Voltage sag compensation method and system based on flywheel energy storage

Abstract

The invention provides a voltage sag compensation method and system based on flywheel energy storage, and belongs to the field of voltage sag compensation. The method comprises the steps of collecting sag characteristic parameters for sag prediction, and obtaining predicted sag parameters and sag prediction errors; acquiring operation characteristic parameters and loss parameters of the plurality of flywheel energy storage units, and performing compensation matching degree analysis; flywheel monitoring error classification is carried out, and error matching degree analysis is carried out in combination with sag prediction errors; and carrying out loss balance degree analysis, and screening an optimal flywheel energy storage unit for compensation in combination with the compensation matching degree and the error matching degree. The technical problems of poor compensation effect and unbalanced use of the flywheel energy storage units caused by insufficient selection basis in the prior art are solved, and the flywheel energy storage units are accurately selected by comprehensively considering the compensation matching degree, the error matching degree and the loss balance degree. The technical effects of improving the voltage sag compensation effect and achieving balanced use of the flywheel energy storage unit are achieved.

Description

Technical Field

[0001] The present invention relates to the field of voltage sag compensation, and in particular to a voltage sag compensation method and system based on flywheel energy storage. Background Art

[0002] Voltage sags are a common power quality issue in power systems, severely impacting sensitive loads. To effectively address voltage sags, flywheel energy storage technology, with its fast response, long cycle life, and high power density, is widely used in voltage sag compensation.

[0003] In existing technologies, flywheel energy storage systems typically utilize multiple flywheel storage units operating in parallel to improve compensation capacity and reliability. When compensating for voltage sags, it's necessary to select the appropriate unit from among the multiple flywheel storage units to perform the compensation task. However, existing selection methods often only consider the remaining energy in the flywheel storage unit, which is an inadequate basis for selection, resulting in poor compensation results and uneven utilization of the flywheel storage units. Summary of the Invention

[0004] The present invention aims to solve the technical problems in the prior art of using flywheel energy storage units for voltage sag compensation, such as insufficient selection basis, poor compensation effect and uneven use of each flywheel energy storage unit, by providing a voltage sag compensation method and system based on flywheel energy storage.

[0005] The technical solution of the present invention to solve the above technical problems is as follows:

[0006] In a first aspect, the present invention provides a method for compensating for voltage sags based on flywheel energy storage, comprising: collecting sag characteristic parameters, performing sag prediction, obtaining predicted sag parameters, and obtaining sag prediction errors; obtaining multiple operating characteristic parameters and multiple loss parameters of multiple flywheel energy storage units, performing compensation matching analysis based on the multiple operating characteristic parameters and the predicted sag parameters, and obtaining multiple compensation matching degrees; classifying flywheel monitoring errors according to the multiple loss parameters respectively, obtaining multiple flywheel monitoring errors, and performing error matching analysis in combination with the sag prediction errors, and obtaining multiple error matching degrees; performing loss balance analysis according to the multiple loss parameters, obtaining multiple loss balance degrees, and calculating compensation scores of multiple flywheel energy storage units in combination with the multiple compensation matching degrees and the multiple error matching degrees, screening the optimal flywheel energy storage unit, and performing compensation when a voltage sag occurs.

[0007] In a second aspect, the present invention provides a voltage sag compensation system based on flywheel energy storage, comprising: a sag prediction module for collecting sag characteristic parameters, performing sag prediction, obtaining predicted sag parameters, and obtaining a sag prediction error; a compensation matching module for obtaining multiple operating characteristic parameters and multiple loss parameters of multiple flywheel energy storage units, performing compensation matching degree analysis based on the multiple operating characteristic parameters and the predicted sag parameters, and obtaining multiple compensation matching degrees; an error matching module for classifying flywheel monitoring errors according to the multiple loss parameters, obtaining multiple flywheel monitoring errors, performing error matching degree analysis in combination with the sag prediction error, and obtaining multiple error matching degrees; a scoring screening module for performing loss balance degree analysis based on the multiple loss parameters, obtaining multiple loss balance degrees, and calculating the compensation scores of multiple flywheel energy storage units in combination with the multiple compensation matching degrees and the multiple error matching degrees, screening the optimal flywheel energy storage unit, and performing compensation when a voltage sag occurs.

[0008] The beneficial effects of the present invention are:

[0009] The system collects sag characteristic parameters, performs sag prediction, obtains predicted sag parameters, and obtains sag prediction errors. Through the prediction mechanism, possible voltage sag situations are identified in advance and the prediction accuracy is quantified. Multiple operating characteristic parameters and multiple loss parameters of multiple flywheel energy storage units are obtained. Based on the multiple operating characteristic parameters and predicted sag parameters, compensation matching analysis is performed to obtain multiple compensation matching degrees, thereby evaluating the adaptability of each flywheel energy storage unit to the predicted sag compensation capability. Flywheel monitoring errors are classified according to multiple loss parameters to obtain multiple flywheel monitoring errors. Error matching analysis is performed in combination with sag prediction errors to obtain multiple error matching degrees, thereby quantifying the coordination between the monitoring accuracy of each flywheel energy storage unit and the sag prediction accuracy. Loss balance analysis is performed based on multiple loss parameters to obtain multiple loss balance degrees. Based on multiple compensation matching degrees and multiple error matching degrees, compensation scores of multiple flywheel energy storage units are calculated to obtain the optimal flywheel energy storage unit, and compensation is performed when a voltage sag occurs, thereby achieving accurate selection and optimized compensation based on multi-dimensional comprehensive evaluation.

[0010] Through the above technical solution, the problem of insufficient basis for selecting flywheel energy storage units in traditional methods is effectively solved, the voltage sag compensation effect is significantly improved and the balanced use of flywheel energy storage units is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 A schematic flow chart of a voltage sag compensation method based on flywheel energy storage provided by the present invention;

[0012] Figure 2 This is a structural schematic diagram of a voltage sag compensation system based on flywheel energy storage provided by the present invention.

