A flywheel energy storage system optimization method based on multi-vector model predictive control

Abstract

The application discloses a flywheel energy storage system optimization method based on a multi-vector model predictive control, belongs to the technical field of energy storage control, and through the collaborative optimization of multi-vector model predictive current control and multi-vector model predictive torque control, adopts three-vector synthesis technology in a flywheel energy storage system grid-side converter, calculates the action time of effective vectors and zero vectors, makes the current accurately track a reference value at the end of a control period, significantly reduces shaft current pulsation, and reduces the harmonic content of grid-connected current, simultaneously fast locks adjacent candidate vectors through a suboptimal voltage vector selection table, reduces the number of calculation cycles, and shortens the DC bus voltage recovery time of the grid-side converter; in the machine-side converter, the action time sequence of the reconstructed voltage vector is dynamically distributed through multi-vectors, torque and flux are convergent to the reference value without error at the end of the period, power pulsation is inhibited, the dynamic response speed and energy conversion efficiency of the flywheel energy storage are effectively improved, and the controller calculation load and system operation loss are reduced.

Description

Technical Field

[0001] This application relates to the field of energy storage control technology, and in particular to an optimization method for flywheel energy storage systems based on multi-vector model predictive control. Background Technology

[0002] In recent years, my country's installed wind power capacity has continued to break historical records. However, the inherent random fluctuations and uneven regional distribution of wind power pose severe challenges to the safe and stable operation of the new power system. With the increasing penetration rate of wind power, the uncertainty of its power output is increasingly affecting grid stability. How to alleviate the contradiction between wind power grid connection and system stability through technological innovation has become a key research focus in the energy sector.

[0003] To address this challenge, the following strategies are currently being adopted: On the grid side, strengthening grid infrastructure construction and optimizing intelligent dispatch algorithms significantly improves the grid's adaptability and anti-interference performance; on the power source side, efforts are being made to develop advanced wind power control technologies, with energy storage systems considered an effective solution to improve wind power quality and supply reliability. This source-grid coordinated optimization approach not only enhances the power system's ability to absorb high proportions of wind power and overcomes the technical bottlenecks restricting the large-scale development of wind power, but also ensures the stable operation of the grid under conditions of high wind power penetration, thereby guaranteeing the security and reliability of power supply.

[0004] The application of energy storage technology has provided crucial support for the development of wind power. By smoothing power fluctuations and participating in system frequency regulation, it effectively reduces the impact of wind power grid connection on the power system. This innovative "wind power + energy storage" model not only promotes the large-scale development and utilization of clean energy and helps solve energy and environmental problems, but also provides key technological guarantees for building a new power system, achieving a win-win situation for both economic and social benefits.

[0005] Although flywheel energy storage systems have a relatively short history of development, they have garnered sustained attention due to their fast response speed, high energy density, and long lifespan. The control of the converter in flywheel energy storage systems has also gradually become a hot topic for researchers. After the flywheel energy storage system absorbs and releases energy to the DC side, the grid-side converter converts the power between the grid and the flywheel energy storage system. Based on the background of flywheel-lithium battery hybrid energy storage for smoothing wind farm grid-connected power, different functional requirements have been proposed for the machine-side and grid-side converters of the flywheel energy storage system.

[0006] The control strategy performance of the flywheel energy storage grid-side converter directly affects the power quality of the flywheel energy storage input to the grid. The control of the grid-side converter in a flywheel energy storage system is mainly divided into four types: vector control, proportional-resonant (PR) control, converter control combined with intelligent algorithms, and repetitive control. Vector control uses a voltage-current dual-closed-loop PI controller for regulation. Because the vector control method for the grid-side converter needs to consider the voltage and current loop time constants during design, its response speed matching with the requirements of the flywheel energy storage system's grid-side converter is relatively low. PR control, or proportional-resonant control, reduces the controller complexity compared to vector control, but its control performance is significantly reduced when there are grid fluctuations. Converter control combined with intelligent algorithms mainly includes nonlinear sliding mode variable structure control and model predictive control. Nonlinear sliding mode variable structure control changes the system's structural state, causing the state variables to enter the sliding surface and undergo sliding mode motion according to a designed trajectory. Model predictive current control (MMDC) uses a mathematical model to predict the current at the next moment under the action of various switching vectors. It then evaluates the switching state pointing towards the reference value using an evaluation function and applies the result at the next moment. However, MMDC always suffers from over- and under-regulation problems in steady-state operation. Researchers aim to improve this by increasing the number of acting vectors within a control cycle. Repetitive control, by periodically accumulating the errors between the reference and feedback signals, achieves zero steady-state error tracking of AC signals. Furthermore, by setting the internal model integral period, some harmonics can be eliminated. This control method is advantageous when the sampling frequency and fundamental frequency are integer ratios; however, it often exhibits larger errors when they are not integer ratios. Summary of the Invention

