Dual-mode control method and system suitable for network construction type energy storage converter

Abstract

The invention relates to a dual-mode control method and system suitable for a network construction type energy storage converter, and the method comprises the steps: building a power and virtual impedance dynamic mapping model, and achieving the dynamic mapping between the power and virtual impedance through building a mathematical relation model between the power and virtual impedance; a state prediction and compensation mechanism is introduced, a feedforward compensation link based on a system state prediction algorithm is designed, transient behaviors possibly occurring in the switching process are subjected to prospective suppression, a self-adaptive virtual impedance adjustment strategy is designed, and dual-mode seamless switching control is achieved. According to the invention, efficient and stable switching of the energy storage converter in the island and grid-connected operation states is ensured. Key technical support is provided for a smart power grid, and the flexibility and reliability of a power system are remarkably improved.

Description

Technical Field

[0001] This invention relates to the field of power electronics and automation control technology, and in particular to a dual-mode control method and system applicable to grid-type energy storage converters. Background Technology

[0002] With the increasing global emphasis on renewable energy utilization and the development of distributed energy systems, the importance of power storage converters (PCS), as key components connecting energy storage devices to the power grid, is becoming increasingly prominent. Power storage converters not only need to efficiently convert energy under normal conditions, but also need to have the ability to flexibly switch between voltage source mode (VSG) and current source mode (CSG) to adapt to different application scenarios, such as microgrids, smart grids, and off-grid power systems.

[0003] However, existing energy storage converter control methods have some limitations and challenges:

[0004] Stability issues during mode switching: During the switching between VSG and CSG modes, transient instability can easily occur due to changes in the dynamic characteristics of the system and load conditions, affecting power quality and system stability.

[0005] Virtual impedance design is not flexible enough: In traditional methods, the virtual impedance value is usually fixed or set based on experience, which makes it difficult to adapt to complex operating environments and changing working conditions, thus limiting the performance optimization space of energy storage converters.

[0006] Low power regulation efficiency: The lack of an effective mechanism in the existing technology to adjust the power output in real time results in low energy conversion efficiency of the energy storage converter under different operating modes.

[0007] Insufficient state prediction capability: Traditional control strategies lack forward-looking countermeasures for potential future load changes or changes in power grid conditions, which reduces the system's response speed and adaptability. Summary of the Invention

[0008] The purpose of this application is to provide a dual-mode control method and system for grid-type energy storage converters, which provides key technical support for smart grids and significantly improves the flexibility and reliability of power systems.

[0009] To achieve the above objectives, this application provides the following technical solution:

[0010] In a first aspect, embodiments of this application provide a dual-mode control method applicable to grid-type energy storage converters, the specific steps of which are as follows, characterized in that:

[0011] Step 1: Construct a dynamic mapping model between power and virtual impedance

[0012] By establishing a mathematical relationship model between power and virtual impedance, dynamic mapping between the two can be achieved;

[0013] Step 2: Introduce state prediction and compensation mechanisms

[0014] A feedforward compensation stage based on a system state prediction algorithm is designed to proactively suppress transient behaviors that may occur during the switching process. By real-time monitoring of key state variables such as voltage, current, and power, the dynamic response trend at the moment of switching is predicted, and a compensation signal is generated to correct the reference value.

[0015] Step 3: Design an adaptive virtual impedance adjustment strategy

[0016] The virtual impedance parameters are dynamically adjusted based on the changes in the system's equivalent impedance and the level of power disturbance. The virtual impedance value is optimized in real time through the impedance regulator to adapt to different operating conditions and effectively suppress transient oscillations and power fluctuations caused by mode switching.

[0017] Step 4: Achieve seamless switching control between dual modes

[0018] The dynamic mapping model, state prediction compensation, and adaptive impedance regulation mechanism are integrated into a complete seamless switching control architecture. The switching logic is executed by the embedded control unit, and the control parameters are adjusted in real time to ensure a smooth and fast transition between voltage source mode and current source mode, which significantly improves the dynamic stability of the energy storage converter in islanded and grid-connected operation.

[0019] The specific steps in step 1 for constructing the dynamic mapping model between power and virtual impedance are as follows:

[0020] Step 1.1 Determine the model input and output variables

[0021] Input variables:

[0022] The real-time collected system power parameters include: active power output P of the energy storage converter, reactive power output Q of the energy storage converter, and mode switching command signal S;

[0023] The energy storage converter outputs active power P, and the three-phase current signal i is acquired by a three-phase current sensor. a i b i c The three-phase voltage signal u collected by the voltage sensor a u b u c Calculate using the following formula:

[0024] ,

[0025] Where T is the acquisition time, t is the time variable, and the reactive power Q output by the energy storage converter is calculated using the following formula:

[0026] ,

[0027] Mode switching command signal S;

[0028] Output variables include: virtual resistance R v Virtual inductance L v ;

[0029] Step 1.2 Establish a mathematical model of power virtual impedance.