[0013] In the accompanying drawings, the components represented by the reference numerals are as follows:

[0014] Temporary sag prediction module 11, compensation matching module 12, error matching module 13, scoring screening module 14. DETAILED DESCRIPTION

[0015] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0016] In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and should not be understood to indicate or imply relative importance or implicitly specify the number of the technical features indicated. Therefore, a feature specified as "first" or "second" may explicitly or implicitly include one or more of the specified features. In the description of the present invention, "plurality" means two or more, unless otherwise specifically defined.

[0017] In the description of the present invention, the term "for example" is used to mean "used as an example, illustration or illustration". Any embodiment of the present invention described as "for example" is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is given to enable any person skilled in the art to implement and use the present invention. In the following description, details are listed for the purpose of explanation. It should be understood that a person of ordinary skill in the art can recognize that the present invention can be implemented without using these specific details. In other examples, well-known structures and processes are not elaborated in detail to avoid obscuring the description of the present invention with unnecessary details. Therefore, the present invention is not intended to be limited to the embodiments shown, but is consistent with the widest scope consistent with the principles and features disclosed herein.

[0018] Example 1, as Figure 1 As shown, an embodiment of the present invention provides a voltage sag compensation method based on flywheel energy storage, comprising:

[0019] S1. Collect sag characteristic parameters, perform sag prediction, obtain predicted sag parameters, and obtain sag prediction errors.

[0020] Specifically, first, when the power system is operating normally or detects abnormal signs, the voltage sag characteristic parameters of each monitoring point in the power system are collected. These sag characteristic parameters include but are not limited to the power system's real-time voltage value, current value, frequency, power factor, load changes, equipment operating status, protection device action signals, line fault indication information, and historical voltage sag records. These sag characteristic parameters can reflect the current operating status of the power system and potential factors that may cause voltage sags. For example, when a key device shows signs of abnormal operation, potential voltage sag risks can be identified by monitoring changes in its electrical parameters.

[0021] Then, based on the collected sag characteristic parameters, a pre-trained sag prediction agent is used to predict sags. This sag prediction agent, built on machine learning, analyzes the input sag characteristic parameters and outputs predicted sag parameters. These parameters primarily include the magnitude of the voltage sag (e.g., a drop from the standard voltage of 100V to 70V), the duration of the sag, and the probability of a sag occurring.

[0022] At the same time, to evaluate the reliability of the predicted sag parameters, the sag prediction error is obtained. This sag prediction error is obtained by searching for sample sag characteristic parameters that are closest to the current sag characteristic parameters in the training data of the sag prediction agent, using these sample sag characteristic parameters for testing and verification, and calculating the deviation between the predicted sag parameters and the actual results. This provides a reliability reference for subsequent compensation selection.

[0023] By obtaining the predicted voltage sag parameters and the voltage sag prediction error, it is possible to predict the impending voltage sag and quantify the uncertainty of the prediction results, laying the foundation for selecting the optimal flywheel energy storage unit for compensation.

[0024] S2. Acquire multiple operating characteristic parameters and multiple loss parameters of multiple flywheel energy storage units, perform compensation matching degree analysis based on the multiple operating characteristic parameters and the predicted sag parameters, and obtain multiple compensation matching degrees.

[0025] Specifically, the operating characteristic parameters of multiple flywheel energy storage units are obtained through a real-time monitoring device. The operating characteristic parameters mainly include the speed, energy storage capacity, output voltage, output current, power output capacity and other parameters of the flywheel energy storage unit. At the same time, the corresponding loss parameters are obtained based on the maintenance records and historical usage data of each flywheel energy storage unit. The loss parameter reflects the historical usage loss of the flywheel energy storage unit, specifically the sum of the voltage change amplitudes of the flywheel energy storage unit for temporary sag compensation. For example, a certain flywheel energy storage unit has accumulated 1000V of voltage compensation in history, and the accumulated compensation voltage value is its loss parameter. The loss parameter can characterize the degree of use of the flywheel energy storage unit. The larger the loss parameter, the more frequently the unit has been used in history, and the currently available compensation capacity is relatively low.

[0026] Subsequently, a compensation matching analysis is performed based on multiple operating characteristic parameters and predicted voltage sag parameters. This compensation matching analysis inputs the operating characteristic parameters of each flywheel energy storage unit into a pre-established voltage sag compensation classification table, which is constructed using a mapping relationship between sample operating characteristic parameters and sample compensation voltages. By looking up the table, the compensation voltage that each flywheel energy storage unit can provide is output. The similarity between each compensation voltage and the obtained predicted voltage sag parameters is then calculated to obtain multiple compensation matching degrees. A higher compensation matching degree indicates that the flywheel energy storage unit is more suitable for compensating for the predicted voltage sag.

[0027] Through the above compensation matching analysis, the degree of compensation adaptation of each flywheel energy storage unit to the predicted voltage sag can be quantitatively evaluated, laying the foundation for the subsequent optimization selection of flywheel energy storage units.

[0028] S3. Classify the flywheel monitoring errors according to the multiple loss parameters to obtain multiple flywheel monitoring errors, and perform error matching analysis in combination with the sag prediction error to obtain multiple error matching degrees.

[0029] Specifically, since the flywheel energy storage unit will produce mechanical wear, electrical aging and other usage losses during long-term operation, these losses will directly affect the accuracy of monitoring the flywheel operation characteristic parameters. The loss parameters of each flywheel energy storage unit are input into a pre-constructed flywheel detection error classification table, which is constructed using sample loss parameters and sample flywheel monitoring errors with a mapping relationship. By looking up the table, the flywheel monitoring error corresponding to each flywheel energy storage unit is output. The flywheel monitoring error mainly includes the degree of monitoring error of flywheel operation characteristic parameters such as flywheel speed monitoring error and output voltage monitoring error. Among them, the flywheel energy storage unit with larger loss parameters has a correspondingly larger flywheel monitoring error. This is because the use loss of the equipment will lead to problems such as decreased sensor accuracy and increased signal transmission interference, thereby affecting the reliability of the monitoring data.

[0030] After obtaining the flywheel monitoring error for each flywheel energy storage unit, an error matching analysis is performed. This error matching analysis calculates the similarity between each flywheel monitoring error and the obtained sag prediction error, resulting in multiple error matching degrees. The error matching degree reflects the degree of match between the flywheel energy storage unit monitoring error and the sag prediction error. A higher error matching degree indicates a closer match between the two errors, enabling better coordination of the error effects during the compensation process and improving compensation accuracy.