[0007] This application provides an optimization method for flywheel energy storage systems based on multi-vector model predictive control. Through innovative designs such as multi-vector synthesis, rapid selection of suboptimal vectors, and dynamic time constraint reconstruction, it achieves high-precision tracking of grid-side current and suppression of turbine-side torque pulsation, providing an efficient and reliable energy storage control solution for high-penetration wind power grid connection.

[0008] To achieve the above objectives, the technical solution of this invention is as follows:

[0009] This invention provides an optimization method for a flywheel energy storage system based on multi-vector model predictive control, comprising: in the grid-side converter of the flywheel energy storage system, multi-vector model predictive current control is used to synthesize two effective voltage vectors and one zero vector in each control cycle; the action time of each vector is calculated based on the deadbeat control concept, so that the grid-connected current accurately tracks the reference value at the end of the control cycle, and suppresses... Shaft current pulsation and grid-connected current harmonics; in the flywheel energy storage system's machine-side converter, torque control is predicted through a multi-vector model. By using a predefined suboptimal vector selection table, adjacent suboptimal vectors are directly determined based on the position of the optimal vector, reducing the number of online calculation cycles, lowering the controller load, and reducing torque pulsation and power fluctuations; by dynamically adjusting the action time of the voltage vector, the sum of the effective vector action times is ensured not to exceed the control cycle, and the action time is reconstructed according to preset rules when the limit is exceeded, in order to maintain system stability.

[0010] In some possible implementations, current control is predicted using a multi-vector model, including: establishing a current dynamic equation based on the deadbeat control concept, and calculating the slope of current change under the action of each voltage vector; the current dynamic equation is expressed as:

[0011] ;

[0012] in, , They represent Always Current component in the axial direction; , These represent the slopes of the current change under the action of each voltage vector; , , These represent the duration of the voltage vector within one control cycle; the duration of each vector is calculated based on the current dynamic equation and expressed as:

[0013] ;

[0014] in, ;

[0015] The time of one control cycle, for time The difference between the current value in the axial direction and the reference value. for time The difference between the current value in the axial direction and the reference value. for Reference value of current in the axial direction. for Reference value of current in the axial direction. It is a substitution variable for complex denominators when solving a system of equations, and has no specific meaning.

[0016] In some possible implementations, the slope of the current change under the action of each voltage vector is calculated as follows:

[0017] ;

[0018] in, For the flywheel system grid-side filter inductor, This is the equivalent resistance of the power switch transistor's losses. The converter-side potential is located in the d-axis direction. The mechanical angular velocity of the flywheel motor. for Current component in the axial direction, for Current component in the axial direction, and These are the two optimal voltage vectors in model predictive control.

[0019] In some possible implementations, in the grid-side converter of the flywheel energy storage system, the duration of the two effective voltage vectors satisfies the following formula:

[0020] ;

[0021] .

[0022] In some possible implementations, torque control is predicted using a multi-vector model, including: establishing dynamic equations for torque and flux linkage based on the deadbeat control concept, and calculating the slope of change under the action of each voltage vector; the dynamic equations for torque and flux linkage are expressed as:

[0023] ;

[0024] in, express Electromagnetic torque at time t, express Three-phase winding flux linkage at any given moment; This represents the slope of torque change under the action of each voltage vector. This represents the slope of the stator flux linkage change under the action of each voltage vector; the duration of action of each vector within one control cycle is calculated based on the dynamic equations of torque and flux linkage, and is expressed as:

[0025] ;

[0026] in, ;

[0027] for The difference between the torque value and the reference value at any given time. for The difference between the flux linkage value and the reference value at any given time. This is a torque reference value. This is the reference value for magnetic flux linkage.

