A power intelligent regulation system based on a centralized flywheel energy storage system

Abstract

This invention relates to the field of power regulation technology, specifically to an intelligent power regulation system based on a centralized flywheel energy storage system. The system includes a state perception module, a data partitioning module, a main control and allocation module, a power switching module, and a stability judgment module. This invention enhances the sensitivity to voltage fluctuations by continuously judging the trend of the first derivative of the flywheel energy storage unit voltage. It improves the partitioning accuracy of the flywheel output state space by combining two-dimensional normalized analysis of the rate of change of rotational kinetic energy and moment of inertia per unit time. Based on the analysis and processing of the flywheel's moment of inertia, electromagnetic coupling coefficient, and remaining kinetic energy, it enhances the accuracy of assessing the power output capacity per unit time. Furthermore, when the power deviation exceeds the limit, it executes output path switching and records the node status, achieving real-time response and bypass access to output power deviation behavior. Combined with the quantitative judgment of inertia balance and bus voltage drop deviation, it improves the stability verification capability after fault switching.

Description

Technical Field

[0001] This invention relates to the field of power regulation technology, and in particular to a power intelligent regulation system based on a centralized flywheel energy storage system. Background Technology

[0002] The field of power regulation technology involves technical aspects of regulating and managing power output in power systems. Its core components include power forecasting, power regulation strategy formulation, load coordination control, power output management, and joint regulation of energy storage systems. This technology covers the research and application of power matching regulation between conventional generation and load sides, power balancing of renewable energy grid connection, and collaborative response strategies for multiple types of energy storage units. The aim is to ensure the stability, security, and efficiency of power system operation. Within the power regulation technology system, particular attention is paid to the optimization methods and execution strategy design of multi-source power balancing mechanisms in the context of large-scale integration of highly volatile and uncertain power sources.

[0003] Among them, the power intelligent control system based on a centralized flywheel energy storage system is a system that uses a central control strategy to uniformly regulate the power of a set number of flywheel energy storage units. Its main technical challenge is to achieve stable output power control of the flywheel energy storage system under load fluctuations or power supply anomalies. Typically, control commands are generated based on real-time acquired voltage and current measurements, using static threshold judgment rules or rule functions fitted from historical data within a specific frequency band, to adjust the flywheel speed and charging / discharging state, and thereby regulate its output power. The main methods used in this process include voltage and frequency determination using a setpoint comparator, generating control curves through interpolation algorithms, and execution methods triggered by logic threshold control commands.

[0004] Traditional control systems rely on static threshold rules or historical fitting functions for specific frequency bands to generate control commands. This makes it difficult to achieve rapid response in scenarios with drastic fluctuations in flywheel output voltage or sudden changes in derivatives. Voltage determination relies on fixed-value comparators, which cannot dynamically adapt to the slope of voltage changes, resulting in power control delays or misjudgments. Furthermore, the lack of normalization or spatial partitioning of rotational inertia and kinetic energy states can easily lead to scheduling imbalances among multiple units. The lack of quantitative judgment methods for stability after power output path switching makes it easy to introduce system voltage anomalies or inertia shocks after path adjustments. Scenarios such as the failure to detect changes in bus voltage drop after the flywheel main path is disconnected will increase the risk of system instability and affect the overall safety and control accuracy of the flywheel energy storage system. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies and propose a power intelligent control system based on a centralized flywheel energy storage system.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: a power intelligent control system based on a centralized flywheel energy storage system, the system comprising:

[0007] The state perception module acquires the real-time voltage values ​​at the stator and rotor ends of the flywheel motor, calculates the first derivative of the voltage over three consecutive time slices, judges the difference between the derivative results, adjusts the sampling period, and generates sampling period adjustment information.

[0008] The data partitioning module extracts the current rotational kinetic energy per unit time and the rate of change of rotational inertia per unit time of the flywheel energy storage unit based on the sampling period adjustment information. After normalizing the two parameters, a two-dimensional coordinate matrix is ​​constructed. Grouping determination is performed on the partition to which the coordinate point belongs to obtain the output partition identifier group.

[0009] The main control and allocation module determines whether the energy output carrying capacity of the corresponding unit is higher than the set unit time power output benchmark value according to the output zone identifier group. If it is higher, the zone power target output value is adjusted to the upper limit of load allocation. If it is lower, it is allocated to the lower limit output boundary and a flywheel group target power value set is generated.

[0010] Based on the target power value set of the flywheel group, the power switching module determines whether the power deviation exceeds the output offset tolerance limit, switches the flywheel output to the redundant bypass and disconnects the original main path, and obtains the flywheel switching path record information.

[0011] The present invention is improved in that the sampling period adjustment information includes voltage change stability indication, voltage derivative trend identification result and voltage sampling rhythm parameter; the output partition identification group specifically includes flywheel operation partition number, power release distribution label and energy storage unit response attribute classification; the flywheel group target power value set includes unit-level energy carrying capacity scheduling parameters, target output power range setting value and partition scheduling weight allocation identifier; and the flywheel switching path record information specifically refers to the path switching node sequence, switching operation effective time and power transmission path mapping relationship.

[0012] The present invention is improved in that the state sensing module includes:

[0013] The voltage acquisition submodule acquires the real-time voltage values ​​of the stator and rotor ends of the flywheel motor, calculates the voltage change in each time period within three consecutive time slices, calls the change in each time period and the change in the previous time period to calculate the numerical difference, obtains the voltage change rate per unit time in the voltage curve, and generates the voltage change rate value.

[0014] The voltage derivative judgment submodule extracts the derivative values ​​of three time slices based on the voltage change rate value, compares the absolute differences pairwise, determines whether all three differences are lower than the set voltage slope change threshold, and determines whether there are any two sets of derivative values ​​with opposite signs and a difference greater than the voltage derivative reversal amplitude threshold, and obtains the derivative trend comparison result.

[0015] Based on the derivative trend comparison results, the period adjustment submodule determines the current derivative fluctuation state. If it is a stable change, it adds the original value of the sampling period to the value of the sampling rhythm growth unit. If it is a trend reversal, it subtracts the value of the rhythm reduction unit to obtain the value after the period rhythm adjustment and generates sampling period adjustment information.

[0016] The present invention is improved in that the data partitioning module includes:

[0017] The kinetic energy extraction submodule collects the rotational speed and moment of inertia of the flywheel energy storage unit at the current moment according to the sampling period adjustment information, calculates the rotational kinetic energy output per unit time, and generates the kinetic energy output value per unit time by combining the time length within the time period with the standardization processing.