[0030] The mathematical relationship is constructed using a piecewise linear mapping function, as shown in the following formula:

[0031] When voltage source mode S=0, the formula is:

[0032] ,

[0033] ,

[0034] Where, k R1 The ratio of active power to virtual resistance is given by the grid resistance R. g Adjustment, P rated R is the rated active power of the converter. v0 As a virtual resistance reference value, the grid resistance R is taken. g 15%; k L1 The ratio of reactive power to virtual inductance is calculated according to... Tuning, Q rated The rated reactive power of the converter; L v0 As the virtual inductance reference value, the grid inductance L is taken. g 15%;

[0035] When the current source mode S=1, the formula is:

[0036] ,

[0037] ,

[0038] Where, k R2 Take k R1 1.3 times; R v1 Take the load resistor R l 10%; k L2 Take k L1 1.5 times; L v1 Take the load inductance L l 10%;

[0039] A smooth transition is achieved using linear interpolation. When the switching time is t0, the transition time is T.tran Then at any time t ∈ [t0,t0+T tran The virtual impedance of ] is:

[0040] ,

[0041] ,

[0042] in, This is represented as a virtual resistance in voltage source mode. This is represented as a virtual resistance in current source mode. Represented as a virtual inductance in voltage source mode. Represented as a virtual inductance in current source mode, this interpolation process ensures that R v (t), L v (t) is continuously differentiable in the transition section, avoiding abrupt changes in the reference signal.

[0043] The state prediction and compensation mechanism introduced in step 2 is as follows:

[0044] Step 2.1 Determine the input variables and prediction target for state prediction

[0045] Input variables include:

[0046] Real-time virtual impedance output from step 1: R v (t), L v (t);

[0047] Instantaneous value of three-phase output voltage of converter: u a (t), u b (t), u c (t), u in the dq coordinate system of three-phase voltage conversion d (t), u q (t);

[0048] Instantaneous value of three-phase output current of converter: i a (t), i b (t), i c (t), i in the dq coordinate system of three-phase current conversion d (t), i q (t);

[0049] Mode switching command signal S: includes 0 = voltage source mode, 1 = current source mode; advance T pred Triggering the prediction algorithm, T pred For prediction time windows;

[0050] The output variables include:

[0051] After switching T predThe state variables at any given time include:

[0052] Predicted voltage value: , ;

[0053] Predicted current value: , ;

[0054] Step 2.2. Design a state prediction algorithm based on recursive least squares.

[0055] The recursive least squares (RLS) algorithm is used to perform short-term prediction of state variables, and historical data is used to fit the dynamic trend. The formula derivation is as follows:

[0056] Step 2.2.1: Construct the observation matrix of state variables

[0057] Define the state vector in the dq coordinate system. According to the power and virtual impedance model in step 1, the state variables satisfy the linear time-varying equation:

[0058] ,

[0059] in, ; express The state vector at time t, express The duty cycle signal of the current converter at any given time; Let A(t) be the observation noise vector; A(t) is a 4×4 state transition matrix, derived from R. v (t), L v (t) and the grid angular frequency w are determined as follows:

[0060] ,

[0061] B(t) is a 4×2 control input matrix, which simplifies to:

[0062] ,

[0063] Among them, U dc T is the DC side voltage of the converter. s The sampling period;

[0064] Step 2.2.2 Recursive calculation using the RLS algorithm

[0065] The goal is to use historical observations x(t), x(tT) s ), ..., x(t-nT) s Given n=50, minimize the prediction error, update the state transition matrix A(t), and predict t+T. pred The state at time: decomposed to obtain Predicted voltage at time , With predicted current , ;

[0066] Step 2.3 Generate feedforward compensation signal and correct reference value

[0067] Step 2.3.1: Calculate the transient error.

[0068] Compare the predicted state with the target steady-state value , , , The error that needs to be compensated is obtained. , , , :

[0069] Voltage error: , ;

[0070] Current error: , ;

[0071] Step 2.3.2: Design the compensation signal transfer function

[0072] To avoid overcompensation leading to new fluctuations, a first-order low-pass filter is used to design the compensation signal. , :

[0073] ,

[0074] ,

[0075] in, , , , , , for , , , The result of the transformation from the time domain to the frequency domain, K u K is the voltage compensation coefficient. i This is the current compensation coefficient; , This is the filtering time constant.

[0076] The adaptive virtual impedance adjustment strategy designed in step 3 is as follows:

[0077] Step 3.1: Define the impedance correction factor

[0078] When grid impedance changes abruptly during grid connection or when load switching occurs during islanding, it is necessary to adjust the tracking sensitivity of the virtual impedance and define an impedance correction factor K. Z (t):

[0079] ,

[0080] Among them, Z eq (t) represents the system's real-time equivalent impedance. The system's rated equivalent impedance is given by the converter's rated power S. rated With rated voltage U rated calculate:

[0081] ,

[0082] Step 3.2: Generate perturbation coefficients

[0083] , The disturbance coefficient K is defined to reflect the intensity of power fluctuations. dist (t):

[0084] ,

[0085] in, The active power disturbance intensity, The active power disturbance threshold. The reactive power disturbance intensity. The reactive power disturbance threshold. , K is the perturbation threshold. dist (t) The piecewise function is used to classify the intensity of weak disturbances, normal adjustment and strong disturbances, so as to avoid excessive adjustment when the disturbance is small and cause new oscillations.