[0031] Through error matching analysis, the compensation reliability of each flywheel energy storage unit can be reasonably evaluated under the premise of considering monitoring uncertainty, providing an error control basis for the final selection of the flywheel energy storage unit.

[0032] S4. Perform loss balance analysis based on multiple loss parameters to obtain multiple loss balance degrees. Combine the multiple compensation matching degrees and the multiple error matching degrees to calculate compensation scores for multiple flywheel energy storage units, screen and obtain the optimal flywheel energy storage unit, and perform compensation when a voltage sag occurs.

[0033] Specifically, first, a loss balance analysis is performed based on multiple loss parameters. Specifically, first, a predicted sag parameter is used to generate a call loss parameter, which represents the amount of voltage compensation required to perform this compensation task; then, the call loss parameter is added to the loss parameter of each flywheel energy storage unit to simulate the loss state of each flywheel energy storage unit after performing the compensation task, and the balance of the loss parameters of all flywheel energy storage units at this time is calculated to obtain multiple loss balances. Among them, the calculation of the loss balance can be determined by the inverse of the variance, which is used to evaluate the uniformity of the loss distribution of each flywheel energy storage unit. The higher the loss balance, the more uniform the loss distribution of the entire flywheel energy storage system is after selecting the flywheel energy storage unit for compensation, which is beneficial to extend the service life of each flywheel energy storage unit and improve the overall reliability and economy of the flywheel energy storage system.

[0034] After obtaining the loss balance degree of each flywheel energy storage unit, the system calculates a comprehensive compensation score for each flywheel energy storage unit, combining the compensation matching degree and error matching degree. This compensation score comprehensively considers the flywheel energy storage unit's compensation capability, error control capability, and loss balance effect, providing a comprehensive assessment of the overall performance of each flywheel energy storage unit. The flywheel energy storage unit with the highest compensation score is then selected as the optimal flywheel energy storage unit. When a voltage sag actually occurs, this optimal flywheel energy storage unit is activated for voltage compensation, achieving efficient and reliable voltage sag compensation control.

[0035] Through the above comprehensive evaluation and optimization selection process, it is possible to achieve balanced use of flywheel energy storage units while ensuring the compensation effect, thereby improving the overall performance and service life of the flywheel energy storage system, and achieving balanced use of flywheel energy storage units while improving the voltage sag compensation effect.

[0036] Furthermore, the sag characteristic parameters are collected, the sag prediction is performed, the predicted sag parameters are obtained, and the sag prediction error is obtained, including:

[0037] S11, collecting sag characteristic parameters;

[0038] S12, inputting the sag characteristic parameters into a sag prediction agent, and outputting predicted sag parameters, wherein the predicted sag parameters include voltage sag amplitude;

[0039] S13, test acquisition: predicting the temporary sag characteristic parameter to obtain a temporary sag prediction error, wherein the temporary sag characteristic parameter of the sample closest to the temporary sag characteristic parameter in the training data of the temporary sag prediction agent is used for testing.

[0040] In one feasible implementation, sag characteristic parameters are first collected. Specifically, real-time data from each monitoring point in the power system is collected every 0.01 second using power monitoring equipment such as voltage transformers, current transformers, and power meters. Collected sag characteristic parameters include: three-phase voltage RMS (e.g., phase A voltage 220V, phase B voltage 218V, phase C voltage 221V), three-phase current RMS, power system frequency (e.g., 50.02Hz), power factor (e.g., 0.95), the on / off status of each key device (e.g., 1 for closed, 0 for open), and load power change rate (e.g., 5% change per second).

[0041] The sag characteristic parameters are then input into the sag prediction agent, which outputs the predicted sag parameters. This sag prediction agent is constructed using a neural network algorithm and may include an input layer, a hidden layer, and an output layer. For example, the number of nodes in the input layer equals the dimensionality of the sag characteristic parameters, e.g., 15 parameters correspond to 15 input nodes. The hidden layer contains 32 neuron nodes and uses the ReLU activation function. The output layer contains three nodes, corresponding to the voltage sag magnitude, sag duration, and sag probability, respectively.

[0042] When the currently collected 15-dimensional sag feature parameters are input, the sag prediction agent calculates them through forward propagation and outputs the predicted sag parameters. For example, the input sag feature parameters are [220, 218, 221, 15.2, 14.8, 15.5, 50.02, 0.95, 1, 1, 0, 5%, 3%, 2%, 0.8], where 220, 218, and 221 represent the RMS voltage (V) of phases A, B, and C, respectively; 15.2, 14.8, and 15.5 represent the RMS current (A) of phases A, B, and C, respectively; 50.02 represents the power system frequency (Hz); 0.95 represents the power factor; 1, 1, and 0 represent the on / off states of three key devices (1 = closed, 0 = open); 5%, 3%, and 2% represent the power change rates of three loads; and 0.8 represents the voltage stability index. After receiving the 15-dimensional sag feature parameters, the sag prediction agent outputs the predicted sag parameter: a voltage sag amplitude of 30V (i.e., from 220V to 190V). Simultaneously, the test obtains the sag prediction error by predicting the sag feature parameters. Specifically, within the training dataset of the sag prediction agent, the Euclidean distance formula is used to calculate the similarity between the current sag feature parameters and the sag feature parameters of all samples. The Euclidean distance calculation formula is:

[0043] ;

[0044] in, is the current sag characteristic parameter, is the sample temporary dip characteristic parameter, and n is the parameter dimension. The sample temporary dip characteristic parameter with the smallest distance is selected as the closest sample temporary dip characteristic parameter. For example, the closest sample temporary dip characteristic parameter with a distance value of 2.3 is found.

[0045] The actual voltage sag parameter corresponding to the closest sample sag characteristic parameter is then compared with the sag prediction agent's prediction of the sample sag characteristic parameter. Assuming the actual voltage sag amplitude of the closest sample sag characteristic parameter is 28V, and the sag prediction agent predicts it as 26V, the sag prediction error is |28-26|=2V. This sag prediction error reflects the accuracy of the sag prediction agent's prediction of the current input sag characteristic parameter. A smaller sag prediction error indicates a more reliable prediction.