[0018] The inertia normalization submodule calls the kinetic energy output value per unit time, and at the same time obtains the change amplitude of the flywheel energy storage unit inertia between two consecutive sampling points in the current moment, calculates the proportional coefficient between the change amplitude and the reference inertia value, and normalizes the proportional coefficient and the kinetic energy output per unit time to obtain the kinetic energy inertia mapping coordinate set value.

[0019] The partitioning judgment submodule establishes a two-dimensional coordinate graph based on the kinetic energy inertia mapping coordinate group value, determines the position of the coordinate point in the four quadrants, calculates the relative gradient distribution density of the points in the quadrant region, classifies the points according to the quadrant number and density weight label, and obtains the output partitioning identifier group.

[0020] The present invention is improved in that the main control and dispatch module includes:

[0021] The inertia extraction submodule obtains the flywheel unit number currently in the high release rate zone based on the output zone identifier group, extracts the angular velocity value and mass distribution coefficient of the corresponding unit, calculates the rotational inertia value of the unit and summarizes them to generate a flywheel inertia data set.

[0022] The energy fusion submodule calls the flywheel inertia data set, extracts the electromagnetic coupling coefficient and remaining kinetic energy value of each flywheel, converts the three data into output values ​​per unit time, and then performs combined calculations to obtain the unit-level energy output carrying capacity value.

[0023] The power distribution submodule determines whether each flywheel unit is greater than or less than the unit-time power output reference value based on the unit-level energy output carrying capacity value. If it is greater, it is marked as the upper limit of load distribution; if it is less, it is marked as the lower limit of load distribution. The target output power range of the flywheel is adjusted based on the marking results to generate a target power value set for the flywheel group.

[0024] The present invention is improved in that the power switching module includes:

[0025] The power extraction submodule extracts the flywheel unit number that needs to be switched based on the target power value set of the flywheel group, collects the current output power value of the flywheel unit and the voltage drop value on the corresponding bus side, stores and archives the collected data according to the number index, and generates a flywheel unit power parameter group.

[0026] The deviation judgment submodule calls the power parameter group of the flywheel unit, extracts the target output power value, actual output power value and bus voltage drop value of the corresponding numbered flywheel, calculates the actual power deviation value and voltage deviation amplitude value respectively, calculates the output deviation normalization intensity value, compares it with the set output deviation tolerance limit value, and if it is greater than the limit value, it is marked as a switching trigger state and generates switching judgment state mark information.

[0027] The path control submodule determines whether the flywheel unit needs to perform a switching operation based on the switching determination status flag information. If the flag is marked as triggered, it records the current switching time point and channel number, and controls the control unit output path to disconnect from the original main path and simultaneously connect to the redundant bypass channel, generating flywheel switching path record information.

[0028] The present invention has an improvement, wherein the system further includes:

[0029] The stability judgment module calls the flywheel switching path recording information, extracts the inertia balance index of each flywheel node in the bypass path after switching and the voltage offset amplitude at the bus end, judges the stability of switching, and obtains the fault switching stability judgment result.

[0030] The fault switching stability judgment results include inertia offset limit judgment results, voltage balance state level marking, and steady-state load offset label after switching.

[0031] The present invention is improved in that the stability determination module includes:

[0032] The path extraction submodule calls the flywheel switching path record information, extracts the flywheel node identifier in the bypass path after switching according to the record number and establishes a mapping relationship, connects to the downstream bus side node number in sequence, obtains complete path information, and generates a bypass path node set.

[0033] The index extraction submodule extracts the inertia balance index of each flywheel node and the voltage offset amplitude of the connected bus node based on the bypass path node set, calculates the normalized amplitude difference between the two respectively, calculates the normalized amplitude difference value, constructs the steady-state response trend vector by amplitude superposition, and generates the steady-state response difference trend quantity.

[0034] The tolerance judgment submodule compares the steady-state response difference trend with the set steady-state tolerance benchmark value item by item, identifies the node number that meets the steady-state condition, counts the proportion of the total number of nodes in the whole path, obtains the path stability index value, and generates the fault switching stability judgment result.

[0035] Compared with the prior art, the advantages and positive effects of the present invention are as follows:

[0036] In this invention, by continuously judging the changing trend of the first derivative of the flywheel energy storage unit voltage, the sampling period is dynamically adjusted, effectively enhancing the sensitivity to voltage fluctuations. Combined with the two-dimensional normalized analysis of the rotational kinetic energy and the rate of change of rotational inertia per unit time, the partitioning accuracy of the flywheel output state space is improved. Based on the analysis and processing of the flywheel rotational inertia, electromagnetic coupling coefficient, and remaining kinetic energy, the accuracy of the assessment of power output capacity per unit time is enhanced. When the power deviation exceeds the limit, the output path is switched and the node status is recorded, realizing real-time response and bypass access to the output power deviation behavior. Combined with the quantitative judgment of inertia balance and bus voltage drop deviation, the stability verification capability after fault switching is improved, realizing dynamic output allocation and stable operation guarantee under the high-frequency power fluctuation conditions of multiple units in the flywheel energy storage cluster. Attached Figure Description

[0037] Figure 1 This is a system flowchart of the present invention;

[0038] Figure 2 This is a flowchart of the state sensing module of the present invention;

[0039] Figure 3 This is a flowchart of the data partitioning module of the present invention;

[0040] Figure 4 This is a flowchart of the main control and allocation module of the present invention;

[0041] Figure 5 This is a flowchart of the power switching module of the present invention;

[0042] Figure 6 This is a flowchart of the stability determination module of the present invention. Detailed Implementation

[0043] 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.

[0044] 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.

[0045] Please see Figure 1 This invention provides a technical solution: a power intelligent control system based on a centralized flywheel energy storage system, the system comprising:

[0046] The state perception module acquires the real-time voltage values ​​at the stator and rotor ends of the flywheel motor, calculates the first derivative of the voltage over three consecutive time slices, and performs a difference judgment on the derivative results. If the absolute difference between the three sets of derivatives is lower than the voltage slope change threshold, the current sampling period is increased. If the signs of any two sets of derivatives are opposite and the amplitude difference is greater than the voltage derivative reversal amplitude threshold, the sampling period is reduced, and sampling period adjustment information is generated.

[0047] The data partitioning module adjusts the information based on the sampling period, extracts the current rotational kinetic energy per unit time and the rate of change of rotational inertia per unit time of the flywheel energy storage unit, normalizes the two parameters respectively, constructs a two-dimensional coordinate matrix, and performs grouping judgment on the partition to which the coordinate point belongs according to the quadrant division rules and differential gradient distribution density to obtain the output partition identification group.