[0086] Step 3.3: Pattern Differentiation Modulation Law

[0087] According to S mode (t) Differentiate between VSG / CSG modes and design PI control laws with different control targets:

[0088] Step 3.3.1 Voltage Source Mode S mode (t)=0

[0089] Virtual resistance R v (t) Regulation law: To address voltage loss deviations caused by active power fluctuations, a regulation law is introduced. PI correction:

[0090] ,

[0091] in, The reference value for the virtual resistor is the output of step 1. for The result of converting from the frequency domain to the time domain For voltage error proportionality coefficient, The voltage error integral coefficient is used to eliminate steady-state voltage error via a PI converter; when When R > 0, increase R v Compensate for voltage loss; conversely, reduce R. v ;

[0092] Virtual inductance L v (t) Regulation law: To address the phase deviation caused by reactive power fluctuations, a regulation law is introduced. PI correction:

[0093] ,

[0094] in, The reference value for the virtual inductance is the output of step 1. For current error proportionality coefficient, The integral coefficient for current error;

[0095] Step 3.3.2 Current Source Mode S mode (t)=1

[0096] Virtual resistance R v (t) Regulation law: To address the current tracking deviation caused by changes in load impedance, a regulation law is introduced. PI correction:

[0097] ,

[0098] in, For current error proportionality coefficient, For the current error integral coefficient; when When R > 0, decrease R v Reduce current resistance; conversely, increase R. v ;

[0099] Virtual inductance L v (t) Regulation law: fixed at 1.1 times that of the previous generation, enhancing the current filtering effect:

[0100] .

[0101] Step 4, which implements seamless switching control between the two modes, specifically involves:

[0102] Upon receiving a mode switching command, a preset transition control procedure is initiated. Within the window period from the moment of switching to the end of the transition, the basic reference signal is used as the baseline frame, and the compensation signal is synchronously superimposed. and Superimposed on each and Up, obtain , , and The calculation formula is expressed as follows; simultaneously, if a sudden impedance change or power fluctuation exceeding the threshold is detected, adaptive adjustment parameters are immediately activated to optimize the virtual impedance in real time and correct the R output in step 1. v L v ;

[0103] The voltage reference signal of the voltage source is calculated from the virtual impedance and current feedback, that is:

[0104] ,

[0105] The current reference signal of the current source is calculated from the virtual impedance and voltage feedback, that is:

[0106] ,

[0107] in, I is the rated voltage, and I is the output current.

[0108] Secondly, embodiments of this application provide a system suitable for a dual-mode control method of a grid-type energy storage converter, comprising a sensing and acquisition module, a control module, an execution and drive module, and a monitoring and feedback module.

[0109] The sensing and acquisition module is responsible for collecting physical quantities of the system in real time, including the three-phase voltage and three-phase current signals output by the energy storage converter, and calculating the active and reactive power using preset formulas. At the same time, it receives mode switching command signals generated by the system controller and transmits all collected and calculated data to the core control module in a synchronized manner, providing basic data support for subsequent modeling and control.

[0110] The core control module integrates three major functional units:

[0111] The dynamic mapping unit performs dynamic correlation calculations between power and virtual impedance, and outputs the basic virtual resistance and virtual inductance through a piecewise linear mapping function and a linear interpolation algorithm.

[0112] The predictive compensation unit includes a corresponding state prediction and compensation mechanism. Based on the recursive least squares algorithm, it uses historical voltage and current data to predict the system state after the switch, generates a feedforward compensation signal, and corrects the voltage and current reference values.

[0113] The adaptive adjustment unit incorporates an adaptive virtual impedance adjustment strategy, calculates the impedance correction coefficient and disturbance coefficient, and dynamically corrects R through a mode-differentiated PI control law. v Lv Parameters to adapt to system impedance changes and power disturbances;

[0114] The execution drive module receives instructions from the control module, undertakes the dual-mode seamless switching control logic, receives the corrected voltage reference value or current reference value after superimposed compensation signal, and generates the corresponding PWM drive signal; controls the on and off of the switching devices inside the converter to realize the physical switching between voltage source mode and current source mode.

[0115] The monitoring and feedback module monitors the system's operating status in real time and constructs a closed-loop regulation mechanism. Its monitoring objects are consistent with the input requirements, tracking the dynamic changes of the converter's output voltage, current, and power, comparing the predicted values ​​with the actual values, and calculating the transient error. It also detects sudden changes in the system's equivalent impedance and abnormal load switching conditions, triggering the core control module to execute adaptive regulation logic to ensure the system's dynamic stability.

[0116] Compared with existing technologies, this invention has significant advantages. 1. By introducing an adaptive virtual impedance mapping mechanism, this invention achieves a smoother and more stable transition during the switching between VSG and CSG modes, effectively reducing the occurrence of transient instability. This not only improves the overall stability of the system but also enhances the quality of the output power, meeting the demand for a highly reliable power supply.

[0117] 2. This invention can automatically adjust the virtual impedance value according to real-time operating conditions, enabling the energy storage converter to maintain optimal performance even in complex and ever-changing working environments. This adaptive characteristic greatly enhances the system's adaptability and flexibility, making it suitable for various application scenarios such as microgrids and smart grids.