[0046] Through the above process, specific predicted sag parameters and corresponding sag prediction errors can be obtained, providing an accurate data basis for subsequent flywheel energy storage unit selection.

[0047] Furthermore, the training steps of the temporary drop prediction agent include:

[0048] S121. Collect a sample sag characteristic parameter set based on voltage sag record data within a historical period, and collect sag parameters under different sample sag characteristic parameters to obtain a sample sag parameter set;

[0049] S122. Constructing the architecture of a temporary sag prediction agent based on machine learning;

[0050] S123. Use the sample temporary sag feature parameter set and the sample temporary sag parameter set as training data to iteratively train the temporary sag prediction agent, and complete the training after the loss test converges.

[0051] In a preferred embodiment, first, based on the voltage sag record data within the historical time, a sample sag characteristic parameter set is collected, and the sag parameters under different sample sag characteristic parameters are collected to obtain a sample sag parameter set. Specifically, all voltage sag event records that occurred in the past 1-3 years are extracted from the historical operation database of the power system. Each sag event contains the power system operation status data before the sag occurs and the actual sag parameters when the sag occurs. For the collection of the sample sag characteristic parameter set, the power system operation data within 5-10 seconds before each sag event is extracted, including parameters such as the three-phase voltage effective value, the three-phase current effective value, the system frequency, the power factor, the equipment switch status, and the load power change rate to form the sample sag characteristic parameters. For example, the sample sag characteristic parameters before a certain sag event are [218, 215, 220, 14.8, 15.1, 14.5, 49.98, 0.92, 1, 0, 1, 8%, 2%, 4%, 0.7]. For the collection of sample sag parameter sets, the actual voltage sag magnitude for each sag event is recorded. For example, the actual voltage sag magnitude corresponding to the sample sag characteristic parameter described above is 35V, which serves as the sample sag parameter. By collecting a large amount of this data, a sample sag characteristic parameter set and the corresponding sample sag parameter set are established.

[0052] Next, the architecture of a machine learning-based voltage sag prediction agent was constructed. Specifically, a multi-layer neural network structure was used to construct the voltage sag prediction agent. For example, the architecture included: an input layer with a node number equal to the dimension of the sample voltage sag characteristic parameters (e.g., 15 nodes); a first hidden layer with 32 neurons and a ReLU activation function; a second hidden layer with 16 neurons and a ReLU activation function; and an output layer with one node and a linear activation function, which outputs the predicted voltage sag parameter, i.e., the predicted voltage sag amplitude. Each layer was fully connected, and the weight parameters were initialized using random initialization.

[0053] Subsequently, the sample sag feature parameter set and the sample sag parameter set are used as training data to iteratively train the sag prediction agent, and the training is completed after the loss test converges. Specifically, the sample sag feature parameter set is used as the training input, and the sample sag parameter set is used as the training output, and the back propagation algorithm is used for training. The mean square error is used as the loss function in the training process, and the calculation formula is:

[0054] ;

[0055] in, is the actual voltage sag amplitude, that is, the sample sag parameter, To predict the voltage sag magnitude, n is the number of samples. Gradient descent is used to optimize network parameters, with a learning rate of 0.001 and a batch size of 32. During training, the loss function is used to calculate the loss value after every 100 iterations. When the loss value changes by less than 0.001 for 10 consecutive times, the loss test is considered converged, and the training of the sag prediction agent is complete. After training, the sag prediction agent can accurately predict the voltage sag magnitude, i.e., the sag parameters, based on the input sag characteristic parameters.

[0056] Furthermore, multiple operating characteristic parameters and multiple loss parameters of multiple flywheel energy storage units are obtained, and compensation matching degree analysis is performed based on the multiple operating characteristic parameters and the predicted sag parameters to obtain multiple compensation matching degrees, including:

[0057] S21. Monitoring and acquiring a plurality of operating characteristic parameters of a plurality of flywheel energy storage units, wherein each operating characteristic parameter includes a rotational speed of the flywheel energy storage unit;

[0058] S22. Acquire multiple loss parameters based on maintenance data of multiple flywheel energy storage units;

[0059] S23. Perform compensation matching analysis based on the multiple operating characteristic parameters and the predicted sag parameters to obtain multiple compensation matching degrees.

[0060] In a preferred embodiment, first, multiple operating characteristic parameters of multiple flywheel energy storage units are monitored and obtained, wherein each operating characteristic parameter includes the rotational speed of the flywheel energy storage unit. Specifically, the operating status information of the flywheel energy storage unit is obtained in real time by monitoring equipment such as speed sensors, voltage sensors, and current sensors installed on each flywheel energy storage unit. The operating characteristic parameter includes the rotational speed of the flywheel energy storage unit. According to the physical principle of flywheel energy storage, the rotational kinetic energy of the flywheel is:

[0061] ;

[0062] Where J is the moment of inertia, is the angular velocity (proportional to the rotational speed). The relationship between angular velocity and rotational speed n is: ω = 2πn / 60, where n is the rotational speed (rpm). Therefore, the energy stored in the flywheel is proportional to the square of the rotational speed. For example, flywheel energy storage unit A rotates at 15,000 rpm, flywheel energy storage unit B rotates at 12,000 rpm, and flywheel energy storage unit C rotates at 18,000 rpm. Higher rotational speeds indicate greater energy storage and greater voltage compensation capability.

[0063] Subsequently, based on the maintenance data of multiple flywheel energy storage units, multiple loss parameters are obtained. Specifically, the usage loss information of each unit is extracted from the historical maintenance record database of the flywheel energy storage unit. The loss parameter is specifically the sum of the voltage change amplitudes for temporary sag compensation of the flywheel energy storage unit. The loss parameter records the total amount of compensation voltage provided by each flywheel energy storage unit in history. For example, the loss parameter of flywheel energy storage unit A is 800V, indicating that the unit has provided a cumulative voltage compensation of 800V in history; the loss parameter of flywheel energy storage unit B is 1200V; and the loss parameter of flywheel energy storage unit C is 600V. The larger the loss parameter, the more frequently the flywheel energy storage unit has been used in history, and the relatively shorter its current remaining service life.