[0048] The main control and allocation module extracts the rotational inertia, electromagnetic coupling coefficient, and remaining kinetic energy value of the flywheel unit currently in the high release rate zone based on the output zone identifier group. It performs a weighted combination calculation on the three values ​​to determine whether the energy output carrying capacity of the corresponding unit is higher than the set power output benchmark value per unit time. If it is higher, the zone power target output value is adjusted to the upper limit of load allocation; if it is lower, it is allocated to the lower limit output boundary, generating a set of target power values ​​for the flywheel group.

[0049] Based on the target power value set of the flywheel group, the power switching module extracts the flywheel unit number, the current output power value of the unit and the voltage drop value of the bus side voltage of the flywheel unit that needs to perform the path switching operation. It determines whether the power deviation exceeds the output offset tolerance limit. If it does, it records the current switching point status, switches the flywheel output to the redundant bypass and disconnects the original main path, and obtains the flywheel switching path record information.

[0050] The stability judgment module calls the flywheel switching path record information, extracts the inertia balance index of each flywheel node in the bypass path after switching and the voltage deviation amplitude at the bus end, judges the amplitude difference between the two sets of data and compares them with the set steady-state tolerance benchmark value to judge the stability of switching and obtain the fault switching stability judgment result.

[0051] Rotational kinetic energy represents the kinetic energy possessed by the flywheel due to its rotation; the rate of change of moment of inertia represents the rate of change of moment of inertia per unit time, reflecting the trend of inertial characteristics change of the flywheel during the control process; the electromagnetic coupling coefficient is defined as the quantitative value of the degree of magnetic flux coupling between the stator and rotor of the motor, usually expressed as a percentage or dimension ratio, and is widely used in the characteristic modeling of synchronous and asynchronous motors; the voltage slope change threshold and the voltage derivative reversal amplitude threshold are both settable parameters, characterizing the stability or fluctuation limits of the voltage derivative per unit time, and are used as criteria for adaptive adjustment of the sampling period; the output offset tolerance limit and the steady-state tolerance benchmark value are defined as the maximum allowable numerical deviation when the system switches between steady-state and disturbance states, often issued by the system scheduling strategy or engineering tolerance setting, and have adjustment range setting standards;

[0052] The sampling period adjustment information includes voltage change stability indication, voltage derivative trend identification results, and voltage sampling rhythm parameters. The output zone identification group specifically includes flywheel operation zone number, power release distribution label, and energy storage unit response attribute classification. The flywheel group target power value set includes unit-level energy carrying capacity scheduling parameters, target output power range setting value, and zone scheduling weight allocation identifier. The flywheel switching path record information specifically refers to the path switching node sequence, switching operation effective time, and power transmission path mapping relationship. The fault switching stability judgment results include inertia offset limit judgment results, voltage balance state level marking, and steady-state load offset label after switching.

[0053] Please see Figure 2 The state awareness module includes:

[0054] The voltage acquisition submodule acquires the real-time voltage values ​​of the stator and rotor ends of the flywheel motor, calculates the voltage change in each time period within three consecutive time slices, calls the change in each time period and the change in the previous time period to calculate the numerical difference, obtains the voltage change rate per unit time in the voltage curve, and generates the voltage change rate value.

[0055] The real-time voltage values ​​at the stator and rotor ends of the flywheel motor are acquired. For each time slice, the stator and rotor voltage values ​​are extracted, recorded, and stored to form a time series array. In the application scenario, the flywheel motor voltage can be sampled at 50ms intervals using a voltage sensor. In the example, the stator voltage values ​​are collected at time slices T1, T2, and T3 respectively. The rotor terminal voltage value is The voltage difference between each pair of samples is calculated to obtain the voltage change between the stator and rotor in each time slot. Furthermore, an adjacent difference calculation operation is performed based on the time-slice voltage change sequence to obtain the rate of voltage change per unit time, i.e., the difference between T2 and T1 is divided by the time interval (e.g., The voltage change rate during time period T2 is calculated as follows: Similarly, the voltage change rate of T3 can be calculated as follows: This allows us to obtain the trend of voltage change per unit time, as shown in Table 1, and finally generate the voltage change rate value.

[0056] Table 1. Voltage Change Data:

[0057] .

[0058] As shown in Table 1, the voltage difference shows an increasing trend over time, and the corresponding voltage change rate increases significantly in time slice T3.

[0059] The threshold for the rate of change of voltage is set to 15. This value is set according to the allowable voltage fluctuation range of the flywheel motor under maximum torque output. The maximum allowable voltage fluctuation is set to 30, and the minimum allowable time is 2. Therefore, the maximum rate of change is 30 / 2=15, that is, the threshold is set to 15, and it fluctuates with the maximum voltage fluctuation value.

[0060] The voltage derivative judgment submodule extracts the derivative values ​​of three time slices based on the voltage change rate value, compares the absolute differences pairwise, determines whether all three differences are lower than the set voltage slope change threshold, and determines whether there are any two sets of derivative values ​​with opposite signs and a difference greater than the voltage derivative reversal amplitude threshold, and obtains the derivative trend comparison result.

[0061] The derivative values ​​of the three time slices are extracted based on the voltage change rate value, and they are named respectively. And record, calculate separately The three derivative differences, with respect to the aforementioned voltage rate value example, are respectively... The calculated differences are respectively These three sets of absolute differences are compared with a set voltage slope change threshold, which is set as follows: The basis for this is the allowable stable slope range of the system voltage curve. This value is derived from the average rate standard deviation range of the voltage sampling system during the stable fluctuation period. Statistical analysis of historical operating data reveals that the average slope change is... The standard deviation is Therefore, the average value plus one standard deviation is taken as the stability threshold, resulting in... Rounding to the nearest whole number The derivative is considered to be in a stable range only when all three sets of differences are less than the threshold. If any set exceeds the threshold, it is marked as a trend reversal tendency. Further analysis is then conducted to determine if there are derivative pairs with opposite signs and differences exceeding a preset voltage derivative reversal amplitude threshold, which is set to... The setting is based on the maximum permissible derivative jump limit of the motor system. This value is calculated based on the correlation between the flywheel's mechanical inertia and the voltage response delay. When the maximum response delay is 0.1s and the maximum permissible voltage increment is 1V, the maximum value of the changing derivative can be set to... Therefore, this is used as the threshold for the inversion magnitude, such as the current derivative pair being The difference is Furthermore, the signs are opposite, exceeding the reversal threshold, thus determining a trend reversal, and finally obtaining the derivative trend comparison result.