[0118] 3. The state prediction module included in this invention can identify potential load changes or power grid condition variations in advance and adjust the control strategy accordingly to quickly respond to changes in the external environment. This not only speeds up the system's response time but also improves its ability to cope with emergencies. Attached Figure Description

[0119] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0120] Figure 1 This is a flowchart of the present invention;

[0121] Figure 2 This is a flowchart of the RLS algorithm of the present invention. Detailed Implementation

[0122] The technical solutions of the embodiments of this application will now be described with reference to the accompanying drawings. It should be noted that similar reference numerals and letters in the following drawings indicate similar items; therefore, once an item is defined in one drawing, it does not need to be further defined and explained in subsequent drawings.

[0123] The terms “comprising,” “including,” or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase “comprising one…” does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0124] The terms “first,” “second,” etc., are used only to distinguish one entity or operation from another, and should not be construed as indicating or implying relative importance, nor as requiring or implying any such actual relationship or order between these entities or operations.

[0125] like Figure 1 and Figure 2 As shown, this embodiment is achieved through the following technical solution: a dual-mode control method and system for grid-type energy storage converters, aiming to solve the instability problem that occurs in energy storage systems during mode switching in existing technologies. By introducing an adaptive virtual impedance mechanism, this invention achieves a smooth and stable transition between VSG and CSG modes, significantly improving system stability and power quality. Simultaneously, this method can automatically adjust parameters according to real-time operating conditions, enhancing the system's adaptability and flexibility to complex environments, and improving energy conversion efficiency and state prediction capabilities. This provides new ideas for the design of energy storage converters and promotes the development of smart grid technology. The invention flowchart is shown below. Figure 1 As shown, the steps of the present invention will be described in detail below.

[0126] To address the aforementioned problems, this invention proposes a dual-mode control method suitable for grid-type energy storage converters, with the specific steps as follows:

[0127] Step 1: Construct a dynamic mapping model between power and virtual impedance

[0128] By establishing a mathematical model relating power and virtual impedance, a dynamic mapping between the two is achieved. This model ensures the continuity and smooth transition of the output voltage and current reference signals during mode switching, avoiding transient shocks caused by sudden changes in reference values.

[0129] By establishing power including active power P, reactive power Q, and virtual resistance. Virtual inductance The dynamic correlation model addresses the output voltage reference signal issue in dual-mode grid-type energy storage converters during the switching between voltage source mode (VSG) and current source mode (CSG). Current reference signal The problem of discontinuity.

[0130] Step 1.1 Determine the model input and output variables

[0131] Input variables:

[0132] The real-time collected system power parameters include: active power P output by the energy storage converter, reactive power Q output by the energy storage converter, and mode switching command signal S.

[0133] The energy storage converter outputs active power P, and the three-phase current signal i is acquired by a three-phase current sensor. a i b i c The three-phase voltage signal u collected by the voltage sensor a u b u c Calculate using the following formula:

[0134]

[0135] Where T is the acquisition time, t is the time variable, and the reactive power Q output by the energy storage converter is calculated using the following formula:

[0136]

[0137] Mode switching command signal S: generated by the system controller according to the operating conditions, including 0=voltage source mode and 1=current source mode.

[0138] Output variables include: virtual resistance R v Virtual inductance L v。

[0139] Step 1.2 Establish a mathematical model of power virtual impedance.

[0140] The mathematical relationship is constructed using a piecewise linear mapping function, as shown in the following formula:

[0141] When voltage source mode S=0, the formula is:

[0142]

[0143]

[0144] Where, k R1The ratio of active power to virtual resistance is given by the grid resistance R. g Adjustment, P rated R is the rated active power of the converter. v0 As a virtual resistance reference value, the grid resistance R is taken. g 15%; k L1 The ratio of reactive power to virtual inductance is calculated according to... Tuning, Q rated The rated reactive power of the converter; L v0 As the virtual inductance reference value, the grid inductance L is taken. g 15%.

[0145] When the current source mode S=1, the formula is:

[0146]

[0147]

[0148] Where, k R2 Take k R1 1.3 times; R v1 Take the load resistor R l 10%; k L2 Take k L1 1.5 times; L v1 Take the load inductance L l 10%.

[0149] A smooth transition is achieved using linear interpolation. When the switching time is t0, the transition time is T. tran Then at any time t ∈ [t0,t0+T tran The virtual impedance of ] is:

[0150]

[0151]

[0152] in, This is represented as a virtual resistance in voltage source mode. This is represented as a virtual resistance in current source mode. Represented as a virtual inductance in voltage source mode. Represented as a virtual inductance in current source mode, this interpolation process ensures that R v (t), L v (t) is continuously differentiable in the transition section, avoiding abrupt changes in the reference signal.

[0153] Step 2: Introduce state prediction and compensation mechanisms

[0154] A feedforward compensation stage based on a system state prediction algorithm is designed to proactively suppress transient behaviors that may occur during switching. By real-time monitoring of key state variables such as voltage, current, and power, the dynamic response trend at the moment of switching is predicted, and a compensation signal is generated to correct the reference value, thereby effectively suppressing voltage and current surges and improving the output waveform quality.

[0155] Based on the virtual impedance parameter R generated in step 1 v L v The system state variables, including voltage U, current I, power P, and Q, are collected in real time. A state prediction algorithm is used to determine the transient trend at the moment of mode switching, and a compensation signal is generated to correct the voltage / current reference value U. ref I ref This further suppresses transient shocks that were not completely eliminated in the mapping model in step 1, and improves the smoothness of the output waveform and the dynamic response speed.

