adjustment time point of the angular velocity direction reversal interval, and form a median adjustment time point set; S113: Read the time points in the median adjustment time point set that meet the rated speed setting range, register them to the switching control node, record the time data corresponding to the corresponding node, and generate energy direction conversion time point data.

4. The power management method for a centralized flywheel energy storage system according to claim 1, characterized in that, The specific steps for obtaining the list of non-standard response nodes are as follows: S211: Collect the RMS value sequence of the Hall current sensor at the acquisition node, acquire the effective current value data of each node in the continuous sampling period at fixed time intervals, extract the current sign direction corresponding to each sampling point, and determine whether the current sign changes directionally between any two adjacent sampling times. If the current direction signs are different between the two times, it is considered a direction jump. Record the time point of the jump and generate a current direction change time sequence. S212: Based on the current direction change time series, pair the two time series according to the node number, calculate the time difference, synchronously extract the phasor data record corresponding to the node, analyze the phase angle change of each sampling point in the phasor sequence, compare the deviation with the theoretical phase angle corresponding to the energy direction conversion time point data, record the node number of all time difference values ​​exceeding the certification response time threshold, and obtain the over-limit response node number sequence. S213: Based on the sequence of out-of-limit response node numbers, sequentially match all node lists within the current monitoring period, extract the node items corresponding to the numbers and attach an out-of-limit response identifier, and mark the response behavior of the corresponding nodes as not meeting the synchronous phasor response requirements, and establish a list of non-standard response nodes.

5. The power management method for a centralized flywheel energy storage system according to claim 1, characterized in that, The specific steps for obtaining the power quality anomaly node report are as follows: S311: Read the current control logic state parameters of each node according to the non-standard response node list, read the corresponding node current parameter data, compare with the current standard value for normalization, and traverse all nodes to form a current parameter dataset. S312: Based on the non-standard response node list, the voltage value of each node is compared with the rated reference voltage value, the voltage deviation is calculated, and the node comprehensive consistency error value is calculated by combining the current parameter dataset to obtain the power quality anomaly assessment sequence. S313: Based on the power quality anomaly assessment sequence, cross-compare the node status and obtain the current start / stop signal mark of each node. If the start / stop status mark is "in operation", then simultaneously determine whether the voltage tolerance status and harmonic deviation status are both compliant. If any one of them is not satisfied, record the corresponding node as abnormal and establish a power quality anomaly node report.

6. The power management method for a centralized flywheel energy storage system according to claim 1, characterized in that, The specific steps for obtaining the list of grid-ready nodes are as follows: S411: Obtain the voltage transient sequence within each power supply cycle, extract the voltage effective value sequence of the current power supply node, compare it with the set voltage fluctuation benchmark value range, count the sampling segments that meet the standard in the continuous power supply time window, and obtain the compliance window statistics. S412: Based on the compliance window statistics, compare with the minimum grid connection waiting time requirement, judge the fluctuation status in the continuous time window corresponding to each power supply node, and determine whether it always falls within the set offset upper limit range. If the condition is met, mark the current power supply cycle of the corresponding node as a stable state cycle and generate a voltage fluctuation compliance identifier set. S413: Based on the voltage fluctuation compliance identifier set, filter the node numbers that meet the compliance identifier and are in the ready-to-start stage, and according to the corresponding control status parameters, detect whether there is an identifier that is configured as a standby grid-connected node. If there is, record it as a candidate number set to obtain a list of grid-connected ready nodes.

7. The power management method for a centralized flywheel energy storage system according to claim 1, characterized in that, The specific steps for obtaining the centralized power management record are as follows: S511: Based on the power quality abnormal node report and the grid-connected ready node list, filter the target nodes with switching control authority, exclude all nodes marked in the abnormal report, read the start / stop status of each node in the set in the latest power supply cycle, and generate a power supply node switching candidate set. S512: Based on the power supply node switching candidate set, obtain the safety interlock information in the configuration corresponding to each candidate node, check the linkage protection status, fault blocking status, grid connection delay status, signal locking status and trigger flag status of each node. If all status items are released, it is determined to be a power supply node, and a controllable interlock power supply node set is obtained. S513: Based on the controllable interlocked power supply node set, read the output configuration and preset grid connection code of each node in the control logic table, construct a unified control instruction sequence for driving node on / off control, write the node number, execution flag, output port number, interlock number and timestamp parameters in sequence, and write them into the instruction cache table and central control dispatch log in ascending order of node number to establish a centralized power management record.

Images