[0062] The period adjustment submodule determines the current derivative fluctuation state based on the derivative trend comparison result. If it is a stable change, it adds the original value of the sampling period to the value of the sampling rhythm growth unit. If the trend reverses, it subtracts the value of the rhythm reduction unit to obtain the value after the period rhythm adjustment and generates the sampling period adjustment information.

[0063] Based on the derivative trend comparison result, the current voltage change derivative fluctuation state is determined. If the comparison result indicates a stable change state, the original value of the current sampling period is added to the system-set sampling rhythm growth unit value, and this value is set to... Its source is the minimum clock frequency resolution beat count within the sampling control system, with a minimum increase of 10ms within the hardware's allowable range; if the comparison result indicates a trend reversal, the original value is subtracted from the rhythm reduction unit value, and this value is set to... The setting is based on the response compensation period under voltage change reversal state. The average system response delay is 0.2s, and it is set to one-tenth of this, i.e., 20ms, as the reduction adjustment cycle. When the original sampling period is 50ms, the adjusted value is 60ms under stable change state and 30ms under trend reversal state. This adjustment value is the optimal sampling period value under the current period, and finally the sampling period adjustment information is generated.

[0064] Please see Figure 3 The data partitioning module includes:

[0065] The kinetic energy extraction submodule adjusts the information based on the sampling period, collects the rotational speed and moment of inertia of the flywheel energy storage unit at the current moment, calculates the rotational kinetic energy output per unit time, and combines the time length within the time period for standardization processing to generate the kinetic energy output value per unit time.

[0066] Obtain sampling period adjustment information, read the current rotational speed and moment of inertia value of the flywheel energy storage unit, corresponding to the energy change behavior within this sampling time point. During data extraction, call the time tag in the system sampling module, corresponding to the duration of each period, based on the recorded current sampling period. The rotational speed of the flywheel unit in the current cycle is obtained. The moment of inertia is Call the formula:

[0067] ;

[0068] Calculate the output kinetic energy per unit time at this moment, and then perform a standardization operation on this value using the formula:

[0069] ;

[0070] Substitute The calculation yields:

[0071] ;

[0072] ;

[0073] ;

[0074] The flywheel kinetic energy output value for the current cycle is obtained. The standardized kinetic energy output per unit time is It is used as an input item for the downstream inertia normalization submodule.

[0075] in, This represents the kinetic energy generated by the flywheel rotation during the current cycle, measured in joules (J). This represents the moment of inertia of the flywheel at the current moment, expressed in kilograms per square meter (kg·m²). ), This indicates the current rotational speed of the flywheel, in revolutions per minute (rpm). The sampling period is measured in seconds (s). Angular velocity, measured in radians per second (rad / s). It represents the kinetic energy output per unit time, and the unit is watts (W).

[0076] The inertia normalization submodule calls the kinetic energy output value per unit time, and at the same time obtains the change amplitude of the flywheel energy storage unit inertia between two consecutive sampling points in the current moment. It calculates the proportional coefficient between the change amplitude and the reference inertia value, and normalizes the proportional coefficient and the kinetic energy output per unit time to obtain the kinetic energy inertia mapping coordinate set value.

[0077] Call the obtained kinetic energy output value per unit time To perform the task of monitoring changes in inertia, the rotational inertia values ​​of two consecutive sampling points were collected, respectively. , The magnitude of the inertia change is calculated by the difference between the two values, and then combined with the system's reference inertia value. Normalization is performed using the following formula:

[0078] ;

[0079] Substituting the values, we get:

[0080] ;

[0081] The normalized proportional gain of the change in inertia is obtained as follows: Then, this coefficient and the kinetic energy output value per unit time are used to form a coordinate pair. This forms a set of coordinate values ​​for the kinetic energy and inertia mapping.

[0082] in, and These represent the moments of inertia at the current and previous sampling points, respectively, in kilograms per square meter (kg·m²). ), This represents the system's set reference inertia value, expressed in kilograms per square meter (kg·m²). ), It represents the normalized scaling factor (dimensionless) between the difference in inertia between two moments and the reference value.

[0083] The partitioning judgment submodule establishes a two-dimensional coordinate graph based on the kinetic energy inertia mapping coordinate group value, determines the position of the coordinate point in the four quadrants, calculates the relative gradient distribution density of the points in the quadrant region, classifies the points according to the quadrant number and density weight label, and obtains the output partitioning identifier group.

[0084] Obtain the kinetic energy inertia mapped coordinate set values ​​( After that, the task of constructing a two-dimensional coordinate graph is executed. For the horizontal axis, Establish a coordinate system for the vertical axis, map the sample data point pairs onto the coordinate graph, and determine the quadrant position of the coordinate points. In this example, Therefore, it is determined that the current point is located in the first quadrant. Then, all coordinate points in the sample set are retrieved, and the point density in each quadrant is estimated. A method of dividing and counting points by fixed unit area is used, and the density weight threshold is defined as follows: ,by The area blocks are divided into grids, the number of sample points is counted, and the density level is marked. If there are 8 points in the unit block where this point is located, the density exceeds the threshold and it is marked as a high-density block "Q1-H", forming the final output zoning identification group. Table 2 shows the classification results of multiple coordinate points.

[0085] Table 2 Output Zone Coordinate Mapping Table:

[0086] .

[0087] As shown in Table 2, each point is assigned a partition number based on its coordinate value and the quadrant in which it is located, and its segment identifier is determined by the point density per unit area. This completes the classification of all mapped points for subsequent control logic to execute judgments.

[0088] Among them, the output zone coordinate group values ​​( The quadrant number represents the ratio of energy output efficiency to inertia change of the flywheel energy storage system per unit time at a certain moment. The quadrant number reflects the synergistic characteristics of energy and inertia, and the point density indicates the degree of concentration of this characteristic state in the overall sample distribution.

[0089] Please see Figure 4 The main control and dispatch module includes:

[0090] The inertia extraction submodule obtains the flywheel unit number currently in the high release rate zone based on the output zone identifier group, extracts the angular velocity value and mass distribution coefficient of the corresponding unit, calculates the rotational inertia value of the unit and summarizes them to generate a flywheel inertia data set.