[0064] Next, a compensation matching analysis is performed based on multiple operating characteristic parameters and predicted sag parameters to obtain multiple compensation matching degrees. Specifically, the operating characteristic parameters of each flywheel energy storage unit are input into a pre-established sag compensation classification table. This sag compensation classification table establishes a mapping relationship between a large number of sample operating characteristic parameters and corresponding sample compensation voltages. For example, a speed of 15,000 rpm corresponds to a compensation voltage of 35V, and a speed of 12,000 rpm corresponds to a compensation voltage of 28V. By looking up the table, the compensation voltage that each flywheel energy storage unit can provide is obtained. The similarity between each compensation voltage and the predicted sag parameter (voltage sag magnitude) is then calculated to obtain multiple compensation matching degrees.

[0065] Through the above steps, the compensation matching degree of each flywheel energy storage unit is obtained. The higher the compensation matching degree, the more closely the compensation capability of the flywheel energy storage unit matches the compensation requirement of the predicted sag parameter.

[0066] Furthermore, a compensation matching degree analysis is performed based on the multiple operating characteristic parameters and the predicted sag parameters to obtain multiple compensation matching degrees, including:

[0067] S231, inputting each operating characteristic parameter into a sag compensation classification table, and outputting a plurality of compensation voltages, wherein the sag compensation classification table is constructed using sample operating characteristic parameters and sample compensation voltages having a mapping relationship;

[0068] S232: Calculate similarities between multiple compensation voltages and the predicted sag parameters to obtain multiple compensation matching degrees.

[0069] In a preferred embodiment, each operating characteristic parameter is first input into a sag compensation classification table, and multiple compensation voltages are output. The sag compensation classification table is constructed using sample operating characteristic parameters and sample compensation voltages with a mapping relationship. Specifically, the sag compensation classification table is constructed by collecting historical data on the actual compensation capabilities of each flywheel energy storage unit under different operating conditions and establishing a mapping relationship between the sample operating characteristic parameters and the sample compensation voltages. For example, through historical data analysis, the following mapping relationship is established: when the flywheel speed in the sample operating characteristic parameters is 10,000 rpm, the corresponding sample compensation voltage is 20V; when the flywheel speed in the sample operating characteristic parameters is 12,000 rpm, the corresponding sample compensation voltage is 28V; when the flywheel speed in the sample operating characteristic parameters is 15,000 rpm, the corresponding sample compensation voltage is 35V; and when the flywheel speed in the sample operating characteristic parameters is 18,000 rpm, the corresponding sample compensation voltage is 42V. This mapping relationship forms the sag compensation classification table, which is stored in the system as a lookup table.

[0070] Subsequently, the operating characteristic parameters of each flywheel energy storage unit are entered into the sag compensation classification table for a lookup operation. For example, the operating characteristic parameters of flywheel energy storage unit A include a speed of 15,000 rpm, and the table lookup shows that its compensation voltage is 35V; the operating characteristic parameters of flywheel energy storage unit B include a speed of 12,000 rpm, and the table lookup shows that its compensation voltage is 28V; the operating characteristic parameters of flywheel energy storage unit C include a speed of 18,000 rpm, and the table lookup shows that its compensation voltage is 42V.

[0071] Then, the similarities between multiple compensation voltages and the predicted sag parameters are calculated to obtain multiple compensation matching degrees. Specifically, a similarity calculation formula is used to assess the degree of matching between the compensation voltage of each flywheel energy storage unit and the predicted sag parameter (voltage sag amplitude). The similarity calculation formula is: Compensation matching degree = 1 - |compensation voltage - predicted sag parameter| / predicted sag parameter. For example, when the predicted sag parameter is 30V, the compensation matching degree of each flywheel energy storage unit is calculated as follows: flywheel energy storage unit A: compensation matching degree = 1-|35-30| / 30 = 1-5 / 30 = 0.83, that is, the compensation matching degree of flywheel energy storage unit A is 0.83; flywheel energy storage unit B: compensation matching degree = 1-|28-30| / 30 = 1-2 / 30 = 0.93, that is, the compensation matching degree of flywheel energy storage unit B is 0.93; flywheel energy storage unit C: compensation matching degree = 1-|42-30| / 30 = 1-12 / 30 = 0.60, that is, the compensation matching degree of flywheel energy storage unit C is 0.60.

[0072] Through the above steps, the compensation matching degree of each flywheel energy storage unit is obtained. The closer the compensation matching degree is to 1, the more closely the compensation capability of the flywheel energy storage unit matches the predicted voltage sag requirement, and the more suitable it is for voltage sag compensation.

[0073] Furthermore, flywheel monitoring errors are classified according to multiple loss parameters to obtain multiple flywheel monitoring errors. Error matching analysis is performed in combination with the sag prediction error to obtain multiple error matching degrees, including:

[0074] S31. Inputting a plurality of loss parameters into a flywheel detection error classification table, and outputting a plurality of flywheel monitoring errors, wherein the flywheel detection error classification table is constructed using sample loss parameters and sample flywheel monitoring errors having a mapping relationship, and the flywheel monitoring errors include monitoring error degrees of flywheel operating characteristic parameters;

[0075] S32. Calculate the similarity between each flywheel monitoring error and the sag prediction error respectively to obtain multiple error matching degrees.

[0076] In a preferred embodiment, multiple loss parameters are first input into a flywheel detection error classification table, and multiple flywheel monitoring errors are output. The flywheel detection error classification table is constructed using a mapping relationship between sample loss parameters and sample flywheel monitoring errors. The flywheel monitoring errors include the degree of monitoring error of the flywheel's operating characteristic parameters. Because flywheel energy storage units experience wear and tear during long-term operation, such as mechanical wear and electrical aging, these losses directly affect the sensor's monitoring accuracy and the reliability of data transmission. The flywheel detection error classification table is constructed by collecting historical monitoring error data corresponding to flywheel energy storage units with different degrees of loss and establishing a mapping relationship between sample loss parameters and sample flywheel monitoring errors. For example, when the sample loss parameter is 500V, the corresponding sample flywheel monitoring error is 1.5V; when the sample loss parameter is 800V, the corresponding sample flywheel monitoring error is 2.2V; and when the sample loss parameter is 1200V, the corresponding sample flywheel monitoring error is 3.1V. This mapping relationship reflects the rule that the larger the loss parameter, the larger the flywheel monitoring error.