[0156] Step 2.1 Determine the input variables and prediction target for state prediction

[0157] Input variables include:

[0158] Real-time virtual impedance output from step 1: R v (t), L v (t);

[0159] Instantaneous value of three-phase output voltage of converter: u a (t), u b (t), u c (t), u in the dq coordinate system of three-phase voltage conversion d (t), u q (t);

[0160] Instantaneous value of three-phase output current of converter: i a (t), i b (t), i c (t), i in the dq coordinate system of three-phase current conversion d (t), i q (t);

[0161] Mode switching command signal S: Generated by the system controller based on operating conditions, including 0 = voltage source mode and 1 = current source mode. Advance T pred Triggering the prediction algorithm, T pred For prediction time windows.

[0162] The output variables include:

[0163] After switching T pred The state variables at any given time include:

[0164] Predicted voltage value: , ;

[0165] Predicted current value: , ;

[0166] Step 2.2. Design a state prediction algorithm based on recursive least squares.

[0167] The Recursive Least Squares (RLS) algorithm is used for short-term prediction of state variables. The RLS algorithm flowchart is as follows: Figure 2 As shown, the formula for fitting dynamic trends using historical data is derived as follows:

[0168] Step 2.2.1: Construct the observation matrix of state variables

[0169] Define the state vector in the dq coordinate system. According to the power and virtual impedance model in step 1, the state variables satisfy the linear time-varying equation:

[0170]

[0171] in, ; express The state vector at time t, express The duty cycle signal of the current converter at any given time; Let A(t) be the observation noise vector; A(t) is a 4×4 state transition matrix, derived from R. v (t), L v (t) and the grid angular frequency w are determined as follows:

[0172]

[0173] B(t) is a 4×2 control input matrix, which simplifies to:

[0174]

[0175] Among them, U dc T is the DC side voltage of the converter. s The sampling period.

[0176] Step 2.2.2 Recursive calculation using the RLS algorithm

[0177] The goal is to use historical observations x(t), x(tT) s ), ..., x(t-nT) s Given n=50, minimize the prediction error, update the state transition matrix A(t), and predict t+T. pred The state at time: decomposed to obtain Predicted voltage at time , With predicted current , .

[0178] Step 2.3 Generate feedforward compensation signal and correct reference value

[0179] Step 2.3.1: Calculate the transient error.

[0180] Compare the predicted state with the target steady-state value , , , The error that needs to be compensated is obtained. , , , :

[0181] Voltage error: , ;

[0182] Current error: , ;

[0183] Step 2.3.2: Design the compensation signal transfer function

[0184] To avoid overcompensation leading to new fluctuations, a first-order low-pass filter is used to design the compensation signal. , :

[0185]

[0186]

[0187] in, , , , , , for , , , The result of the transformation from the time domain to the frequency domain, K u K is the voltage compensation coefficient. i This is the current compensation coefficient; , This is the filtering time constant.

[0188] Step 3: Design an adaptive virtual impedance adjustment strategy

[0189] The virtual impedance parameters are dynamically adjusted based on changes in the system's equivalent impedance and the level of power disturbance. The virtual impedance value is optimized in real time by an impedance regulator to adapt to different operating conditions, effectively suppressing transient oscillations and power fluctuations caused by mode switching, and improving the converter's dynamic response performance and operational stability.

[0190] The system employs a three-level logic approach: impedance monitoring, disturbance assessment, and mode-differentiated adjustment. R is achieved through a proportional-integral (PI) control law. v L v The dynamic correction, formula derivation, and module design are as follows:

[0191] Step 3.1: Define the impedance correction factor

[0192] When grid impedance changes abruptly during grid connection or when load switching occurs during islanding, it is necessary to adjust the tracking sensitivity of the virtual impedance and define an impedance correction factor K. Z (t):

[0193]

[0194] Among them, Z eq (t) represents the system's real-time equivalent impedance. The system's rated equivalent impedance is given by the converter's rated power S. rated With rated voltage U rated calculate:

[0195]

[0196] Step 3.2: Generate perturbation coefficients

[0197] , The disturbance coefficient K is defined to reflect the intensity of power fluctuations. dist (t):

[0198]

[0199] in, The active power disturbance intensity, The active power disturbance threshold. The reactive power disturbance intensity. The reactive power disturbance threshold. , K is the perturbation threshold. dist (t) The piecewise function is used to classify the intensity of weak disturbances, normal adjustment and strong disturbances, so as to avoid excessive adjustment when the disturbance is small and cause new oscillations.

[0200] Step 3.3: Pattern Differentiation Modulation Law

[0201] According to S mode(t) Differentiate between VSG / CSG modes and design PI control laws with different control targets:

[0202] Step 3.3.1 Voltage Source Mode S mode (t)=0

[0203] Virtual resistance R v (t) Regulation law: To address voltage loss deviations caused by active power fluctuations, a regulation law is introduced. PI correction:

[0204]

[0205] in, The reference value for the virtual resistor is the output of step 1. for The result of converting from the frequency domain to the time domain For voltage error proportionality coefficient, The voltage error integral coefficient is used to eliminate steady-state voltage error via a PI converter; when When R > 0, increase R v Compensate for voltage loss; conversely, reduce R. v .