[0028] In some possible implementations, the method for calculating the slope of torque change under the action of each voltage vector is expressed as follows:

[0029] ;

[0030] The calculation method for the slope of the stator flux linkage change under the action of various voltage vectors is expressed as follows:

[0031] ;

[0032] in, The number of magnetic pole pairs, It is a permanent magnet flux linkage. This refers to the rotor angular velocity of the permanent magnet synchronous motor. In order to be in Voltage component in the axial direction, This is the grid-side filter inductor.

[0033] In some possible implementations, in the flywheel energy storage system's generator-side converter, the duration of the two effective voltage vectors satisfies the following formula:

[0034] ;

[0035] .

[0036] In some possible implementations, the suboptimal voltage vector selection table is generated based on the division of the Thiessen polygon region. The suboptimal vector is the adjacent vector of the optimal vector. The specific selection rules include: if the reference voltage vector is located within the Thiessen polygon region of a certain optimal vector, then the suboptimal vector is limited to two adjacent valid vectors of the optimal vector; the candidate range of suboptimal vectors is obtained directly by looking up the table, without having to traverse all possible vectors.

[0037] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:

[0038] In this embodiment of the invention, through the synergistic optimization of multi-vector model predictive current control and multi-vector model predictive torque control, a three-vector synthesis technique is employed in the grid-side converter of the flywheel energy storage system to accurately calculate the action time of the effective vector and the zero vector. This ensures that the current precisely tracks the reference value at the end of the control cycle, significantly reducing [the impact of the current loss]. The system reduces shaft current pulsation and grid-connected current harmonic content. Simultaneously, it rapidly locks adjacent candidate vectors using a suboptimal voltage vector selection table, reducing the number of calculation cycles and shortening the DC bus voltage recovery time of the grid-side converter. In the generator-side converter, the timing of the voltage vector's action is reconstructed through multi-vector dynamic allocation, ensuring that torque and flux converge to the reference value without delay at the end of the cycle, suppressing power pulsation. This effectively improves the dynamic response speed and energy conversion efficiency of flywheel energy storage, while reducing the controller's computational load and system operating losses. This provides technical support for the stable operation and long lifespan requirements of energy storage systems in high-proportion wind power grid-connected scenarios. Attached Figure Description

[0039] To more clearly illustrate the embodiments of the present invention, the accompanying drawings used in the embodiments of the present invention will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 A schematic flowchart illustrating an embodiment of an optimization method for a flywheel energy storage system based on multi-vector model predictive control provided for the implementation of this invention;

[0041] Figure 2 This is a block diagram of the grid-side converter control in an embodiment of the present invention;

[0042] Figure 3 This is a schematic diagram of the Thiessen polygon region division in an embodiment of the present invention;

[0043] Figure 4 This is a flowchart of the three-vector model predictive torque control in an embodiment of the present invention;

[0044] Figure 5 This is a schematic diagram of network-side control based on MPCC in the prior art;

[0045] Figure 6 This is a schematic diagram of network-side control based on MV-MPCC in an embodiment of the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0047] In the relevant descriptions of this embodiment, the terms "including," "containing," and "possessing" are all open terms and are generally understood to include but not be limited to; the term "at least one" is generally understood to mean one or more, where "multiple" refers to two or more; the term "at least one of the following" or similar expressions refer to any combination of these items, including any combination of single or multiple items, for example, "at least one of a, b, or c", or "at least one of a, b, and c", which can all mean: a, b, c, ab (i.e., a and b), ac, bc, or abc, where a, b, and c can be single or multiple; the symbol "A / B" is used to describe the selection relationship of associated objects, generally indicating an "or" relationship.

[0048] In the following description of the embodiments, the terminology used in the embodiments of this application is for the purpose of describing particular embodiments only and is not intended to limit the application. The singular forms "a" and "the" as used in the embodiments of this application and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise.