[0077] Then, the loss parameters of each flywheel energy storage unit are entered into the flywheel detection error classification table for a lookup operation. For example, if the loss parameter of flywheel energy storage unit A is 800V, the flywheel monitoring error obtained by table lookup is 2.2V; the loss parameter of flywheel energy storage unit B is 1200V, the flywheel monitoring error obtained by table lookup is 3.1V; and the loss parameter of flywheel energy storage unit C is 600V, the flywheel monitoring error obtained by table lookup is 1.8V.

[0078] Subsequently, the similarity between each flywheel monitoring error and the sag prediction error is calculated to obtain multiple error matching degrees. Specifically, a similarity calculation formula is used to assess the degree of matching between the flywheel monitoring error and the sag prediction error for each flywheel energy storage unit. The similarity calculation formula is: Error matching degree = 1 - |flywheel monitoring error - sag prediction error| / sag prediction error. For example, when the sag prediction error is 2V, the error matching degree of each flywheel energy storage unit is calculated as follows: flywheel energy storage unit A: error matching degree = 1-|2.2-2| / 2 = 1-0.2 / 2 = 0.90, that is, the error matching degree of flywheel energy storage unit A is 0.90; flywheel energy storage unit B: error matching degree = 1-|3.1-2| / 2 = 1-1.1 / 2 = 0.45, that is, the error matching degree of flywheel energy storage unit B is 0.45; flywheel energy storage unit C: error matching degree = 1-|1.8-2| / 2 = 1-0.2 / 2 = 0.90, that is, the error matching degree of flywheel energy storage unit C is 0.90.

[0079] Through the above steps, the error matching degree of each flywheel energy storage unit is obtained. The higher the error matching degree, the closer the monitoring error and sag prediction error of the flywheel energy storage unit are, which can better coordinate the error effects during the compensation process and improve compensation accuracy.

[0080] Furthermore, a loss balancing analysis is performed based on multiple loss parameters to obtain multiple loss balancing degrees, including:

[0081] S41, using the predicted sag parameter to generate a call loss parameter;

[0082] S42. Add the called loss parameter to each loss parameter in sequence, and calculate the loss parameter balance degree to obtain multiple loss balance degrees.

[0083] In a preferred embodiment, a call loss parameter is first generated using the predicted sag parameter. Specifically, the voltage compensation amount required to execute the current compensation task is calculated based on the voltage sag amplitude in the predicted sag parameter. This voltage compensation amount is the call loss parameter. For example, if the voltage sag amplitude in the predicted sag parameter is 30V, a call loss parameter of 30V is generated, indicating that 30V of voltage compensation capacity is required to execute the current compensation task.

[0084] Then, the loss parameters are added to each loss parameter in turn, and the loss parameter balance is calculated to obtain multiple loss balance degrees. Specifically, the loss state of each flywheel energy storage unit after performing the compensation task is simulated, and the expected loss parameters of each flywheel energy storage unit after performing the compensation are obtained by adding the loss parameters to the current loss parameters of each flywheel energy storage unit. For example, assuming that the current loss parameters of each flywheel energy storage unit are: flywheel energy storage unit A is 800V, flywheel energy storage unit B is 1200V, and flywheel energy storage unit C is 600V. When the loss parameter is called as 30V, the expected loss parameters of each flywheel energy storage unit after compensation are as follows: select flywheel energy storage unit A for compensation: A is 830V, B is 1200V, and C is 600V; select flywheel energy storage unit B for compensation: A is 800V, B is 1230V, and C is 600V; select flywheel energy storage unit C for compensation: A is 800V, B is 1200V, and C is 630V.

[0085] Subsequently, the balance of the loss parameters of all flywheel energy storage units under various selection schemes is calculated to obtain multiple loss balance degrees. The loss parameter balance degree is calculated using the inverse of the variance, and the formula is: loss balance degree = 1 / σ², where σ² is the loss parameter variance. For example, after selecting flywheel energy storage unit A for compensation, the loss parameter sequence is [830, 1200, 600], and its variance σ² = [(830-876.7)² + (1200-876.7)² + (600-876.7)²] / 3 = 91511, and the loss balance degree = 1 / 91511 = 1.09× According to the above calculation method, multiple loss balance degrees corresponding to multiple flywheel energy storage units can be obtained in sequence.

[0086] Through the above steps, the corresponding loss balance degree of each flywheel energy storage unit is obtained. The higher the loss balance degree, the more evenly the loss distribution of the entire flywheel energy storage system is after selecting the flywheel energy storage unit for compensation, which is conducive to extending the service life of each flywheel energy storage unit.

[0087] Furthermore, combining the multiple compensation matching degrees and the multiple error matching degrees, calculating and obtaining compensation scores of multiple flywheel energy storage units, screening and obtaining the optimal flywheel energy storage unit, and performing compensation when a voltage sag occurs, including:

[0088] S43. Calculate and obtain multiple compensation scores based on the multiple compensation matching degrees, the multiple error matching degrees, and the multiple loss leveling degrees;

[0089] S42. Select the flywheel energy storage unit with the largest compensation score as the optimal flywheel energy storage unit to compensate when a voltage sag occurs.

[0090] In a preferred embodiment, first, multiple compensation scores are calculated based on multiple compensation matching degrees, multiple error matching degrees and multiple loss balancing degrees. Specifically, a weighted summation method is adopted to comprehensively calculate the compensation matching degree, error matching degree and loss balancing degree of each flywheel energy storage unit to obtain a comprehensive compensation score of each flywheel energy storage unit. The calculation formula is: compensation score = α×compensation matching degree + β×error matching degree + γ×loss balancing degree. Among them, α, β, and γ are weight coefficients, satisfying α+β+γ=1. For example, α=0.5, β=0.3, and γ=0.2 can be set to represent the importance of compensation matching degree, error matching degree and loss balancing degree in the comprehensive evaluation, respectively.