[0206] Virtual inductance L v (t) Regulation law: To address the phase deviation caused by reactive power fluctuations, a regulation law is introduced. PI correction:

[0207]

[0208] in, The reference value for the virtual inductance is the output of step 1. For current error proportionality coefficient, The integral coefficient for current error;

[0209] Step 3.3.2 Current Source Mode S mode (t)=1

[0210] Virtual resistance R v (t) Regulation law: To address the current tracking deviation caused by changes in load impedance, a regulation law is introduced. PI correction:

[0211]

[0212] in, For current error proportionality coefficient, For the current error integral coefficient; when When R > 0, decrease R v Reduce current resistance; conversely, increase R. v .

[0213] Virtual inductance L v (t) Regulation law: fixed at 1.1 times that of the previous generation, enhancing the current filtering effect:

[0214]

[0215] Step 4: Achieve seamless switching control between dual modes

[0216] By integrating dynamic mapping models, state prediction compensation, and adaptive impedance regulation mechanisms, a complete seamless switching control architecture is constructed. The switching logic is executed by an embedded control unit, which adjusts control parameters in real time to ensure a smooth and rapid transition between voltage source mode and current source mode, significantly improving the dynamic stability of the energy storage converter in both islanded and grid-connected operation.

[0217] When a mode switching command is received, a preset transition control process is initiated. During the window period from the moment of switching to the end of the transition, the control unit first calls the basic reference signal from step 1 as the reference frame, and synchronously superimposes the compensation signal from step 2. and Superimposed on each and Up, obtain , , and The calculation formula is expressed as follows; simultaneously, if a sudden impedance change or power fluctuation exceeding the threshold is detected, the adaptive adjustment parameters in step 3 are immediately activated to optimize the virtual impedance in real time and correct the R output in step 1. v L v .

[0218] The voltage reference signal of the voltage source is calculated from the virtual impedance and current feedback, that is:

[0219]

[0220] The current reference signal of the current source is calculated from the virtual impedance and voltage feedback, that is:

[0221]

[0222] in, I is the rated voltage, and I is the output current.

[0223] A dual-mode control system for grid-type energy storage converters consists of a sensing and acquisition module, a core control module, an execution and drive module, and a monitoring and feedback module. The functions of each module are detailed below:

[0224] The sensing and acquisition module is responsible for collecting the physical quantities of the system in real time, matching the input variable requirements of step 1.1. It collects the three-phase voltage and three-phase current signals output by the energy storage converter, and calculates the active power and reactive power using preset formulas. At the same time, it receives the mode switching command signal generated by the system controller and synchronously transmits all collected and calculated data to the core control module, providing basic data support for subsequent modeling and control.

[0225] The core control module is the decision-making core of the system, integrating three major functional units:

[0226] The dynamic mapping unit performs the power and virtual impedance dynamic correlation calculation in step 1, and outputs the basic virtual resistance and virtual inductance through the piecewise linear mapping function and linear interpolation algorithm in step 1.2 to ensure the continuity of the reference signal during mode switching.

[0227] The predictive compensation unit includes the state prediction and compensation mechanism corresponding to step 2. Based on the recursive least squares algorithm in step 2.2, it uses historical voltage and current data to predict the system state after the switch, generates a feedforward compensation signal, and corrects the voltage and current reference values.

[0228] The adaptive adjustment unit includes the adaptive virtual impedance adjustment strategy in step 3, calculating the impedance correction coefficient in step 3.1 and the disturbance coefficient in step 3.2, and dynamically correcting R through the mode-differentiated PI adjustment law in step 3.3. v L v Parameters are adapted to changes in system impedance and power disturbances.

[0229] The execution drive module receives instructions from the core control module and carries over the dual-mode seamless switching control logic from step 4. It receives the corrected voltage or current reference value after superimposing the compensation signal in step 4, generates the corresponding PWM drive signal, and controls the on and off of the switching devices inside the converter to achieve physical switching between voltage source mode and current source mode.

[0230] The monitoring and feedback module monitors the system's operating status in real time and constructs a closed-loop regulation mechanism. Its monitoring objects are consistent with the input requirements of steps 2 and 3. It tracks the dynamic changes of the converter's output voltage, current, and power, compares the predicted values ​​in step 2.3.1 with the actual values, and calculates the transient error. It detects sudden changes in the system's equivalent impedance and abnormal load switching conditions, triggering the core control module to execute the adaptive regulation logic in step 3 to ensure the system's dynamic stability.

[0231] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A dual-mode control method applicable to grid-type energy storage converters, comprising the following specific steps, characterized in that: Step 1: Construct a dynamic mapping model between power and virtual impedance By establishing a mathematical relationship model between power and virtual impedance, dynamic mapping between the two can be achieved; Step 2: Introduce state prediction and compensation mechanisms A feedforward compensation stage based on a system state prediction algorithm is designed to proactively suppress transient behaviors that may occur during the switching process. By real-time monitoring of key state variables such as voltage, current, and power, the dynamic response trend at the moment of switching is predicted, and a compensation signal is generated to correct the reference value. Step 3: Design an adaptive virtual impedance adjustment strategy The virtual impedance parameters are dynamically adjusted based on the changes in the system's equivalent impedance and the level of power disturbance. The virtual impedance value is optimized in real time through the impedance regulator to adapt to different operating conditions and effectively suppress transient oscillations and power fluctuations caused by mode switching. Step 4: Achieve seamless switching control between dual modes By integrating the dynamic mapping model, state prediction compensation and adaptive impedance regulation mechanism, a complete seamless switching control architecture is constructed. The switching logic is executed by the embedded control unit, and the control parameters are adjusted in real time to ensure a smooth and fast transition between voltage source mode and current source mode, which significantly improves the dynamic stability of the energy storage converter in islanded and grid-connected operation.