[0049] Those skilled in the art should understand that, in the following description of the embodiments of this application, the sequence of numbers does not imply the order of execution. Some or all steps may be executed in parallel or sequentially. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0050] Those skilled in the art will understand that the numerical ranges in the embodiments of this application should be understood to specifically disclose each intermediate value between the upper and lower limits of the range. Any stated value or intermediate value within a stated range, as well as any other stated value or each smaller range between intermediate values ​​within a range, are also included within this invention. The upper and lower limits of these smaller ranges may be independently included or excluded from the range.

[0051] Unless otherwise stated, the technical / scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. While this application describes only preferred methods and materials, any methods and materials similar or equivalent to those described herein may be used in the implementation or testing of this application. All references to this specification are incorporated by way of citation to disclose and describe the methods and / or materials associated with those references. In the event of any conflict with any incorporated reference, the content of this specification shall prevail.

[0052] To illustrate the technical solution of the present invention, specific embodiments are described below.

[0053] Figure 1A schematic flowchart illustrating an embodiment of the flywheel energy storage system optimization method based on multi-vector model predictive control provided for the implementation of this invention is shown below. Figure 1 As shown, the above-mentioned flywheel energy storage system optimization method based on multi-vector model predictive control can include:

[0054] In S101, the grid-side converter of the flywheel energy storage system uses multi-vector model predictive current control to synthesize two effective voltage vectors and one zero vector in each control cycle. Based on the deadbeat control concept, the action time of each vector is calculated, ensuring that the grid-connected current accurately tracks the reference value at the end of the control cycle and suppresses... Shaft current pulsation and grid-connected current harmonics;

[0055] This invention improves upon the traditional single-vector model predictive current control scheme by proposing Multi-Vector Model Prediction Current Control (MV-MPCC), which enables the current to reach the reference value at the end of the control cycle, reducing the risks associated with actual control. The shaft current pulsation reduces the high-order harmonic content in the grid-connected three-phase current.

[0056] MV-MPCC synthesizes two effective vectors and one zero vector in each control cycle. Based on the deadbeat control concept, it accurately calculates the action time of each vector, thereby ensuring that the current accurately tracks the reference value at the end of the control cycle. This effectively suppresses the problems inherent in traditional single-vector control. The problems of shaft current pulsation and grid-connected current harmonics are overcome by using omnidirectional adjustable voltage vector synthesis technology to overcome over-regulation and under-regulation phenomena. While maintaining the dynamic response speed of the system, the steady-state performance is significantly improved, and high-precision grid-connected control of the grid-side converter is realized.

[0057] In some embodiments, step S101 above, predicting current control using a multi-vector model, includes:

[0058] Based on the concept of deadbeat control, a current dynamic equation is established to calculate the slope of current change under the action of each voltage vector; the current dynamic equation is expressed as:

[0059] ;

[0060] in, , They represent Always Current component in the axial direction; , These represent the slopes of the current change under the action of each voltage vector; , , These represent the duration of the voltage vector's action within one control cycle;

[0061] The duration of action of each vector is calculated based on the current dynamic equation, and expressed as follows:

[0062] ;

[0063] in, ;

[0064] The time of one control cycle, for time The difference between the current value in the axial direction and the reference value. for time The difference between the current value in the axial direction and the reference value. for Reference value of current in the axial direction. for Reference value of current in the axial direction. It is a substitution variable for complex denominators when solving a system of equations, and has no specific meaning.

[0065] In some embodiments, the current change slope (current change slope in the d-axis direction) under the action of each voltage vector is calculated as follows:

[0066] ;

[0067] in, For the flywheel system grid-side filter inductor, This is the equivalent resistance of the power switch transistor's losses. The converter-side potential is located in the d-axis direction. The mechanical angular velocity of the flywheel motor. for Current component in the axial direction, for Current component in the axial direction, and These are the two optimal voltage vectors in model predictive control.

[0068] The duration of action of the two effective vectors must satisfy the conditions given by the following formula:

[0069] ;

[0070] In addition, the sum of the action times of the two valid vectors must be limited to no more than one sampling period. Therefore, additional restrictions need to be added at the end of each control cycle, and a judgment needs to be made. If this condition is met, the effective vector action time is reconstructed according to the following formula:

[0071] ;

[0072] For example, the control block diagram of the grid-side converter control strategy for a flywheel energy storage system based on MV-MPCC is as follows: Figure 2 As shown. The basic control strategy is to input the eight voltage vectors that the grid-side converter can act upon into the grid-side converter model, and then, based on the evaluation function design, calculate the voltage vector that minimizes the evaluation function value to act at the next time step. (See figure.) and These represent the DC-side voltage of the converter and its reference value, respectively. and They represent and Shaft current reference value.