[0091] Based on the flywheel numbers of the high release rate zones recorded in the output zone identifier group, extract the numbers sequentially. The current angular velocity value of the flywheel With mass distribution coefficient ,in The mass distribution coefficients are respectively Flywheel radius value The quality is According to the standard formula for moment of inertia:

[0092] ;

[0093] The following can be calculated separately:

[0094] ;

[0095] The maximum inertia of the system is After normalization, we get:

[0096] ;

[0097] The above data forms a flywheel inertia data set, which is used for subsequent energy fusion calculations.

[0098] The energy fusion submodule calls the flywheel inertia data set, extracts the electromagnetic coupling coefficient and remaining kinetic energy value of each flywheel, converts these three data points into output quantities per unit time, and then performs combined calculations using the following formula:

[0099] ;

[0100] Calculate and obtain the unit-level energy output carrying capacity value;

[0101] in, For the first Energy output capacity of each flywheel unit For the first The normalized value of the moment of inertia of each flywheel unit is obtained by dividing the actual moment of inertia by the normalized maximum moment of inertia of the system. For the first The angular velocity of each flywheel unit For the first The electromagnetic coupling coefficient of each flywheel unit The average electromagnetic coupling coefficient of all flywheel units in the high release rate region. For the first The normalized value of the remaining kinetic energy of each flywheel unit is obtained by the ratio of the current remaining kinetic energy value to the maximum energy storage capacity of that unit. For the first The normalized value of the time length of each flywheel unit within the current sampling period is obtained by the ratio of the length of its current sampling period to the maximum sampling period of the system.

[0102] Call the number in the inertia data set The term, extracting the electromagnetic coupling coefficient With remaining kinetic energy value ,in The corresponding remaining kinetic energy is The system's maximum energy storage capacity is The normalized kinetic energy is as follows:

[0103] ;

[0104] Sampling period is After normalization, it becomes The average value of the electromagnetic coupling coefficient is calculated as follows:

[0105] ;

[0106] Substitute the above parameters into the following energy fusion formula:

[0107] ;

[0108] Its meaning and parameter explanation are as follows:

[0109] The moment of inertia of a flywheel, measured in units of Normalization to This reflects the flywheel's ability to respond to changes in kinetic energy;

[0110] Angular velocity, unit: This represents the rotational state of the flywheel;

[0111] Electromagnetic coupling coefficient, dimensionless, reflects electromechanical conversion efficiency;

[0112] Average electromagnetic coupling coefficient, used to identify the degree of deviation;

[0113] Remaining kinetic energy, in units of Normalization reflects the release capacity;

[0114] Time-normalized value, unitless, measures the sampling period;

[0115] The absolute value is used to calculate the degree of dispersion of the coupling coefficient and identify its impact on system consistency.

[0116] The square root treatment of kinetic energy makes its growth rate nonlinearly related to the kinetic energy itself, thus controlling the energy output change curve;

[0117] The logic of the overall summation term and division structure is as follows: the inertia and angular velocity terms are used as the basic kinetic energy expression of the flywheel, combined with the electromagnetic stability index (absolute value term) and energy storage availability (square root term), and then summed. The summation is then normalized by the sampling time period to obtain a dimensionless expression of the output capacity per unit time.

[0118] The advantage of the formula is that by taking into account the basic kinetic energy term, electromagnetic conversion performance and remaining energy storage value together, and by introducing a joint index of electromagnetic consistency and output stability through normalization and nonlinear processing, the accuracy and robustness of the unit energy output carrying capacity expression are improved, and the interference of a single index on the overall judgment under abnormal conditions is avoided.

[0119] The values ​​calculated using the formula are as follows:

[0120] ;

[0121] ;

[0122] ;

[0123] The power output reference value per unit time is Therefore, the output capability offset is calculated as follows:

[0124] This belongs to the upper limit of the offset;

[0125] This is a lower limit offset;

[0126] This belongs to the lower limit of the offset.

[0127] This result indicates that the flywheel unit Given its high output potential exceeding the system's baseline capability under current conditions, its target output power needs to be limited and compressed to prevent system overload, and the flywheel... When the load is in a lower output range, the target output power should be appropriately increased to achieve a dynamic and balanced distribution of the overall load. This difference directly affects the assignment of flywheel output range values ​​in subsequent power allocation strategies and is a key basis for determining the target power value of the flywheel group.

[0128] The power distribution submodule determines whether each flywheel unit is greater than or less than the power output benchmark value per unit time based on the unit-level energy output carrying capacity value. If it is greater, it is marked as the upper limit of load distribution; if it is less, it is marked as the lower limit of load distribution. The target output power range of the flywheel is assigned and adjusted based on the marking results to generate a set of target power values ​​for the flywheel group.

[0129] After calculating the output capability offset range, the system adjusts the values ​​based on the values ​​of each flywheel unit. power output reference value per unit time Based on the comparison results, perform an interval determination operation, specifically, call the already calculated interval... System default Perform a comparison operation on each unit, if If so, the corresponding unit will be marked as the load allocation upper limit zone. If it is, then it is marked as the lower limit zone of load distribution. The current judgment result is: It belongs to the upper limit of load distribution. It belongs to the lower limit zone of load distribution, and the corresponding flags are set as follows: The system then calls the aforementioned flag and associates it with the currently set flywheel power adjustment step size. The flywheel output power is adjusted by assigning values ​​in integer order, specifically for the area marked as the upper limit. Set the target output power to For those marked as the lower limit region Then set the target output power to respectively. Finally, a set of target power values ​​for the flywheel group is generated. This serves as the input target for the power regulation system in the next regulation cycle.

[0130] The flywheel target power value set directly affects the control parameters of the subsequent motor voltage excitation and serves as the execution reference for the closed-loop control logic in the system. All adjustments are calculated from this value and applied to the flywheel adjustment execution module. Therefore, its formation requires precise integration with the current unit-level offset trend and the establishment of an effective mapping mechanism with the output performance. The power adjustment step size used in this submodule... The system preset parameters are set according to the system load response sensitivity. The power change of 1500W per unit offset in the current set reference system response characteristics can be controlled by the system to respond smoothly without causing voltage oscillation. The set value is obtained by fitting the performance simulation curve and the adjustment gain model. It is suitable for the flywheel control system with a rated power of 400000W. The offset tolerance is controlled within ±5%, so it is reasonable.

[0131] The results show that by linking the output capability offset range with the target power set, the system can quickly set the target power based on the current flywheel state without relying on a complex prediction model. This enables dynamic balance correction of high / low output units within the flywheel group, meets the system rhythm control requirements, and establishes a stable initial value basis for the next cycle scheduling.