[0091] For example, the score of each flywheel energy storage unit is calculated as follows:

[0092] Flywheel energy storage unit A: compensation score = 0.5 × 0.83 + 0.3 × 0.90 + 0.2 × 1.09 × =0.415+0.27+0.000002=0.685;

[0093] Flywheel energy storage unit B: compensation score = 0.5 × 0.93 + 0.3 × 0.45 + 0.2 × 8.52 × =0.465+0.135+0.000002=0.600;

[0094] Flywheel energy storage unit C: compensation score = 0.5 × 0.60 + 0.3 × 0.90 + 0.2 × 1.25 × =0.30+0.27+0.0000025=0.570.

[0095] The flywheel energy storage unit with the highest compensation score is then selected as the optimal flywheel energy storage unit to compensate for voltage sags. Specifically, the compensation scores of each flywheel energy storage unit are compared, and the unit with the highest score is selected as the optimal flywheel energy storage unit. In the above example, flywheel energy storage unit A has the highest compensation score of 0.685, so it is selected as the optimal flywheel energy storage unit.

[0096] When a voltage sag actually occurs in the power system, the optimal flywheel energy storage unit A is immediately activated to compensate for the voltage. By releasing the rotational kinetic energy of flywheel A and converting it into electrical output, it provides the required voltage support for the power system, thus achieving rapid and accurate compensation for the voltage sag.

[0097] Through a comprehensive evaluation and optimization selection process, it is possible to comprehensively consider multiple factors such as compensation capacity, error control and loss balance while ensuring the compensation effect, to achieve intelligent selection and efficient utilization of flywheel energy storage units, thereby improving the voltage sag compensation effect and achieving balanced use of flywheel energy storage units.

[0098] Example 2, as Figure 2 As shown, based on the same inventive concept as the voltage sag compensation method based on flywheel energy storage provided in Example 1, an embodiment of the present invention further provides a voltage sag compensation system based on flywheel energy storage, comprising:

[0099] The sag prediction module 11 is used to collect sag characteristic parameters, perform sag prediction, obtain predicted sag parameters, and obtain sag prediction errors;

[0100] The compensation matching module 12 is used to obtain multiple operating characteristic parameters and multiple loss parameters of multiple flywheel energy storage units, perform compensation matching degree analysis based on the multiple operating characteristic parameters and the predicted sag parameters, and obtain multiple compensation matching degrees;

[0101] an error matching module 13 for classifying flywheel monitoring errors according to a plurality of loss parameters, obtaining a plurality of flywheel monitoring errors, and performing error matching degree analysis in combination with the sag prediction error to obtain a plurality of error matching degrees;

[0102] The scoring and screening module 14 is used to perform loss balance analysis based on multiple loss parameters to obtain multiple loss balance degrees. In combination with the multiple compensation matching degrees and multiple error matching degrees, the module calculates compensation scores for multiple flywheel energy storage units, selects the optimal flywheel energy storage unit, and compensates for voltage sags when voltage sags occur.

[0103] Furthermore, the sag prediction module 11 includes the following execution steps:

[0104] Collect sag characteristic parameters;

[0105] Inputting the sag characteristic parameters into a sag prediction agent, and outputting predicted sag parameters, wherein the predicted sag parameters include the voltage sag amplitude;

[0106] The test obtains the temporary sag characteristic parameter by predicting the temporary sag prediction error, wherein the temporary sag characteristic parameter of the sample closest to the temporary sag characteristic parameter in the training data of the temporary sag prediction agent is used for testing.

[0107] Furthermore, the training steps of the temporary drop prediction agent include:

[0108] According to the voltage sag record data in the historical time, a sample sag characteristic parameter set is collected, and the sag parameters under different sample sag characteristic parameters are collected to obtain a sample sag parameter set;

[0109] Build the architecture of a machine learning-based sag prediction agent;

[0110] The sample sag feature parameter set and the sample sag parameter set are used as training data to iteratively train the sag prediction agent, and the training is completed after the loss test converges.

[0111] Furthermore, the compensation matching module 12 includes the following execution steps:

[0112] Monitoring and acquiring a plurality of operating characteristic parameters of a plurality of flywheel energy storage units, wherein each operating characteristic parameter includes a rotational speed of the flywheel energy storage unit;

[0113] Acquiring multiple loss parameters based on maintenance data of multiple flywheel energy storage units;

[0114] A compensation matching degree analysis is performed based on the multiple operating characteristic parameters and the predicted sag parameters to obtain multiple compensation matching degrees.

[0115] Furthermore, the compensation matching module 12 further includes the following execution steps:

[0116] Input each operating characteristic parameter into a sag compensation classification table, and output a plurality of compensation voltages, wherein the sag compensation classification table is constructed using sample operating characteristic parameters and sample compensation voltages having a mapping relationship;

[0117] Similarities between a plurality of compensation voltages and the predicted sag parameters are calculated to obtain a plurality of compensation matching degrees.

[0118] Furthermore, the error matching module 13 includes the following execution steps:

[0119] Inputting a plurality of loss parameters into a flywheel detection error classification table and outputting a plurality of flywheel monitoring errors, wherein the flywheel detection error classification table is constructed using sample loss parameters and sample flywheel monitoring errors having a mapping relationship, and the flywheel monitoring errors include monitoring error degrees of flywheel operating characteristic parameters;

[0120] The similarity between each flywheel monitoring error and the sag prediction error is calculated separately to obtain multiple error matching degrees.

[0121] Furthermore, the scoring and screening module 14 includes the following steps:

[0122] Using the predicted sag parameter, generating a call loss parameter;

[0123] The call loss parameter is added to each loss parameter in turn, and the loss parameter balance degree is calculated to obtain multiple loss balance degrees.

[0124] Furthermore, the scoring and screening module 14 further includes the following execution steps:

[0125] Calculating and obtaining a plurality of compensation scores according to the plurality of compensation matching degrees, the plurality of error matching degrees, and the plurality of loss leveling degrees;

[0126] The flywheel energy storage unit with the largest compensation score is selected as the optimal flywheel energy storage unit to compensate when a voltage sag occurs.

[0127] It should be noted that, in the above embodiments, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant description of other embodiments.

[0128] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0129] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0130] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0131] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1A step that specifies a function in one or more boxes.

[0132] Although preferred embodiments of the present invention have been described, additional changes and modifications to these embodiments may occur to those skilled in the art once the basic inventive concepts become known.