[0073] S102, in the flywheel energy storage system machine-side converter, predictive torque control is achieved through multi-vector model. By using a predefined suboptimal vector selection table, adjacent suboptimal vectors are directly determined based on the position of the optimal vector, reducing the number of online calculation cycles, lowering the controller load, and reducing torque ripple and power fluctuations.

[0074] It should be noted that, since the machine-side converter control strategy based on Model Predictive Torque Control (MPTC) cannot theoretically reach the reference value at the end of each control cycle during the optimization process, this invention designs a power control strategy for the flywheel energy storage system based on Multi-Vector Model Predictive Torque Control (MV-MPTC), which enables the torque and flux linkage to reach the reference value at the end of the control cycle, reduces torque ripple in actual control, and thus reduces power ripple during the power throughput process of the flywheel energy storage system.

[0075] In some embodiments, step S102 above, predicting torque control using a multi-vector model, includes:

[0076] Based on the concept of deadbeat control, dynamic equations for torque and flux linkage are established, and the slope of change under the action of each voltage vector is calculated; the dynamic equations for torque and flux linkage are expressed as follows:

[0077] ;

[0078] in, express Electromagnetic torque at time t, express Three-phase winding flux linkage at any given moment; This represents the slope of torque change under the action of each voltage vector. This represents the slope of the stator flux linkage change under the action of various voltage vectors;

[0079] The duration of action of each vector within one control cycle is calculated based on the dynamic equations of torque and flux linkage, and is expressed as follows:

[0080] ;

[0081] in, ;

[0082] for The difference between the torque value and the reference value at any given time. for The difference between the flux linkage value and the reference value at any given time. This is a torque reference value. This is the reference value for magnetic flux linkage.

[0083] In some embodiments, the method for calculating the slope of torque change under the action of each voltage vector is expressed as follows:

[0084] ;

[0085] The calculation method for the slope of the stator flux linkage change under the action of various voltage vectors is expressed as follows:

[0086] ;

[0087] in, The number of magnetic pole pairs, It is a permanent magnet flux linkage. This refers to the rotor angular velocity of the permanent magnet synchronous motor. In order to be in Voltage component in the axial direction, This is the grid-side filter inductor.

[0088] In some embodiments, the effective vector action times need to satisfy the following formula:

[0089] ;

[0090] In addition, the sum of the effects of the effective vectors must be limited to no more than one control cycle. Otherwise, the effective voltage vector duration is reconstructed according to the following formula:

[0091] .

[0092] In some embodiments, to reduce the number of loops when MT-MPTC selects a suboptimal voltage vector and reduce the computational load, the present invention optimizes the process by introducing a suboptimal voltage vector selection table, which is then read offline when selecting a suboptimal voltage vector.

[0093] In the process of selecting the suboptimal vector for MP-MPTC control, the suboptimal voltage vector selection table can be generated based on the Thiessen polygon region division. The suboptimal vector is the adjacent vector of the optimal vector. The specific selection rules include:

[0094] If the reference voltage vector is located within the Thiessen polygon region of an optimal vector, then the suboptimal vector is limited to two adjacent valid vectors of that optimal vector;

[0095] The range of candidate suboptimal vectors can be obtained directly by looking up a table, without having to traverse all possible vectors.

[0096] Figure 3 See the diagram showing the division of the Thiessen polygon region. Figure 3 As shown, in model predictive control, when the reference voltage vector... lie in When the region is selected, the system will choose As the optimal vector, the suboptimal vector must then appear in the adjacent vectors. and Based on this principle, the calculation process can be significantly optimized by establishing a suboptimal vector selection table, as shown in Table 1. Table 1 is the suboptimal voltage vector selection table. That is, after determining the optimal vector, only the two adjacent valid vectors need to be examined to determine the suboptimal vector, without having to re-traverse all possible vectors.