[0132] Please see Figure 5 The power switching module includes:

[0133] The power extraction submodule extracts the flywheel unit number that needs to be switched based on the target power value set of the flywheel group, collects the current output power value of the flywheel unit and the voltage drop value on the corresponding bus side, stores and archives the collected data according to the number index, and generates a flywheel unit power parameter group.

[0134] The flywheel power extraction submodule uses the target power value set of the flywheel group as an index to obtain the flywheel unit number that needs adjustment or is in a critical state through a number lookup method. The system identifies flywheel units numbered 4 and 7 as the units to be monitored, and sequentially calls their numbers to perform state synchronization acquisition, obtaining the current output power values ​​of flywheel units numbered 4 and 7 respectively. Simultaneously, the instantaneous voltage readings on the corresponding bus side were collected. The system reference voltage is Based on this, the pressure drop values ​​are calculated as follows: The system reference voltage The settings are based on the system design bus stable voltage standard, with a fixed value of 370V. The technical indicators for ±5% voltage stability tolerance are determined according to the national standard GB / T 3859.1, and the fluctuation tolerance requirements are adapted to the maximum load current environment. All acquisition results are recorded in the data archiving area within the main control node according to the number structure, forming structured power parameter items. The archived data fields include: flywheel number, current power value, bus voltage drop, and sampling timestamp. The corresponding item storage format is shown in Table 3.

[0135] Table 3 Power Sampling Data Table:

[0136] .

[0137] As shown in Table 3, the power parameter group provides the basic sampling source for the subsequent deviation judgment module. Its storage timestamp corresponds to the voltage offset, ensuring that the subsequent judgment process has traceability and indexing and positioning capabilities.

[0138] The deviation judgment submodule calls the flywheel unit power parameter group to extract the target output power value, actual output power value, and bus voltage drop value of the corresponding numbered flywheel. It then calculates the actual power deviation value and voltage offset amplitude value using the following formula:

[0139] ;

[0140] The normalized strength value of the output offset is obtained by calculation and compared with the set output offset tolerance limit. If it is greater than the limit, it is marked as a switching trigger state and a switching judgment state mark information is generated.

[0141] in, Indicates the first The normalized strength value of the output offset of each flywheel unit For the first The actual output power of each flywheel unit For the first The target output power value of each flywheel unit The normalized time length of the current sampling period. For the first Voltage drop on the bus side of each flywheel unit For the first The number of oscillations output by each flywheel unit within a short cycle For the first Output offset tolerance limit for each flywheel unit;

[0142] After the deviation judgment submodule inputs the power parameter group, it extracts the target power of the flywheel unit with the current number in sequence according to the flywheel number. Actual power Voltage drop and the number of short-cycle output fluctuations of the corresponding flywheel unit. The target power of flywheel number 4 is set as follows: The current sampling period normalized length is Its short-cycle fluctuation frequency The short cycle is defined as a period of no more than 5 seconds. If the flywheel output power deviates by more than 3%, it is counted as one fluctuation. If three fluctuations are detected in the current cycle, the corresponding setting is... Output offset tolerance limit This tolerance value The setting is based on the allowable fluctuation ratio of the system's target output reference power. If the maximum single flywheel output power of the system is 400,000W, then the tolerance is set to 3%, or 12,000W. Currently, it is operating in a fine control state, so it is compressed to 0.75%. To improve the sensitivity of switching detection. According to the formula:

[0143] ;

[0144] ;

[0145] The meanings of each parameter are explained below:

[0146] No. The real-time output power of each flywheel unit is measured by the power extraction submodule.

[0147] The target output power value is set by the power distribution module and represents the current desired stable state.

[0148] The current sampling period normalization time value is normalized based on the system's maximum sampling period, with a standard of 10s period. That is, the current 0.9 indicates that the current period is 9s in length.

[0149] The bus voltage drop is calculated from the flywheel unit output and the bus reference voltage.

[0150] Output fluctuation count indicates the number of times the flywheel experiences power fluctuations within the current short cycle. This item measures output stability and affects the system's sensitivity to small fluctuations.

[0151] For tolerance limits, please refer to the above description;

[0152] Operational logic explanation: The first term quantifies the impact of power error on sampling rhythm (multiplied by the square root of the time factor), while the second term is used to enhance the impact of voltage deviation and fluctuation stability (by multiplying the absolute value by the logarithmic function). The denominator is standardized to its respective allowable floating tolerance.

[0153] The advantage of the formula is that by embedding voltage offset and output fluctuation parameters into the normalization calculation, a new output evaluation index that simultaneously reflects power accuracy and fluctuation safety is formed, thereby strengthening the response mechanism of switching decisions.

[0154] The result indicates that the current output offset intensity of flywheel unit number 4 is 0.5106. If the system's judgment threshold is set to 0.5, this result exceeds the standard and needs to be marked as "switch trigger" and input to the path control module.

[0155] The path control submodule determines whether the flywheel unit needs to perform a switching operation based on the switching determination status flag information. If the flag is marked as triggered, it records the current switching time point and channel number, and controls the control unit output path to disconnect from the original main path and simultaneously connect to the redundant bypass channel, generating flywheel switching path record information.

[0156] The path control submodule receives a status flag value from the deviation judgment module. Upon recognizing it as a "triggered" state, it immediately executes the switching process, first recording the current system clock time. and the current flywheel number The original channel number is set to The system executes control commands to shut down the control relay of the main output path CH01 and activate the redundant bypass channel. The output relay simultaneously records the switching path information entries as follows: Number 4, Main Path CH01, Switching Path CH09, Switching Time 14:01:35, Status TRIGGERED. The switching status value is used as the basis for execution during path switching, and the switching judgment strictly relies on the information obtained from the deviation judgment module. Numerical values ​​and tolerance thresholds If the comparison results show that the former exceeds the latter, the power output stability is considered unacceptable, and the system will activate the bypass switching logic. This threshold is set. The foundation is derived from a compression scheme that allows for a 3% fluctuation in maximum stable output power, making it suitable for scenarios with concentrated loads. This path information serves as the basis for subsequent system redundancy strategy matching and switching chain tracing. It is input into the system output chain dynamic scheduling logic module to ensure that the flywheel switching logic can be continuously referenced in output stability control, possessing dynamic reusability and state reconstruction capabilities.

[0157] Please see Figure 6 The stability assessment module includes:

[0158] The path extraction submodule calls the flywheel switching path record information, extracts the flywheel node identifier in the bypass path after switching according to the record number, establishes a mapping relationship, connects to the downstream bus side node number in sequence, obtains complete path information, and generates a bypass path node set.