2. The dual-mode control method for grid-type energy storage converters according to claim 1, characterized in that: The specific steps in step 1 for constructing the dynamic mapping model between power and virtual impedance are as follows: Step 1.1 Determine the model input and output variables Input variables: The real-time collected system power parameters include: active power output P of the energy storage converter, reactive power output Q of the energy storage converter, and mode switching command signal S; The energy storage converter outputs active power P, and the three-phase current signal i is acquired by a three-phase current sensor. a i b i c The three-phase voltage signal u collected by the voltage sensor a u b u c Calculate using the following formula: ， Where T is the acquisition time, t is the time variable, and the reactive power Q output by the energy storage converter is calculated using the following formula: ， Mode switching command signal S; Output variables include: virtual resistance R v Virtual inductance L v ; Step 1.2 Establish a mathematical model of power virtual impedance. The mathematical relationship is constructed using a piecewise linear mapping function, as shown in the following formula: When voltage source mode S=0, the formula is: ， ， Where, k R1 The ratio of active power to virtual resistance is the ratio factor, based on the grid resistance R. g Adjustment, P rated R is the rated active power of the converter. v0 As a virtual resistance reference value, the grid resistance R is taken. g 15%; k L1 The ratio of reactive power to virtual inductance is calculated according to... Tuning, Q rated L is the rated reactive power of the converter. v0 As the virtual inductance reference value, the grid inductance L is taken. g 15%; When the current source mode S=1, the formula is: ， ， Where, k R2 Take k R1 1.3 times; R v1 Choose the load resistor R l 10%; k L2 Take k L1 1.5 times; L v1 Take the load inductance L l 10%; A smooth transition is achieved using linear interpolation. When the switching time is t0, the transition time is T. tran Then at any time t ∈ [t0,t0+T tran The virtual impedance of ] is: ， ， in, This is represented as a virtual resistance in voltage source mode. This is represented as a virtual resistance in current source mode. Represented as a virtual inductance in voltage source mode. Represented as a virtual inductance in current source mode, this interpolation process ensures that R v (t), L v (t) is continuously differentiable in the transition section, avoiding abrupt changes in the reference signal.

3. The dual-mode control method for grid-type energy storage converters according to claim 1, characterized in that: The state prediction and compensation mechanism introduced in step 2 is as follows: Step 2.1 Determine the input variables and prediction target for state prediction Input variables include: Real-time virtual impedance output from step 1: R v (t), L v (t); Instantaneous value of three-phase output voltage of converter: u a (t), u b (t), u c (t), u in the dq coordinate system of three-phase voltage conversion d (t), u q (t); Instantaneous value of three-phase output current of converter: i a (t), i b (t), i c (t), i in the dq coordinate system of three-phase current conversion d (t), i q (t); Mode switching command signal S: includes 0 = voltage source mode, 1 = current source mode; advance T pred Triggering the prediction algorithm, T pred For prediction time windows; The output variables include: After switching T pred The state variables at any given time include: Predicted voltage value: , ; Predicted current value: , ; Step 2.

2. Design a state prediction algorithm based on recursive least squares. The recursive least squares (RLS) algorithm is used to perform short-term prediction of state variables, and historical data is used to fit the dynamic trend. The formula derivation is as follows: Step 2.2.1: Construct the observation matrix of state variables Define the state vector in the dq coordinate system. According to the power and virtual impedance model in step 1, the state variables satisfy the linear time-varying equation: ， in, ; express The state vector at time t, express The duty cycle signal of the current converter at any given time; Let A(t) be the observation noise vector; A(t) is a 4×4 state transition matrix, derived from R. v (t), L v (t) and the grid angular frequency w are determined: ， B(t) is a 4×2 control input matrix, which simplifies to: ， Among them, U dc T is the DC side voltage of the converter. s The sampling period; Step 2.2.2 Recursive calculation using the RLS algorithm The goal is to use historical observations x(t), x(tT) s ), ..., x(t-nT) s Given n=50, minimize the prediction error, update the state transition matrix A(t), and predict t+T. pred The state at time: decomposed to obtain Predicted voltage at time , With predicted current , ; Step 2.3 Generate feedforward compensation signal and correct reference value Step 2.3.1: Calculate the transient error. Compare the predicted state with the target steady-state value , , , The error that needs to be compensated is obtained. , , , : Voltage error: , ; Current error: , ; Step 2.3.2: Design the compensation signal transfer function To avoid overcompensation leading to new fluctuations, a first-order low-pass filter is used to design the compensation signal. , : ， ， in, , , , , , for , , , The result of the transformation from the time domain to the frequency domain, K u K is the voltage compensation coefficient. i This is the current compensation coefficient; , This is the filtering time constant.