[0097] Table 1:

[0098]

[0099] Figure 4 The flowchart for the three-vector model predictive torque control in this embodiment of the invention is shown below. Figure 4 As shown in the improved control block diagram, the optimization method of this invention obtains the candidate range of suboptimal vectors by directly looking up a table, effectively reducing the number of loop calculations in the control algorithm. This not only improves the controller's computational efficiency and shortens the calculation delay of the control signal, but also enhances the system's dynamic response capability and operational stability. This intelligent selection mechanism based on spatial vector region partitioning significantly improves the real-time performance of the three-vector model predictive control system while ensuring control accuracy.

[0100] S103 ensures that the sum of the effective vector action times does not exceed the control cycle by dynamically adjusting the action time of the voltage vector, and reconstructs the action time according to preset rules when the limit is exceeded, so as to maintain system stability.

[0101] Specifically, through unified dynamic time constraints and reconfiguration rules, deep collaboration is established between the grid-side converter (precise current tracking) and the machine-side converter (torque ripple suppression). This enables rapid stabilization of the DC bus voltage on the grid side, providing a low-harmonic energy transmission environment for the machine side. The machine side reduces mechanical losses, which in turn contributes to power quality optimization on the grid side. A suboptimal vector selection mechanism enhances the real-time performance of the hybrid energy storage system. By using the effective voltage vector's duration constraint formulas in both the grid-side and machine-side converters, and through preset constraints and reconfiguration rules, the duration of the effective vector is dynamically adjusted to ensure the total duration is strictly limited within the control cycle, thereby maintaining system dynamic stability and avoiding overload or loss-of-synchronization risks.

[0102] In some embodiments, the flywheel energy storage system optimization method based on multi-vector model predictive control of the present invention can be further adapted to flywheel-lithium battery hybrid energy storage systems.

[0103] Understandably, lithium-ion batteries have high charge / discharge efficiency, low loss, and considerable energy storage capacity and density, but they are slightly inferior in response speed. Applying the method of this invention to a flywheel-lithium battery hybrid energy storage system can significantly improve the dynamic performance and energy efficiency of hybrid energy storage in smoothing power fluctuations in wind farm grid connection. Specific implementation methods include:

[0104] (1) Implement a stratified power smoothing mechanism for high-frequency and low-frequency power:

[0105] In hybrid energy storage scenarios, flywheel energy storage is responsible for responding to high-frequency, short-term power fluctuations, while lithium battery energy storage handles low-frequency, long-term energy throughput. This invention achieves millisecond-level power response for flywheel energy storage through MV-MPCC control of the grid-side converter, accurately tracking high-frequency power commands. Simultaneously, MV-MPTC control of the turbine-side converter ensures stable flywheel speed by suppressing torque ripple, avoiding mechanical losses caused by frequent charge-discharge switching. Lithium battery energy storage performs slow dynamic compensation based on the smoothed power curve of the flywheel. The two systems achieve power-energy decoupling optimization through a hierarchical control architecture, further reducing the power fluctuation rate of wind farm grid connection.

[0106] (2) Achieving coordinated control across multiple time scales:

[0107] In a hybrid system, the grid-side converter generates power allocation commands for the flywheel and lithium battery based on the real-time power deviation of the wind farm. The MV-MPCC algorithm of this invention rapidly adjusts the flywheel output using a three-vector synthesis technique and utilizes a suboptimal voltage vector selection table to match dynamic demands in real time, ensuring the flywheel's dominant role in the hybrid system. Simultaneously, dynamic constraints ensure conflict-free power switching between the flywheel and lithium battery. For example, when a persistent power deficit is detected, the system automatically reduces the flywheel output ratio, smoothly transitioning to lithium battery dominance, avoiding power gaps caused by control delays.

[0108] (3) Achieve adaptive parameter adjustment and lifetime optimization:

[0109] To address the lifespan degradation characteristics of flywheels and lithium batteries, the control strategy of this invention incorporates an adaptive parameter adjustment module. For example, when the flywheel speed approaches its limit, the flux linkage reference value of the MV-MPTC is dynamically adjusted to limit the flywheel's charge-discharge depth, while simultaneously increasing the compensation weight of the lithium battery, thereby extending the flywheel's mechanical lifespan. Furthermore, the low-harmonic output of the grid-side converter reduces polarization losses in the lithium battery caused by high-frequency harmonics, further improving the overall energy efficiency of the hybrid system.