[0133] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the present invention and its equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A voltage sag compensation method based on flywheel energy storage, characterized in that: The method comprises: Collect sag characteristic parameters, perform sag prediction, obtain predicted sag parameters, and obtain sag prediction error; Acquiring multiple operating characteristic parameters and multiple loss parameters of multiple flywheel energy storage units, performing compensation matching degree analysis based on the multiple operating characteristic parameters and the predicted sag parameters, and obtaining multiple compensation matching degrees; Classifying the flywheel monitoring errors according to the multiple loss parameters to obtain multiple flywheel monitoring errors, and performing error matching analysis in combination with the sag prediction error to obtain multiple error matching degrees; A loss balance analysis is performed based on multiple loss parameters to obtain multiple loss balance degrees. The compensation scores of multiple flywheel energy storage units are calculated by combining the multiple compensation matching degrees and multiple error matching degrees. The optimal flywheel energy storage unit is screened and compensated when a voltage sag occurs.

2. The voltage sag compensation method based on flywheel energy storage according to claim 1, characterized in that: Collect sag characteristic parameters, perform sag prediction, obtain predicted sag parameters, and obtain sag prediction errors, including: Collect sag characteristic parameters; Inputting the sag characteristic parameters into a sag prediction agent, and outputting predicted sag parameters, wherein the predicted sag parameters include the voltage sag amplitude; The test obtains the temporary sag characteristic parameter by predicting the temporary sag prediction error, wherein the temporary sag characteristic parameter of the sample closest to the temporary sag characteristic parameter in the training data of the temporary sag prediction agent is used for testing.

3. The voltage sag compensation method based on flywheel energy storage according to claim 2, characterized in that: The training steps of the temporary drop prediction agent include: According to the voltage sag record data in the historical time, a sample sag characteristic parameter set is collected, and the sag parameters under different sample sag characteristic parameters are collected to obtain a sample sag parameter set; Build the architecture of a machine learning-based sag prediction agent; The sample sag feature parameter set and the sample sag parameter set are used as training data to iteratively train the sag prediction agent, and the training is completed after the loss test converges.

4. The voltage sag compensation method based on flywheel energy storage according to claim 1, characterized in that: Acquire multiple operating characteristic parameters and multiple loss parameters of multiple flywheel energy storage units, perform compensation matching analysis based on the multiple operating characteristic parameters and predicted sag parameters, and obtain multiple compensation matching degrees, including: Monitoring and acquiring a plurality of operating characteristic parameters of a plurality of flywheel energy storage units, wherein each operating characteristic parameter includes a rotational speed of the flywheel energy storage unit; Acquiring multiple loss parameters based on maintenance data of multiple flywheel energy storage units; A compensation matching degree analysis is performed based on the multiple operating characteristic parameters and the predicted sag parameters to obtain multiple compensation matching degrees.

5. The voltage sag compensation method based on flywheel energy storage according to claim 4, characterized in that: Perform compensation matching analysis based on the multiple operating characteristic parameters and the predicted sag parameters to obtain multiple compensation matching degrees, including: Input each operating characteristic parameter into a sag compensation classification table, and output a plurality of compensation voltages, wherein the sag compensation classification table is constructed using sample operating characteristic parameters and sample compensation voltages having a mapping relationship; Similarities between a plurality of compensation voltages and the predicted sag parameters are calculated to obtain a plurality of compensation matching degrees.

6. The voltage sag compensation method based on flywheel energy storage according to claim 1, characterized in that: Flywheel monitoring errors are classified according to multiple loss parameters to obtain multiple flywheel monitoring errors. Error matching analysis is performed in combination with the sag prediction error to obtain multiple error matching degrees, including: Inputting a plurality of loss parameters into a flywheel detection error classification table and outputting a plurality of flywheel monitoring errors, wherein the flywheel detection error classification table is constructed using sample loss parameters and sample flywheel monitoring errors having a mapping relationship, and the flywheel monitoring errors include monitoring error degrees of flywheel operating characteristic parameters; The similarity between each flywheel monitoring error and the sag prediction error is calculated separately to obtain multiple error matching degrees.

7. The voltage sag compensation method based on flywheel energy storage according to claim 1, characterized in that: Perform loss balancing analysis based on multiple loss parameters to obtain multiple loss balancing degrees, including: Using the predicted sag parameter, generating a call loss parameter; The call loss parameter is added to each loss parameter in turn, and the loss parameter balance degree is calculated to obtain multiple loss balance degrees.

8. The voltage sag compensation method based on flywheel energy storage according to claim 1, characterized in that: Combining the multiple compensation matching degrees and the multiple error matching degrees, calculating the compensation scores of the multiple flywheel energy storage units, screening and obtaining the optimal flywheel energy storage unit, and performing compensation when a voltage sag occurs, including: Calculating and obtaining a plurality of compensation scores according to the plurality of compensation matching degrees, the plurality of error matching degrees, and the plurality of loss leveling degrees; The flywheel energy storage unit with the largest compensation score is selected as the optimal flywheel energy storage unit to compensate when a voltage sag occurs.

9. A voltage sag compensation system based on flywheel energy storage, characterized in that: A system for implementing a voltage sag compensation method based on flywheel energy storage according to any one of claims 1 to 8, the system comprising: The sag prediction module is used to collect sag characteristic parameters, perform sag prediction, obtain predicted sag parameters, and obtain sag prediction errors; A compensation matching module is used to obtain multiple operating characteristic parameters and multiple loss parameters of multiple flywheel energy storage units, perform compensation matching degree analysis based on the multiple operating characteristic parameters and predicted sag parameters, and obtain multiple compensation matching degrees; an error matching module, configured to classify flywheel monitoring errors according to a plurality of loss parameters, obtain a plurality of flywheel monitoring errors, and perform error matching degree analysis in combination with the sag prediction error to obtain a plurality of error matching degrees; The scoring and screening module is used to perform loss balance analysis based on multiple loss parameters to obtain multiple loss balance degrees. In combination with the multiple compensation matching degrees and multiple error matching degrees, the module calculates the compensation scores of multiple flywheel energy storage units, selects the optimal flywheel energy storage unit, and compensates for voltage sags in the event of a voltage sag.

Images