4. The dual-mode control method for grid-type energy storage converters according to claim 1, characterized in that: The adaptive virtual impedance adjustment strategy designed in step 3 is as follows: Step 3.1: Define the impedance correction factor When grid impedance changes abruptly during grid connection or when load switching occurs during islanding, it is necessary to adjust the tracking sensitivity of the virtual impedance and define an impedance correction factor K. Z (t): ， Among them, Z eq (t) represents the system's real-time equivalent impedance. The system's rated equivalent impedance is given by the converter's rated power S. rated With rated voltage U rated calculate: ， Step 3.2: Generate perturbation coefficients , The disturbance coefficient K is defined to reflect the intensity of power fluctuations. dist (t): ， in, The active power disturbance intensity, The active power disturbance threshold. The reactive power disturbance intensity. The reactive power disturbance threshold. , K is the perturbation threshold. dist (t) The piecewise function is used to classify the intensity of weak disturbances, normal adjustment and strong disturbances, so as to avoid excessive adjustment when the disturbance is small and cause new oscillations. Step 3.3: Pattern Differentiation Modulation Law According to S mode (t) Differentiate between VSG / CSG modes and design PI control laws with different control targets: Step 3.3.1 Voltage Source Mode S mode (t)=0 Virtual resistance R v (t) Regulation law: To address voltage loss deviations caused by active power fluctuations, a regulation law is introduced. PI correction: ， in, The reference value for the virtual resistor is the output of step 1. for The result of converting from the frequency domain to the time domain For voltage error proportionality coefficient, The voltage error integral coefficient is used to eliminate steady-state voltage error via a PI converter; when When R > 0, increase R v Compensate for voltage loss; conversely, reduce R. v ; Virtual inductance L v (t) Regulation law: To address the phase deviation caused by reactive power fluctuations, a regulation law is introduced. PI correction: ， in, The reference value for the virtual inductance is denoted as , and the output is denoted as . For current error proportionality coefficient, The integral coefficient for current error; Step 3.3.2 Current Source Mode S mode (t)=1 Virtual resistance R v (t) Regulation law: To address the current tracking deviation caused by changes in load impedance, a regulation law is introduced. PI correction: ， in, For current error proportionality coefficient, For the current error integral coefficient; when When R > 0, decrease R v Reduce current resistance; conversely, increase R. v ; Virtual inductance L v (t) Regulation law: fixed at 1.1 times that of the previous generation, enhancing the current filtering effect: 。 5. The dual-mode control method for grid-type energy storage converters according to claim 1, characterized in that: Step 4, which implements seamless switching control between the two modes, specifically involves: Upon receiving a mode switching command, a preset transition control procedure is initiated. Within the window period from the moment of switching to the end of the transition, the basic reference signal is used as the baseline frame, and the compensation signal is synchronously superimposed. and Superimposed on each and Up, obtain , , and The calculation formula is expressed as follows; simultaneously, if a sudden impedance change or power fluctuation exceeding the threshold is detected, adaptive adjustment parameters are immediately activated to optimize the virtual impedance in real time and correct the R output in step 1. v L v ; The voltage reference signal of the voltage source is calculated from the virtual impedance and current feedback, that is: ， The current reference signal of the current source is calculated from the virtual impedance and voltage feedback, that is: ， in, I is the rated voltage, and I is the output current.

6. The system for dual-mode control of grid-type energy storage converters according to any one of claims 1-5, comprising a sensing and acquisition module, a control module, an execution and drive module, and a monitoring and feedback module, characterized in that: The sensing and acquisition module is responsible for collecting physical quantities of the system in real time, including the three-phase voltage and three-phase current signals output by the energy storage converter, and calculating the active and reactive power using preset formulas. At the same time, it receives mode switching command signals generated by the system controller and transmits all collected and calculated data to the core control module in a synchronized manner, providing basic data support for subsequent modeling and control. The core control module integrates three major functional units: The dynamic mapping unit performs dynamic correlation calculations between power and virtual impedance, and outputs the basic virtual resistance and virtual inductance through a piecewise linear mapping function and a linear interpolation algorithm. The predictive compensation unit includes a corresponding state prediction and compensation mechanism. Based on the recursive least squares algorithm, it uses historical voltage and current data to predict the system state after the switch, generates a feedforward compensation signal, and corrects the voltage and current reference values. The adaptive adjustment unit incorporates an adaptive virtual impedance adjustment strategy, calculates the impedance correction coefficient and disturbance coefficient, and dynamically corrects R through a mode-differentiated PI control law. v L v Parameters to adapt to system impedance changes and power disturbances; The execution drive module receives instructions from the control module, undertakes the dual-mode seamless switching control logic, receives the corrected voltage reference value or current reference value after superimposed compensation signal, and generates the corresponding PWM drive signal; controls the on and off of the switching devices inside the converter to realize the physical switching between voltage source mode and current source mode. The monitoring and feedback module monitors the system's operating status in real time and constructs a closed-loop regulation mechanism. Its monitoring objects are consistent with the input requirements, tracking the dynamic changes of the converter's output voltage, current, and power, comparing the predicted values ​​with the actual values, and calculating the transient error. It also detects sudden changes in the system's equivalent impedance and abnormal load switching conditions, triggering the core control module to execute adaptive regulation logic to ensure the system's dynamic stability.

Images