[0110] In this embodiment of the invention, through the synergistic optimization of multi-vector model predictive current control and multi-vector model predictive torque control, a three-vector synthesis technique is employed in the grid-side converter of the flywheel energy storage system to accurately calculate the action time of the effective vector and the zero vector. This ensures that the current precisely tracks the reference value at the end of the control cycle, significantly reducing [the impact of the current loss]. The system reduces shaft current pulsation and grid-connected current harmonic content. Simultaneously, it rapidly locks adjacent candidate vectors using a suboptimal voltage vector selection table, reducing the number of calculation cycles and shortening the DC bus voltage recovery time of the grid-side converter. In the generator-side converter, the timing of the voltage vector's action is reconstructed through multi-vector dynamic allocation, ensuring that torque and flux converge to the reference value without delay at the end of the cycle, suppressing power pulsation. This effectively improves the dynamic response speed and energy conversion efficiency of flywheel energy storage, while reducing the controller's computational load and system operating losses. This provides technical support for the stable operation and long lifespan requirements of energy storage systems in high-proportion wind power grid-connected scenarios.

[0111] Instance data verification:

[0112] In the grid-side converter control of the flywheel energy storage system based on MV-MPCC, the charging and discharging switching of the flywheel energy storage system is simulated. The simulation time is 2s. In the first 1s, the flywheel energy storage system absorbs energy from the grid side at rated power. At 1s, it switches to the flywheel energy storage system releasing energy to the grid side at rated power.

[0113] Figure 5 This is a schematic diagram of network-side control based on MPCC in the prior art. Figure 6 This is a schematic diagram of grid-side control based on MV-MPCC in an embodiment of the present invention. During the process of switching from the flywheel energy storage system absorbing energy at rated power to releasing energy at rated power, the direct-axis current of the grid-side converter switches its flow direction at the rated value, while the quadrature-axis current remains at 0. For example... Figure 5 As shown, the cross-axis and quadrature-axis currents of the grid-side converter control strategy based on MPCC have a ripple size of approximately 0.08 pu, which is relatively large. Figure 6As shown, under the grid-side converter control strategy based on MV-MPPC, the ripple magnitudes of the direct-axis and quadrature-axis currents are approximately 0.04 (pu), which is significantly reduced compared to the grid-side dq-axis currents based on MPCC. MPCC exhibits larger current ripples than MV-MPCC, impacting the power quality of the grid-side converter's input power system. Simultaneously, under the grid-side converter control based on MV-MPCC, the DC bus voltage recovery time is shorter, requiring only 0.04s to recover and maintain stability, an improvement of 0.01s compared to the MPCC-based grid-side converter control method.

[0114] In the control of the generator-side converter of the flywheel energy storage system based on TV-MPTC, the simulation time is 2 seconds, the initial speed is 5000 r / min, and the simulation simulates the flywheel energy storage system absorbing changing power. In the first second, the flywheel energy storage absorbs power on a ramp-up, and in the 1 second, it switches to absorbing constant power. Consistent with the flywheel energy storage charging and discharging switching condition, the power control strategies of the generator-side converter of the flywheel energy storage system based on DTC, MPTC, and MV-MPTC control are simulated respectively. The power ripple and torque ripple under the three control strategies are compared, and corresponding ripple evaluation criteria are set to compare and analyze the power ripple during the charging and discharging process of the flywheel energy storage system. The simulation results are shown in Table 2.

[0115] Table 2:

[0116]

[0117] As shown in Table 2, on the one hand, the TV-MPTC model predictive torque control solves the over-regulation problem of traditional single-vector model predictive torque control by reducing the torque ripple of the flywheel motor, thereby reducing power fluctuations during the charging and discharging process of the flywheel energy storage system. On the other hand, by introducing a suboptimal voltage vector switching table, the computational complexity of the power control strategy for the generator-side converter of the flywheel energy storage system is reduced, the controller load is decreased, and the risk of system instability caused by the controller not being able to complete the calculations within a single control cycle is avoided.

[0118] The various embodiments in this specification are described in a progressive manner. For the same or similar parts between the various embodiments, please refer to each other. Each embodiment focuses on describing the differences from other embodiments.