[0159] The path extraction submodule receives and retrieves flywheel switching path record information. By parsing the flywheel number, switching time, path number, and other information in the switching record, it retrieves the index numbers of all switched flywheel units one by one. It then identifies the redundant channel addresses of each number in the system path. Through the mapping rules between the number and the switching channel, it determines the bypass path channel number after the switch, completing the end-to-end mapping of the path. For example, if flywheels numbered 5 and 8 switch to bypass channels 09 and 11, the system establishes a mapping as 5→CH09 and 8→CH11. This is done by querying the preset bus end node binding table in the main control module. A bidirectional binding relationship is established between flywheel numbers and bus node numbers. For example, flywheel number 5 is connected to bus node MB3, and flywheel number 8 is connected to MB4, forming a path connection structure as: 5→CH09→MB3, 8→CH11→MB4. All such path relationships are aggregated sequentially through the number index to construct a bypass path node set. The path set contains fields such as: flywheel number, channel number, bus node number, binding time, and current channel status. The uniqueness of the number is verified for each path relationship to ensure that the node mapping in the multi-channel structure under redundant bypass is unambiguous, forming a data structure that can be directly called.

[0160] The index extraction submodule, based on the bypass path node set, extracts the inertia balance index of each flywheel node and the voltage offset amplitude of the connected bus node, and calculates the normalized amplitude difference between the two using the following formula:

[0161] ;

[0162] The difference value of the normalized amplitude is obtained by calculation, and the steady-state response trend vector is constructed by superimposing the amplitude values ​​to generate the steady-state response difference trend quantity;

[0163] in, Indicates the first The difference in the normalized image value corresponding to each flywheel node. For the first Normalized value of the inertia balance index of each flywheel node For the first The normalized value of the voltage offset amplitude at the bus terminal nodes corresponding to each node. For the first Normalized sum of squared voltage changes in each node path For the first Frequency of historical switching actions in the path of each node For the first The normalized inverse ratio of the node switching time interval;

[0164] After receiving the bypass path node set, the index extraction submodule sequentially extracts the inertia balance index value of each node in the current cycle and the voltage offset amplitude of the bound bus node. In specific operation, the normalized value of the inertia balance index of flywheel node 5 is set. The normalized value of the voltage deviation amplitude of the connected bus node MB3 The sum of squared voltage changes at the corresponding nodes in the path is The frequency of path history switching actions is The corresponding inverse ratio of the switching time interval is According to the formula:

[0165] ;

[0166] The meanings of each parameter are explained below:

[0167] Indicates the first The difference in response amplitude between inertia and voltage state of each flywheel node is used to measure its steady-state consistency deviation.

[0168] The normalized value of the flywheel node inertia balance index is obtained by comparing the flywheel rotational inertia with the inertia threshold of its adjacent path and then performing normalization processing.

[0169] The normalized value of the voltage offset of the bound bus is derived from the normalized calculation of the offset of the instantaneous bus voltage relative to the system reference voltage;

[0170] The sum of squares of voltage changes within the sampling period in the path is uniformly normalized to the standard unit voltage square value.

[0171] Indicates the number of path history switching actions, collected from recent data. The frequency of switching within a cycle, after normalization, is set within the range of [0,10].

[0172] The normalized inverse ratio of the switching time interval is defined as follows: ,in This represents the time elapsed since the last switch on the current path. The maximum switching interval is set to 600 seconds as a reference value.

[0173] If the current node has been switched for 120 seconds since the last switch, then After normalization, it can be set to 0.2;

[0174] The formula's operational logic is explained as follows: The numerator quantifies the difference between the inertia index and the voltage state through absolute value, and multiplies it by the voltage change value as the response intensity. The denominator is the historical fluctuation probability influence factor, which suppresses the high error disturbance introduced by frequent path switching in the form of sigmoid, so that the structure has the function of dynamic compensation for path stability.

[0175] The advantage of the formula is that by introducing an exponential decay function into the denominator, a weight penalty is implemented for frequent path switching, thereby enhancing the ability to identify steady-state responses under historical path oscillations and forming a more robust stability evaluation quantity.

[0176] The results show that the amplitude difference of flywheel node No. 5 is 0.0689, which is in the low amplitude difference range. If the system tolerance benchmark is 0.12, its state is considered to be close to steady state, and it can be used as a candidate for the next step of stable node screening.

[0177] The tolerance judgment submodule compares the steady-state response difference trend with the set steady-state tolerance benchmark value item by item, identifies the node number that meets the steady-state condition, counts the proportion of the total number of nodes in the whole path, obtains the path stability index value, and generates the fault switching stability judgment result.

[0178] The tolerance judgment submodule performs node identification and filtering operations based on the steady-state response difference trend, comparing the values ​​of each flywheel node sequentially. Compared with the system's set steady-state tolerance benchmark value ,in The setting is based on the system's maximum permissible amplitude response error ratio of 5%. Under the conditions of maximum inertia index and voltage change offset, the average difference amplitude measured is 0.23. After statistical correction, 50% of this is taken as the tolerance boundary value, i.e., 0.12 is used as the threshold for steady-state identification. All conditions meeting these criteria are considered. The flywheel node numbers are recorded as "steady-state nodes." The ratio of this type of node to the total number of nodes along the path is used to form the stable node ratio. For example, if there are 12 nodes on the current path, and 9 of them meet the criteria, then the path stability index is [value missing]. After being normalized to the standard range, it is recorded as 0.75. If the system sets the recognition threshold to 0.6, the path is judged as a steady-state path and can be used as a candidate for the preferred path of redundancy switching in the fault-tolerant stage. The path is marked and written into the steady-state word structure, and the output is the path stability judgment result for subsequent path management logic matching.