[0119] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of this application.

Claims

1. An optimization method for a flywheel energy storage system based on multi-vector model predictive control, characterized in that, include: In the grid-side converter of the flywheel energy storage system, multi-vector model predictive current control is used to synthesize two effective voltage vectors and one zero vector in each control cycle. Based on the deadbeat control concept, the action time of each vector is calculated, so that the grid-connected current accurately tracks the reference value at the end of the control cycle, suppressing... Shaft current pulsation and grid-connected current harmonics; The current control predicted by the multi-vector model includes: Based on the concept of deadbeat control, a current dynamic equation is established to calculate the slope of current change under the action of each voltage vector; the current dynamic equation is expressed as: ； in, , They represent Always Current component in the axial direction; , These represent the slopes of the current change under the action of each voltage vector; , , These represent the duration of the voltage vector's action within one control cycle; The duration of action of each vector is calculated based on the current dynamic equation, and expressed as follows: ； in, ; The time of one control cycle, for time The difference between the current value in the axial direction and the reference value. for time The difference between the current value in the axial direction and the reference value. for Reference value of current in the axial direction. for Reference value of current in the axial direction. Substitute variables for complex denominators when solving a system of equations; In the flywheel energy storage system's generator-side converter, multi-vector model predictive torque control is implemented. This involves using a predefined suboptimal vector selection table to directly determine adjacent suboptimal vectors based on the position of the optimal vector, reducing the number of online calculation cycles, lowering the controller load, and minimizing torque ripple and power fluctuations. The multi-vector model predictive torque control includes: Based on the concept of deadbeat control, dynamic equations for torque and flux linkage are established, and the slope of change under the action of each voltage vector is calculated; the dynamic equations for torque and flux linkage are expressed as follows: ； in, express Electromagnetic torque at time t, express Three-phase winding flux linkage at any given moment; This represents the slope of torque change under the action of each voltage vector. This represents the slope of the stator flux linkage change under the action of various voltage vectors; The duration of action of each vector within one control cycle is calculated based on the dynamic equations of torque and flux linkage, and is expressed as follows: ； in, ; for The difference between the torque value and the reference value at any given time. for The difference between the flux linkage value and the reference value at any given time. This is a torque reference value. This is a reference value for magnetic flux linkage; By dynamically adjusting the action time of the voltage vector, the sum of the effective vector action times is ensured not to exceed the control cycle, and the action time is reconstructed according to preset rules when the limit is exceeded, so as to maintain system stability.

2. The method according to claim 1, characterized in that, The calculation method for the slope of the current change under the action of each voltage vector is expressed as follows: ； in, For the flywheel system grid-side filter inductor, This is the equivalent resistance of the power switch transistor's losses. The converter-side potential is located in the d-axis direction. The mechanical angular velocity of the flywheel motor. for Current component in the axial direction, for Current component in the axial direction, and These are the two optimal voltage vectors in model predictive control.

3. The method according to claim 2, characterized in that, In the grid-side converter of the flywheel energy storage system, the duration of action of the two effective voltage vectors satisfies the following formula: ； 。 4. The method according to claim 3, characterized in that, The method for calculating the slope of torque change under the action of each voltage vector is expressed as follows: ； The calculation method for the slope of the stator flux linkage change under the action of various voltage vectors is expressed as follows: ； in, The number of magnetic pole pairs, It is a permanent magnet flux linkage. This refers to the rotor angular velocity of the permanent magnet synchronous motor. In order to be in Voltage component in the axial direction, This is the grid-side filter inductor.

5. The method according to claim 4, characterized in that, In the generator-side converter of a flywheel energy storage system, the duration of action of the two effective voltage vectors satisfies the following formula: ； 。 6. The method according to claim 5, characterized in that, The suboptimal vector selection table is generated based on the Thiessen polygon region division, and the suboptimal vector is the adjacent vector of the optimal vector. The specific selection rules include: If the reference voltage vector is located within the Thiessen polygon region of an optimal vector, then the suboptimal vector is limited to two adjacent valid vectors of that optimal vector; The range of candidate suboptimal vectors can be obtained directly by looking up a table, without having to traverse all possible vectors.

Images