[0179] 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 intelligent control system based on a centralized flywheel energy storage system, characterized in that, The system includes: The state perception module acquires the real-time voltage values ​​at the stator and rotor ends of the flywheel motor, calculates the first derivative of the voltage over three consecutive time slices, judges the difference between the derivative results, adjusts the sampling period, and generates sampling period adjustment information. The data partitioning module extracts the current rotational kinetic energy per unit time and the rate of change of rotational inertia per unit time of the flywheel energy storage unit based on the sampling period adjustment information. After normalizing the two parameters, a two-dimensional coordinate matrix is ​​constructed. Grouping determination is performed on the partition to which the coordinate point belongs to obtain the output partition identifier group. The main control and allocation module determines whether the energy output carrying capacity of the corresponding unit is higher than the set power output benchmark value per unit time based on the output zone identifier group. If it is higher, the zone power target output value is adjusted to the upper limit of load allocation. If it is lower, it is allocated to the lower limit output boundary and a flywheel group target power value set is generated. Based on the target power value set of the flywheel group, the power switching module determines whether the power deviation exceeds the output offset tolerance limit, switches the flywheel output to the redundant bypass and disconnects the original main path, and obtains the flywheel switching path record information. The sampling period adjustment information includes voltage change stability indication, voltage derivative trend identification result, and voltage sampling rhythm parameter. The output partition identification group specifically includes flywheel operation partition number, power release distribution label, and energy storage unit response attribute classification. The flywheel group target power value set includes unit-level energy carrying capacity scheduling parameters, target output power range setting value, and partition scheduling weight allocation identifier. The flywheel switching path record information specifically refers to the path switching node sequence, switching operation effective time, and power transmission path mapping relationship.

2. The power intelligent control system based on a centralized flywheel energy storage system according to claim 1, characterized in that, The state awareness module includes: The voltage acquisition submodule acquires the real-time voltage values ​​of the stator and rotor terminals of the flywheel motor, calculates the voltage change in each time period within three consecutive time slices, calculates the difference between the change in each time period and the change in the previous time period, obtains the voltage change rate per unit time in the voltage curve, and generates the voltage change rate value. The voltage derivative judgment submodule extracts the derivative values ​​of three time slices based on the voltage change rate value, compares the absolute differences pairwise, determines whether all three differences are lower than the set voltage slope change threshold, and determines whether there are any two sets of derivative values ​​with opposite signs and a difference greater than the voltage derivative reversal amplitude threshold, and obtains the derivative trend comparison result. Based on the derivative trend comparison results, the period adjustment submodule determines the current derivative fluctuation state. If it is a stable change, it adds the original value of the sampling period to the value of the sampling rhythm growth unit. If it is a trend reversal, it subtracts the value of the rhythm reduction unit to obtain the value after the period rhythm adjustment and generates sampling period adjustment information.

3. The intelligent power control system based on a centralized flywheel energy storage system according to claim 2, characterized in that, The data partitioning module includes: The kinetic energy extraction submodule collects the rotational speed and moment of inertia of the flywheel energy storage unit at the current moment according to the sampling period adjustment information, calculates the rotational kinetic energy output per unit time, and generates the kinetic energy output value per unit time by combining the time length within the time period with the standardization processing. The inertia normalization submodule calls the kinetic energy output value per unit time, and at the same time obtains the change amplitude of the flywheel energy storage unit inertia between two consecutive sampling points in the current moment, calculates the proportional coefficient between the change amplitude and the reference inertia value, and normalizes the proportional coefficient and the kinetic energy output per unit time to obtain the kinetic energy inertia mapping coordinate set value. The partitioning judgment submodule establishes a two-dimensional coordinate graph based on the kinetic energy inertia mapping coordinate group value, determines the position of the coordinate point in the four quadrants, calculates the relative gradient distribution density of the points in the quadrant region, classifies the points according to the quadrant number and density weight label, and obtains the output partitioning identifier group.

4. The intelligent power control system based on a centralized flywheel energy storage system according to claim 3, characterized in that, The main control and dispatch module includes: The inertia extraction submodule obtains the flywheel unit number currently in the high release rate zone based on the output zone identifier group, extracts the angular velocity value and mass distribution coefficient of the corresponding unit, calculates the rotational inertia value of the unit and summarizes them to generate a flywheel inertia data set. The energy fusion submodule calls the flywheel inertia data set, extracts the electromagnetic coupling coefficient and remaining kinetic energy value of each flywheel, converts the three data into output values ​​per unit time, and then performs combined calculations to obtain the unit-level energy output carrying capacity value. The power distribution submodule determines whether each flywheel unit is greater than or less than the unit-time power output reference value based on the unit-level energy output carrying capacity value. If it is greater, it is marked as the upper limit of load distribution; if it is less, it is marked as the lower limit of load distribution. The target output power range of the flywheel is adjusted based on the marking results to generate a target power value set for the flywheel group.

5. The intelligent power control system based on a centralized flywheel energy storage system according to claim 4, characterized in that, The power switching module includes: The power extraction submodule extracts the flywheel unit number that needs to be switched based on the target power value set of the flywheel group, collects the current output power value of the flywheel unit and the voltage drop value on the corresponding bus side, stores and archives the collected data according to the number index, and generates a flywheel unit power parameter group. The deviation judgment submodule calls the power parameter group of the flywheel unit, extracts the target output power value, actual output power value and bus voltage drop value of the corresponding numbered flywheel, calculates the actual power deviation value and voltage deviation amplitude value respectively, calculates the output deviation normalization intensity value, compares it with the set output deviation tolerance limit value, and if it is greater than the limit value, it is marked as a switching trigger state and generates switching judgment state mark information. The path control submodule determines whether the flywheel unit needs to perform a switching operation based on the switching determination status flag information. If the flag is marked as triggered, it records the current switching time point and channel number, and controls the control unit output path to disconnect from the original main path and simultaneously connect to the redundant bypass channel, generating flywheel switching path record information.

6. The intelligent power control system based on a centralized flywheel energy storage system according to claim 5, characterized in that, The system also includes: The stability judgment module calls the flywheel switching path recording information, extracts the inertia balance index of each flywheel node in the bypass path after switching and the voltage offset amplitude at the bus end, judges the stability of switching, and obtains the fault switching stability judgment result. The fault switching stability judgment results include inertia offset limit judgment results, voltage balance state level marking, and steady-state load offset label after switching.

7. The intelligent power control system based on a centralized flywheel energy storage system according to claim 6, characterized in that, The stability determination module includes: The path extraction submodule calls the flywheel switching path record information, extracts the flywheel node identifier in the bypass path after switching according to the record number and establishes a mapping relationship, connects to the downstream bus side node number in sequence, obtains complete path information, and generates a bypass path node set. The index extraction submodule extracts the inertia balance index of each flywheel node and the voltage offset amplitude of the connected bus node based on the bypass path node set, calculates the normalized amplitude difference between the two respectively, calculates the normalized amplitude difference value, constructs the steady-state response trend vector by amplitude superposition, and generates the steady-state response difference trend quantity. The tolerance judgment submodule compares the steady-state response difference trend with the set steady-state tolerance benchmark value item by item, identifies the node number that meets the steady-state condition, counts the proportion of the total number of nodes in the whole path, obtains the path stability index value, and generates the fault switching stability judgment result.

Images